JP2005189981A - Information processor and its method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To properly execute statistical information processing. <P>SOLUTION: A summarization cycle deciding part 340 decides a summarization cycle(one week or the like) used for dividing the summarization period of time series data being the target of statistical processing for the statistical processing. A summarization block deciding part 342 divides the summarization period into a plurality of summarization blocks. The summarization blocks are set with a variable length so that the number of time series data suitable for the statistical processing can be included as an expected value. A singular value detecting part 344 totals and verifies the time series data based on the acquired summarization blocks, and an output processing part 346 outputs the totaling/verification result. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an information processing apparatus and method for performing information processing on information that occurs over time.

Patent Document 1 discloses a method of performing abnormality diagnosis of a control system using a time series model.
Further, Patent Document 2 discloses a method for finding an artificial or continuous portion in temporal data based on a threshold value.
Patent Document 3 discloses a method of analyzing information based on a plurality of periods and decomposing periodic fluctuations, non-periodic fluctuations, and noise.
However, any of the methods disclosed in Patent Documents 1 to 3 cannot detect unknown fluctuations included in information from an appropriate amount of information with appropriate sensitivity.

JP-A-6-187030 JP-A-10-293757 JP 2002-344447 A

  The present invention has been made from the background described above, and an object thereof is to provide an information processing apparatus and method capable of appropriately performing statistical information processing.

  In order to achieve the above object, an information processing apparatus according to the present invention performs information processing on a plurality of pieces of processing target information that are generated over time in a predetermined period and can be processed. The information processing is performed based on one or more processing sections included in the predetermined period, and one or more processing sections are variable in length according to the generation mode of the target information. A processing section setting means for setting, and an information processing means for performing the information processing on the processing target information based on the set variable length processing section.

In the information processing apparatus according to the present invention, the processing section setting means divides a period during which information is collected into processing sections for performing information processing (such as statistical processing).
The processing section setting means variably sets the length of the processing section so that each processing section includes a number of information suitable for information processing (statistical processing).
For example, when the chi-square goodness-of-fit test is performed on the information included in the processing section, the length of the processing section is set so that each processing section includes 10 or more pieces of information suitable for this test as expected values. Set the length of the processing section.
The information processing means performs processing (such as statistical processing) on information included in the processing section set in this way.

  Preferably, the predetermined period includes one or more periods, and the processing section setting unit sets the processing section based on a period included in the period.

  Preferably, the information processing apparatus further includes a fluctuation period detecting unit that detects a fluctuation period of the processing target information, and the processing section setting unit sets the processing section based on the detected fluctuation period.

  Preferably, the information processing is performed based on processing sections each including a predetermined number or more of the processing target information, and the processing section setting means is configured such that each of the processing sections includes the predetermined number or more of the processing target information. The processing section is set to a variable length so as to include

  Preferably, the information processing is performed based on processing sections each including a predetermined number or more of the processing target information, and the processing section setting means has the processing sections having the same number of processing target information as expected values. The processing section is set to a variable length so as to include

  Preferably, the processing object is further transmitted within a plurality of units that can be evaluated, or between units, and further includes a processing object information storage unit that stores the processing object information, and the information processing unit includes: The information processing is performed on the recorded processing target information, and the influence of the evaluation target unit, which is the target of evaluation, on the other units among the plurality of units is evaluated.

  Preferably, the information processing is statistical processing.

  The information processing method according to the present invention is an information processing method for performing the information processing on a plurality of pieces of processing target information that occurs over time in a predetermined period and can be a target of information processing of a predetermined method. The information processing is performed based on one or more processing sections included in the predetermined period, and one or more processing sections are set with a variable length according to the generation mode of the target information. Then, based on the set variable length processing section, the information processing is performed on the processing target information.

  The program according to the present invention is an information processing apparatus that performs the information processing on a plurality of pieces of processing target information that is generated over time in a predetermined period and can be an information processing target of a predetermined method. The information processing is performed based on one or more processing sections included in the predetermined period, and one or more of the processing sections are set with a variable length according to the generation mode of the target information. Based on the setting step and the set variable-length processing section, the computer executes the information processing for the processing target information.

  According to the information processing apparatus and method according to the present invention, statistical information processing can be appropriately performed.

[First Embodiment]
In order to facilitate the understanding of the present invention, prior to the description of the first embodiment, first, the background that led to the present invention will be described.
Until now, the value of an organization (whether it is an organization with an entity such as a company department or a virtual organization such as a mailing list) is the value of an official transaction and service outside of an organization, or It has been evaluated by sales to the outside.
For example, in a company, organizations that are functionally differentiated with respect to problems to be solved arranged hierarchically are arranged hierarchically, and the problem is solved by transmitting a command to such an organization.

Within such a corporate structure, the assessment of an organization's value is, for example, the input of resources such as people, goods and money that are purely invested in the organization, and the resulting economic value and services. It has been evaluated as the difference and ratio with the output.
However, it has been pointed out from both the academic and actual business aspects that the value of an organization cannot be assessed only from the perspective of the difference and ratio between input and output.
Also, in the field of business administration, it is pointed out that such a problem solving method using a hierarchical organization cannot quickly respond to various requests of customers.
In response to such problems, IT (Information Technology) has been introduced, the organization within the company has a flat structure, autonomous activities are recommended to each organization, and resources are allocated to solve the problem. A solution to flexibly change and optimize is proposed.

However, when the structure of the organization is made flat in this way, it becomes extremely difficult to evaluate the value of the organization and its members based on the difference and ratio between the input and the output.
This is because, as described above, in a flat structure, the structure changes flexibly, so that an organization is not set up aiming at solving some problems from the beginning, and the members of the organization always change. Or a member may formally or informally belong to more than one organization, and after the problem is solved, the organization itself will be dissolved. This is because it is difficult to evaluate what kind of output was made.

The present invention focuses on communication within an organization and between organizations (organizational communication) regardless of whether it is formal or informal in order to evaluate the value of the organization.
Specifically, the present invention pays attention to what range and scale of communication that occurs (transmitted) in one organization is used in other organizations. Evaluate the value of
In other words, the present invention pays attention to organization communication and determines the value of a certain organization using the influence of the organization on another organization.

Focusing on these points, the value of organizations and organizational communications that did not produce direct economic value, such as providing information and services not officially imposed on a certain organization, or that were difficult to evaluate. Can alternatively be evaluated.
More specifically, when an organization is solving a problem, by analyzing the content of the communication in this organization, information in one organization can be communicated to other organizations and their members, Focusing on what is used for the activities of other organizations and their members, it is possible to objectively evaluate what is different between the value evaluation of the organization and the value evaluation of the organization communication.
For example, in a company, not only the value of the department that is actually making profits, but at first glance it seems that the value is low because it does not make profits, but in fact it is beneficial to many departments of the company and is indirectly Since the value of departments that are linked to large profits can be quantitatively evaluated, grasping the value of the organization from this perspective is useful for appropriate allocation of investments and budgets, etc. Can contribute.

The applicant of the present application has already applied for an invention that supports various analyzes of organization communication performed on a daily basis as Japanese Patent Application No. 2001-275808 (organization communication analysis apparatus and method; Application 1).
The invention according to application 1 is a factor that activates or inactivates organizational communication by performing quantitative analysis on organizational communication such as a mailing list / electronic bulletin board and associating the analysis result with qualitative information. In an exploratory analysis, the organization communication operation is realized based on the analysis results.
The present invention is a further development of the invention according to the application 1 from the viewpoints described so far, and realizes an objective value evaluation of an organization by performing an analysis on the organization communication.
Examples of organizational communication means (media) include verbal, telephone, video telephone systems, and computer networks (e-mail, electronic bulletin board, chat, instant message, etc.).
In order to realize the present invention, it is premised that the communication performed through these media is recorded.
In the following, for the sake of concreteness and simplification of explanation, a case where an electronic mail is used in a company is taken as a specific example.
When e-mail is used for communication, generally the message in the natural language included in the e-mail, the sender and receiver of the e-mail, the date and time when the e-mail was transmitted, etc. Can be logged.

Distribution of electronic mail on the Internet is performed via a transmission server called an SMTP (Simple Mail Transfer Protocol) server.
Therefore, in the company, by recording the message log of the e-mail with the SMTP server in a unified manner, when and who communicated with whom and what kind of content for all employees in the company Can be grasped.
The present invention is configured to evaluate the value of an organization in a company by analyzing a message log of an SMTP server that distributes electronic mail in the company, for example.

[Network system 1]
Hereinafter, a first embodiment of the present invention will be described.
FIG. 1 is a diagram illustrating a configuration of a network system 1 to which an evaluation method according to the present invention is applied.
The network system 1 is, for example, a wide area network (WAN) constructed across a plurality of offices in the same company. As shown in FIG. 1, each of a plurality of organizations 1 to n that can be evaluated The departmental systems 2-1 to 2-n (n ≧ 2) and the analysis apparatus 3 are connected via the network 100.
Hereinafter, in the case where any one of a plurality of components such as the departmental systems 2-1 to 2-n is indicated without being specified, it is simply abbreviated as the departmental system 2.
Each departmental system 2 has a configuration in which, for example, client computers 20-1 to 20-m (m ≧ 1) used by m members of each organization are connected to the server 24 via the departmental LAN 102. Take.

[Hardware configuration]
FIG. 2 is a diagram illustrating a hardware configuration of the client computer 20, the server 24, and the analysis / evaluation apparatus 3 illustrated in FIG.
As shown in FIG. 2, the client computer 20, the server 24, and the analysis device 3 include a main body 200 including a CPU 202 and a memory 204, a display / input device 206 including a liquid crystal display, a keyboard and a mouse (not shown), an HDD It comprises a recording device 208 such as a CD device, and a communication device 212 that communicates with the network 100 and the departmental LAN 102.
That is, the client computer 20, the server 24, and the analysis device 3 include components as a general computer that can communicate via a network.

[Client program 22]
FIG. 3 is a diagram showing the configuration of the client program 22 that runs on the client computer 20 shown in FIGS.
As shown in FIG. 3, the client program 22 includes a user interface unit (UI unit) 220, a mail program 222, a web browser 224, and a LAN communication control unit 226.
For example, the client program 22 is supplied to the recording device 208 of the client computer 20 via the recording medium 210, loaded into the memory 204, and executed.
The client program 22 provides a mail transmission / reception function and a WWW browsing function to members (users) of an organization that uses the client computer 20 by using these components.

In the client program 22, the UI unit 220 receives a user operation on the display / input device 206 (FIG. 2) and controls processing of each component of the client program 22.
In addition, the UI unit 220 displays the electronic mail received by the mail program 222 and the data from the WWW received by the web browser 224 to the user.

The mail program 222 provides the user of the client computer 20 with a function for sending and receiving electronic mail.
The web browser 224 provides a WWW browsing function to the user of the client computer 20.
The LAN communication control unit 226 communicates with other client computers 20 or servers 24 in the same organization via the departmental LAN 102 (FIG. 1) and the network 100 (generally constituting components of communication main body are also referred to as communication nodes). And communication with communication nodes of other organizations.

[Server program 26]
FIG. 4 is a diagram showing the configuration of the server program 26 that operates on the server 24 shown in FIGS.
As shown in FIG. 4, the server program 26 includes a LAN communication control 260, a network communication control 262, a mail server program 264, a web server 266, a log management unit 268, and a log database (log DB) 270.
As with the client program 22 (FIG. 3), the server program 26 is supplied to the recording device 208 (FIG. 2) of the server 24 via the recording medium 210 (FIG. 1), loaded into the memory 204, and executed.

  With these components, the server program 26 provides a mail server function to the client computers 20 (members) belonging to the same departmental system 2 (organization), and the same or different departmental system 2 (organization). The WWW server function is provided to the client computer 20 (member).

In the server program 26, the LAN communication control 260 performs communication control with the departmental LAN 102 (FIG. 1).
The network communication control 262 performs communication control with the network 100.
The mail server program 264 implements a mail server function.
The web server 266 implements a WWW server function.

FIG. 5 is a diagram illustrating an e-mail message log recorded in the log DB 270 illustrated in FIG. 4.
FIG. 6 is a diagram showing a specific example of the message log shown in FIG.
The log management unit 268 records communication (communication) performed by the mail server program 264 and the web server 266 as a message log in the log DB 270 according to control data input from the analysis / evaluation apparatus 3 via the network communication control 262. To do.

Further, the log management unit 268 transmits the message log recorded in the log DB 270 to the analysis / evaluation apparatus 3 via the network communication control 262 and the network 100 according to the control of the analysis / evaluation apparatus 3.
As shown in FIGS. 5 and 6, the message log recorded by the log management unit 268 includes IDs (communication IDs) attached to the respective e-mails, transmission side of the e-mail, and reception side. The identification data (ID), the content of the e-mail message (communication content), the date and time when the e-mail was transmitted, and the like are recorded.

[Analysis / Evaluation Program 30]
FIG. 7 is a diagram showing a configuration of a first analysis / evaluation program 30 that operates on the analysis / evaluation apparatus 3 shown in FIGS. 1 and 2.
As shown in FIG. 7, the analysis / evaluation program 30 includes a network communication control 300, a log collection / management unit 302, a log DB 304, a first analysis / evaluation unit 306, an analysis / evaluation result DB 308, an organization / personal DB 310, and a UI. Part 312.
Similar to the client program 22 (FIG. 3) and the server program 26 (FIG. 4), the analysis / evaluation program 30 is supplied to the recording device 208 of the analysis / evaluation device 3 via the recording medium 210 and loaded into the memory 204. To be executed.

With these components, the analysis / evaluation program 30 collects e-mail message logs, WWW browsing logs, and the like from the respective servers 24 (FIG. 1) of the departmental systems 2-1 to 2-n. Analyzing message logs, etc. to evaluate the value of the organization.
It should be noted that even if the WWW log is analyzed, the value evaluation of the organization according to the present invention can be realized. However, as described above, in the description of the embodiment, the analysis is performed on the message log of the email, A specific example is a method for evaluating the value of an organization.

In the analysis / evaluation program 30, the network communication control 300 performs communication control with the network 100.
The UI unit 312 accepts a user operation on the display / input device 206 and controls processing of each component of the analysis / evaluation program 30.
Further, the UI unit 312 displays the log stored in the log DB 304 and the analysis result / evaluation result stored in the analysis / evaluation result DB 308 on the display / input device 206 in accordance with a user operation.

FIG. 8 is a diagram showing organization information stored in the organization / personal DB 310 shown in FIG.
FIG. 9 is a diagram showing a specific example of the organization information shown in FIG.
FIG. 10 is a diagram showing personal information stored in the organization / personal DB 310 shown in FIG.
FIG. 11 is a diagram showing a specific example of the personal information shown in FIG.
The organization / individual DB 310 includes organization information (FIGS. 8 and 9) of each of the organizations 1 to n (FIG. 1) using the departmental systems 2-1 to 2-n, and the members 1 of the organizations 1 to n. Each personal information (FIG. 10, FIG. 11) of .about.m is stored.

As shown in FIGS. 8 and 9, the organization / individual DB 310 has, as organization information of each organization 1 to n, an identifier (organization ID), organization name, organization form, and organization period (existence) of the organizations 1 to n. Period), and if there is a higher organization in the organizations 1 to n, the higher organization is stored.
As shown in FIGS. 10 and 11, the organization / individual DB 310 includes, as personal information of the members 1 to m of the organizations 1 to n, identifiers (personal ID / employee ID), names, and emails of the members 1 to m. The address and the organization ID (FIG. 8) of the organization to which the members 1 to m belong are stored.

In the organizational form item of the organizational information shown in FIGS. 8 and 9, the organization appearing in the existing organization chart is represented as formal.
In addition, an organization formed to perform cross-organizational information exchanges such as well-end meetings and mailing lists that do not appear in the organization chart is represented as semi-formal.
An organization that carries out cross-cutting activities at intervals is called a project.

The organization ID is uniquely associated with each organization and is used to indicate the correspondence with the upper organization.
The organization name is the name of the organization or informal or project organization.
The lifetime of the organization is assigned from when to when it has been.
In addition, the organizational form shown in FIGS. 8 and 9 includes an official organization as described in the organization chart of a company, and a project organization in which a plurality of official organizations are gathered to achieve a certain purpose. Attribute of organization such as community based on voluntary participation, organization such as information sharing group with common interests (normal organization, project, community, etc.) is stored.

FIG. 12 is a diagram showing organization communication information (organization communication information) stored in the log DB 304 by the log collection / management unit 302.
FIG. 13 is a diagram showing a specific example of the organization communication information (organization communication information) shown in FIG.
In the organization communication information (organization communication information) shown in FIGS. 12 and 13, the organization communication ID is uniquely assigned to each communication.
The sending side ID and the receiving side ID correspond to the personal information shown in FIG. 10 and the like, and the sending side organization ID and the receiving side organization ID are based on the correspondence with the organization information associated with the personal information. can get.
Further, in the item of communication content, information such as a conversation performed between the sender ID and the receiver ID is recorded, and further, the date and time when this communication is performed is associated.

FIGS. 12 and 13 show a case where there is one sender and one receiver. However, by appropriately changing the data structure shown in FIG. Organizational communication information of many electronic mails can also be stored in the log DB 304.
The log collection / management unit 302 collects the message logs shown in FIG. 5 from each of the servers 24 of the departmental systems 2-1 to 2-n via the network 100 and stores them in the log DB 304.

FIG. 14 is a flowchart showing the organization communication information creation process (S44) by the log collection / management unit 302 shown in FIG.
As described above, the log collection / management unit 302 refers to the organization / individual DB 310, the message log obtained from the server 24, the organization information shown in FIGS. 8 and 9, and the individuals shown in FIGS. 10 and 11. The information is associated with the information, and the organization communication information (FIGS. 12 and 13) is created and stored in the log DB 304 as described below.

In step 240 (S440), the log collection / management unit 302 reads personal information.
In step 442 (S442), the log collection / management unit 302 reads organization information.
In step 444 (S444), for example, the log collection / management unit 302 sets the organization communication information as a processing target line by line from the beginning, and the organization communication information that has not yet been set as the processing target in the processing so far. The first line is read for processing.

In step 446 (S446), the log collection / management unit 302 associates the organization communication information with the transmission side ID, the reception side ID, the transmission side organization ID, and the reception side organization ID of the organization communication information, and stores the organization communication information of the row to be processed. create.
In step 448 (S448), the log collection / management unit 302 determines whether the processing has been completed for all the rows of the organization communication information.
The log collection / management unit 302 ends the process when the process is completed for all rows, and returns to the process of S444 otherwise.

  For example, when the processing shown in FIG. 14 is applied to communication by normal e-mail, the log collection / management unit 302 uses the e-mail originator e-mail address and the e-mail e-mail address of the e-mail to the organization / personal DB 310. The personal ID (employee ID) of the member i who sent the electronic mail by searching the stored personal information (FIG. 10) and the personal ID of the member j (i, j = 1 to m) who received the electronic mail ( The personal ID (employee ID) of the member i who sent the e-mail is the sender ID, and the personal ID (employee ID) of the member i who received the e-mail is the receiver ID.

Furthermore, the log collection / management unit 302 searches the organization information (FIG. 8) stored in the organization / personal DB 310 using the sender ID and sender ID, and the organization to which the member i who sent the e-mail belongs belongs. The organization ID of p is the sender organization ID, and the organization ID of the organization q (p, q = 1 to n) to which the member j who has received the email belongs is the receiver organization ID.
Further, the log collection / management unit 302 adds an identifier (organization communication ID) to each piece of information obtained as a result of the above search, adds the message content of the e-mail as communication content, The sent date / time and organization form are added, and the organization communication information shown in FIG. 12 is created and stored in the log DB 304.

[Analysis / Evaluation Unit 306]
Hereinafter, analysis processing and evaluation processing performed in the analysis / evaluation unit 306 will be described.

[Analysis of common words]
The analysis / evaluation unit 306 extracts organization communication information having the same transmission-side organization ID and reception-side organization ID from the organization communication information (FIG. 12) stored in the log DB 304.
In other words, the analysis / evaluation unit 306 transmits the electronic message transmitted between the internal members i and j (client computers 20-i, j) in each of the organizations p (departmental systems 2-1 to 2-m). Organization communication information (intra-organization communication information) is taken out.

Further, the analysis / evaluation unit 306 displays the organization (transmission side organization ID and reception side organization ID) to which the transmission side (ID) and the receiver side (ID) included in the intra-organization communication information belong to FIGS. Analyzing the communication contents of the organization (corresponding to the organization information shown), and extracting the common concept used in the organization for all organizations (sender side organization ID) included in the intra-organization communication information Tally.
As a common concept of this organization, communication contents of in-house communication information are generally treated as a unit in natural language processing, etc., words, sentences, synonyms of common words, semantic networks based on common words and synonyms, and , Ontology (a common concept set used in the organization p), and the like can be used.

FIG. 15 is a flowchart showing processing (S46) in which the analysis / evaluation unit 306 extracts and counts commonly used words (common words) as a common concept.
As shown in FIG. 15, as shown in step 460 (S460), the analysis / evaluation unit 306, for example, sets organization communication information as a processing target one line at a time from the beginning. The first line among the lines of organization communication information that is not the target of is read as a processing target.

In step 462 (S462), the analysis / evaluation unit 306 determines whether or not the sender ID and the receiver ID of the line read by the process of S460 are the same.
The analysis / evaluation unit 306 proceeds to the process of S48 when the sender ID is the same as the receiver ID, and returns to the process of S460 otherwise.

In step 48 (S48), the analysis / evaluation unit 306 performs individual tabulation processing.
This individual tabulation process will be described later with reference to FIG.
In step 464 (S464), the analysis / evaluation unit 306 adds the frequency and type of the word obtained as a result of the individual tabulation to the list as a common word candidate of the sender organization (ID).

In step 466 (S466), the analysis / evaluation unit 306 determines whether or not the processing has been completed for all rows of the organization communication information.
The analysis / evaluation unit 306 proceeds to the process of S468 when the process is completed for all the rows, and returns to the process of S460 otherwise.
In step 468 (S468), the analysis / evaluation unit 306 sets the top five common words in each organization as common words.
The list obtained by the process of S464 can include an enormous number of words, but the number of words can be limited by the process of S468.

FIG. 16 is a diagram showing the individual tabulation process (S48) shown in FIG.
As shown in FIG. 16, in step 480 (S480), the analysis / evaluation unit 306 extracts communication contents from the row of organization communication information to be processed, and divides it into word units.
In this process, for example, the analysis / evaluation unit 306 divides the communication content into parts of speech by morphological analysis, and extracts a noun as an expression element.
Note that the analysis / evaluation unit 306 can use not only a noun but also other parts of speech such as a verb as an expression element, a combination of several parts of speech as an expression element, or a phrase or the like instead of a part of speech. It can also be used as an expression element.

In step 482 (S482), the analysis / evaluation unit 306 counts the number of word types from the expression elements obtained by the processing of S480.
In step 484 (S484), the analysis / evaluation unit 306 counts the number of appearances of words for each type of word obtained by the process of S482.
In the individual tabulation process, the analysis / evaluation unit 306 may extract proper nouns as well as simple word appearance frequencies, or may count the frequency of proper nouns.

FIG. 17 is a first diagram illustrating common word information in the organization stored in the analysis / evaluation result DB 308 by the analysis / evaluation unit 306 shown in FIG.
18A to 18C are diagrams showing specific examples of the common word information shown in FIG.
For example, the analysis / evaluation unit 306 has, as the common words, three words that are frequently included in the communication contents (FIGS. 12 and 13) of the in-organization communication information of the organization p, for example, three words that are used more frequently. The words x, y, and z are selected, and as shown in FIG. 17, the organization ID of the organization p (FIG. 8) is added to each of the words x, y, and z to create common word information (attributes) and analyze Store in the evaluation result DB 308.
As shown in FIGS. 18A to 18C, the common word information is associated with the organization ID and stored as a list.

FIG. 19 is a second diagram illustrating a common concept in the organization stored in the analysis / evaluation result DB 308 by the analysis / evaluation unit 306 shown in FIG.
FIG. 20 is a diagram showing a specific example of the common concept shown in FIG.
As described above, for example, the analysis / evaluation unit 306 includes a sentence included in the communication content of the communication information in the organization, a synonym of the common word, a semantic network based on the common word and the synonym, and an ontology (used in the organization p. Information indicating other concepts of the common word, such as a common concept set).

The analysis / evaluation unit 306 records the extracted words independently, for example, as shown in FIG.
Alternatively, for example, as shown in FIG. 20, the analysis / evaluation unit 306 records the extracted words in association with other concepts, for example, common word information (FIGS. 17, 18A to 18C). May be.

In addition, the specific example which recorded the common synonym independently as a common concept is shown in FIG. 20, This common synonym has the same meaning using the common synonym dictionary etc. which are marketed. The analysis / evaluation unit 306 performs the same processing as the common word (however, a step for matching all the corresponding words in the synonym dictionary is included as a target to be matched) for the organization communication information. Thus, it is possible to extract and aggregate.
The matching here corresponds to “search” in terms of the database field, and specifically, for example, a search for a common concept such as word A from a certain communication unit group.

FIGS. 21 and 23 are first and second diagrams illustrating the inter-organization communication information generated by the analysis / evaluation unit 306.
22 is a diagram showing a specific example of the inter-organization communication information shown in FIG.
As shown in FIG. 21 and FIG. 23, the analysis / evaluation unit 306 adds organization IDs (sending organization ID, receiving organization ID) to the extracted common word and adds inter-organization communication information. Is created and stored in the analysis / evaluation result DB 308.
The analysis / evaluation unit 306 stores, in the analysis / evaluation result DB 308, the organization communication information exchanged (transmitted / received; transmitted) between two different organizations p and q (here, p ≠ q) (FIG. 12, The common concept of FIG. 13) is extracted and registered.

FIG. 24 is a flowchart showing a matching process (S50) in which the analysis / evaluation unit 306 extracts a common concept of the organization communication information (FIGS. 12 and 13).
In the following, the analysis / evaluation unit 306 extracts the common concept of the organization communication information (FIGS. 12 and 13), taking as an example a case where a word is obtained as a common concept from emails sent and received between two different organizations. Processing to be performed will be described.
In step 500 (S500), the analysis / evaluation unit 306 sets, for example, the organizational communication information (FIGS. 12 and 13) as a processing target one line at a time from the beginning. Of the lines of organization communication information that have not been processed, the first line is read as a processing target.

In step 502 (S502), the analysis / evaluation unit 306 determines whether or not the transmission-side organization ID and the reception-side organization ID included in the processing-target organization communication information are different.
The analysis / evaluation unit 306 proceeds to the process of S504 when the originating organization ID and the receiving organization ID are different, and proceeds to the process of S512 otherwise.
In other words, the sender organization ID is different from the sender organization ID, and the sender organization receives the word (common concept) of the sender organization between the receiver organization and the sender organization. Is considered to have communicated (influenced) the party organization.

In step 504 (S504), the analysis / evaluation unit 306 reads the common word list of the sender organization ID (= receiver organization ID) included in the organization communication information to be processed and sets it as a matching processing target. Set the common word i.
In step 506 (S506), the analysis / evaluation unit 306 determines whether or not the common word i that is the processing target is included in the communication content included in the row of the organization communication information that is the processing target.
The analysis / evaluation unit 306 proceeds to the process of S508 when the common word i to be processed is included in the row of the organization communication information to be processed, and otherwise proceeds to the process of S510.

In step 508 (S508), the analysis / evaluation unit 306 records the common word i as a processing target as inter-organization information shown in FIG.
In step 510 (S510), the analysis / evaluation unit 306 sets the common word i + 1 that has not yet been processed among the common words included in the word list as the next common word i to be processed.
In step 512 (S512), the analysis / evaluation unit 306 determines whether or not the processing has been completed for all rows of the organization communication information.
The analysis / evaluation unit 306 ends the process when the process has been completed for all rows, and returns to the process of S500 otherwise.

To summarize the matching process shown in FIG. 24, first, the analysis / evaluation unit 306 reads the organization communication information line by line, extracts the corresponding communication contents, and creates a common word (FIG. 17, FIG. 18) is matched with the communication content included in the read line.
This matching processing is executed when the transmission-side organization ID included in the row that is the processing target of the organization communication information is different from the reception-side organization ID.

As a result of this matching processing, when the common word of the sender organization ID is included in the communication content, it is assumed that the in-organization information exchanged by the sender organization has affected the recipient organization. 22 is recorded as inter-organization communication information.
The matching process is executed for all the common words included in the common word list associated with the caller organization ID included in the row of the organization communication information to be processed.

By performing the above matching processing for all the organization communication information, the organization communication information indicating which organization influences which organization via which common word (influence degree) (Fig. 21 and FIG. 22) are created.
Note that, when the same processing as the processing shown in FIG. 24 is performed on the concept (sentence, synonym of common word, semantic network by common word and synonym, and ontology) instead of the common word, two organizations Inter-organizational communication information (FIG. 23) indicating what kind of influence is given from which to which is created.

In addition, regarding the organization communication information exchanged between two or more organizations, by applying the common concept (word) extraction / aggregation processing shown in FIG. 15 and the matching processing shown in FIG. It is possible to know what kind of common word affects which organization from which organization.
For example, when information is transmitted from the sending organization p to the receiving organizations q and r, the sending organization ID is the organization ID of the organization p, and the receiving organization ID is the organization ID of the organization q. When the receiving side organization ID is the organization ID of the organization r and the processes shown in FIGS. 15 and 24 are performed separately, inter-organization communication information can be obtained.

FIGS. 25 and 26 are first and second diagrams illustrating personal common word information stored in the analysis / evaluation result DB 308 by the analysis / evaluation unit 306 illustrated in FIG. 7.
The analysis / evaluation unit 306 also extracts common words and concepts of organization communication information (FIG. 12) of e-mail exchanged between members, as well as common word information within and between organizations. be able to.
As shown in FIG. 25 or FIG. 26, the analysis / evaluation unit 306 attaches personal IDs (sender personal ID, receiver personal ID) to the extracted common word and concept, and performs personal communication. Information is created and stored in the analysis / evaluation result DB 308.

[Organization impact assessment]
The analysis / evaluation unit 306 includes the intra-organization communication information of the organization p to be evaluated in the intra-organization communication information (FIGS. 17 and 19) and the personal communication information (FIGS. 25 and 26) generated as described above. Personal communication information of individual i, intra-organization communication information (FIGS. 17 and 19) of each organization q (q = 1 to n; q ≠ p), and inter-organization communication information (FIGS. 21 and 23) Compare.
Further, the analysis / evaluation unit 306 includes the organization / individual common word / concept (FIGS. 17, 19, 25, and 26) included in the intra-organization communication information as the common word / concept, and An organization that has been influenced by the organization or individual to be evaluated includes an organization that includes the common word / concept of the organization / individual to be evaluated as a common word / concept in the inter-organization communication information that uses the organization as the sender or receiver. It is determined that

27 and 30 are first and second diagrams illustrating the impact evaluation results stored in the analysis / evaluation result DB 308 by the analysis / evaluation unit 306 illustrated in FIG.
FIG. 28 is a diagram showing a specific example of the influence range shown in FIG.
FIG. 29 is a diagram illustrating a specific example of the degree of influence.
Here, a specific example is a case where the influence is divided into the influence range and the influence degree.
As shown in FIGS. 27 and 28, the influence range is calculated by counting how many other organizations use a concept commonly used in a certain organization (common concept).
As shown in FIG. 29, the degree of influence is calculated by counting how many concepts commonly used in one organization appear in other organizations.

Hereinafter, how to evaluate the inter-organization communication information (FIGS. 21 and 22) created by the processing shown in FIG. 24 will be described.
FIG. 31 is a flowchart showing the process (S52) for calculating the influence range shown in FIG.
As shown in FIG. 31, in step 520 (S520), the analysis / evaluation unit 306, for example, sets inter-organization communication information (FIGS. 21 and 22) as processing targets one by one in order from the beginning. The first line among the lines of inter-organization communication information that has not yet been processed is read as a process target.

In step 522 (S522), the analysis / evaluation unit 306 determines whether or not a flag is attached to the receiving-side organization ID included in the row of inter-organization communication information read by the processing of S520.
The analysis / evaluation unit 306 proceeds to the process of S528 when the flag is attached to the receiving-side organization ID, and proceeds to the process of S524 otherwise.

In step 524 (S524), the analysis / evaluation unit 306 increments the value of the influence range of the caller organization ID.
In step 526 (S526), the analysis / evaluation unit 306 flags the receiving organization ID.

In step 528 (S528), the analysis / evaluation unit 306 determines whether or not the processing has been completed for all rows of the inter-organization communication information.
The analysis / evaluation unit 306 terminates the process when all the lines of the inter-organization communication information are terminated, and returns to the process of S520 otherwise.

The processing shown in FIG. 31 is summarized.
First, the analysis / evaluation unit 306 reads inter-organization communication information (FIGS. 21 and 22) line by line.
Next, the analysis / evaluation unit 306 counts the number of receiving organizations without duplication from the inter-organization communication information.
In order to count the number of receiving side organizations without duplication, the analysis / evaluation unit 306 adds a flag to each receiving side organization ID that has been counted once, and does not count the receiving side organizations with the flag.
The analysis / evaluation unit 306 performs such counting for all the rows of the inter-organization communication information, and sets the total value as an influence range for all the organizations of a certain organization (sender organization).

FIG. 32 is a flowchart showing the process (S54) for calculating the influence degree shown in FIG.
As shown in FIG. 32, in step 540 (S540), the analysis / evaluation unit 306, for example, sets inter-organization communication information (FIGS. 21 and 22) as processing targets one line at a time from the beginning. The first line among the lines of inter-organization communication information that has not yet been processed is read as a process target.

In step 542 (S542), the analysis / evaluation unit 306 increments the value of the influence degree of the originating organization.
In step 544 (S544), the analysis / evaluation unit 306 determines whether or not the processing has been completed for all rows of the inter-organization communication information.
The analysis / evaluation unit 306 terminates the process when all the lines of the inter-organization communication information are terminated, and returns to the process of S540 otherwise.

The process shown in FIG. 32 is summarized.
The analysis / evaluation unit 306 reads inter-organization communication information (FIGS. 21 and 22) one line at a time, and cumulatively counts how many times a common word is used in the receiving organization for each of the transmitting organizations.
The analysis / evaluation unit 306 performs such counting for all the rows of the inter-organization communication information, and sets the total value as the degree of influence on all the organizations of a certain organization (sender organization).
The analysis / evaluation unit 306 counts the cumulative value of the frequency of common words in the sending organization, so that when more commonly used words in one organization are used in other organizations Alternatively, the degree of influence may be calculated by estimating the degree of influence high.

FIG. 33 is a diagram illustrating a specific example of the influence range obtained for each concept.
FIG. 34 is a diagram illustrating a specific example of the degree of influence obtained for each concept.
28 and 29 show the total results actually obtained by the processes shown in FIGS. 31 and 32. FIG.
The inter-organization communication information shown in FIG. 21 and FIG. 22 corresponds to the organizational communication information shown in FIG. 12 and FIG. 13, and thus the same sender organization ID and receiver organization ID exist. 28 and FIG. 29 are total values, there is no duplication of organization ID (sender organization ID).
In addition, as shown in FIGS. 27 and 29, the influence range and the degree of influence can be calculated for each organization ID, or as shown in FIGS. 33 and 34, the total value can be obtained for each concept. it can.
The aggregate values shown in FIGS. 33 and 34 are equivalent to the aggregate value of the influence range and the influence degree for each organization ID shown in FIGS. 28 and 29.

Note that the number of appearances of the common word x may be analyzed noting the entire network system 1 (company) but focusing only on the organization q where the employee j communicates with the employee i of the organization p.
It is wrong to judge that organization p is affected by organization q when the same word is just a coincidence and used only in organization q other than organization p. The influence range and the influence level of such an error can be removed from the influence range and the influence degree that an organization has on other influences.

Furthermore, even if the organization p and the organization q are not in direct organizational communication (organizational communication), but are in an indirect communication (communication etc.) via another organization r, It is also possible to evaluate the influence range and the influence degree in a chained manner and evaluate the total value as the influence of the organization p on the organization q.
For example, each of the organizations A to D uses the word x in communication inside and outside the organization. In the inter-organization communication information (FIGS. 21 and 22) calculated from the organization communication information shown in FIGS. When the organizations A and B and B and C are connected by organization communication (communication) including the common word x, the appearance of the common word x in the organization C is the range of influence that the organization A has on the organization C. And can be counted as an influence degree.

For such chained evaluation, the number of chains included in the evaluation may be set in advance.
For example, if the number of chains is 2, the word y is included between the organization A and the organization B, between the organization B and the organization C, and between the organization C and the organization D in the inter-organization communication information (FIGS. 21 and 22). When included in the communication content, the influence range and influence of the organization A are not counted between the organizations CD.
If a loop such as organization A to organization B and organization B to organization A occurs, the process ends at that point.

Furthermore, for the evaluation of the chained influence range and the degree of influence, weighting depending on the number of chains may be performed.
For example, when communication (communication) including the common word x from organization A to organization B, organization B to organization D, and from organization D to organization E is performed in a chained manner, the influence range and influence of organization A In order to evaluate the degree, 1 is added to the number of organizations as the degree of influence on the organization B. When counting the influence range and the degree of influence on the organization D via the organization B, 1/2 is added to the number of organizations. When the influence range and degree of influence on the organization E that passes through the organization B and further passes through the organization D are counted, the evaluation may be performed by adding 1/4 to the number of organizations.

[Other analysis and evaluation]
The analysis / evaluation unit 306 includes the number of inter-organization communication information, intra-organization communication information, and personal communication information shown in FIGS. 17 to 26, the frequency, the organization form (business), and the common words / concepts of any organization / individual. It is possible to analyze how an organization / individual to be evaluated is utilized in which organization based on which organization it is transmitted to.
The organization communication information (FIG. 12) includes a sender organization ID, a recipient organization ID, and information on the date and time when the e-mail was exchanged, and the analysis / evaluation unit 306 tracks these. Thus, it is possible to analyze in what time course and in which route the common word / concept of the organization / individual to be evaluated has spread.
The analysis / evaluation unit 306 combines the above analysis / evaluation results as necessary into a comprehensive analysis result, and stores it in the analysis / evaluation result DB 308.
Each analysis / evaluation shown above is selectively used and can be combined as appropriate.

[Overall operation of network system 1]
Hereinafter, the overall operation of the network system 1 will be described.
FIG. 35 is a diagram showing an analysis / evaluation sequence (S10) in the network system 1.
As shown in FIG. 35, the user operates the display / input device 206 (FIGS. 1 and 2) of the analysis / evaluation apparatus 3 to select one of the organizations 1 to n (departmental systems 2-1 to 2-n). These are designated as the evaluation target organization p (department-specific system 2-p), which medium of the WWW and e-mail is to be evaluated (S100, S102).

For example, the user of the analysis / evaluation apparatus 3 can specify what influence the organization communication of the planning organization Z of the project X has in the company.
Further, for example, the user designates a range for analyzing the organization communication of the planning organization Z of the project X not only as an influence on the company-wide organization communication but also as an influence on a plurality of internal organizations and a group of employees. be able to.
As described above, in this embodiment, a specific example is a case where the user designates an e-mail as a medium.

The UI unit 312 of the analysis / evaluation program 30 (FIG. 7) accepts this operation and controls the log collection / management unit 302 to control each of the organizations 1 to n (departmental systems 2-1 to 2-n). Then, a signal for accumulating an e-mail message log is transmitted (S104).
When this signal is received, the log management unit 268 of the server program 26 operating on the server 24 of each of the departmental systems 2-1 to 2-n causes the inside of each of the departmental systems 2-1 to 2-n and A message log of electronic mail transmitted between the departmental systems 2-1 to 2-n is accumulated.

The user performs an operation to instruct the display / input device 206 (FIG. 2) of the analysis / evaluation apparatus 3 to start analysis and evaluation (S106).
In response to this operation, the log collection / management unit 302 receives a message log from each of the departmental systems 2-1 to 2-n (S108).

The user designates the evaluation range (S110).
In other words, for example, the user analyzes or evaluates the influence or the like given to the partial range of the organizations 1 to n by the organization p to be evaluated in S100, or the influence on all the ranges of the organizations 1 to n. Whether to analyze / evaluate, etc., whether to analyze only the organization whose evaluation target affects other organizations, or whether to analyze the impact and utilization, etc., on the display / input device 206 specify.

FIG. 36 is a diagram showing an analysis / evaluation process (S20) of the analysis / evaluation unit 306 (FIG. 7) shown in FIG.
The analysis / evaluation unit 306 performs analysis / evaluation in accordance with this designation, as shown in FIG. 36, and stores the result in the analysis / evaluation result DB 308.
The UI unit 312 displays the analysis / evaluation result obtained by the process of S20 on the display device of the display / input device 206 or records it on the recording medium 210 via the recording device 208.

  As shown in FIG. 36, in step 200 (S200), as described with reference to FIGS. 17 and 19, the analysis / evaluation unit 306 analyzes the communication contents of the email transmitted within the organization, Extract common words (attributes) for each organization.

  In step 202 (S202), the analysis / evaluation unit 306 analyzes the communication content of the email transmitted between the organizations as described with reference to FIGS. 21 and 23, and is transmitted between the organizations. Extract common words (attributes) in emails.

  In step 204 (S204), as described with reference to FIGS. 25 and 26, the analysis / evaluation unit 306 analyzes the communication content of the e-mail related to the individual, and determines the common word (attribute) of the e-mail related to the individual. Extract.

  In step 206 (S206), the analysis / evaluation unit 306 compares the common word / concept of the organization / individual to be evaluated with the common word / concept of another organization.

  In step 208 (S208), as described with reference to FIGS. 27 and 30, the analysis / evaluation unit 306 evaluates the influence of the organization / person to be evaluated on other organizations.

  In step 210 (S210), the analysis / evaluation unit 306 analyzes the utilization of the organization / individual to be evaluated.

In step 212 (S212), the analysis / evaluation unit 306 determines the value of the organization / individual to be evaluated from the number of organizations (FIGS. 27 and 30) affected by the organization / individual to be evaluated and the degree of utilization. evaluate.
The analysis / evaluation unit 306 can also evaluate organization communication in an arbitrary time range by referring to the existence period of the organization communication information (FIG. 12) and the date and time of communication.

  In step 214 (S214), the analysis / evaluation unit 306 analyzes how the influence of the organization / individual to be evaluated has changed over time.

  In step 216 (S216), the analysis / evaluation unit 306 stores the comprehensive analysis / evaluation result of the organization to be evaluated obtained in the above processing in the analysis / evaluation result DB 308.

[Modification]
In the above description, e-mail and WWW have been exemplified as the media. However, the present invention can be applied to other media as well, such as live audio in a conference room or audio for audio communication. May be evaluated based on text data obtained by voice recognition.
The present invention can also be applied to communication via an electronic bulletin board, a chat system, an instant message, and the like.

[Concrete example]
In addition, for example, according to the present invention, as a voluntary community that shares case information on its products, the influence of the organization formed in the form of a mailing list, the exchange of emails of the organization developing the product, It is possible to evaluate the degree of influence by analyzing all communication including voice recording at a meeting and chatting at a seat.
In addition, according to the present invention, it is possible to grasp how the content of discussion on the mailing list for free discussion on company-wide missions affects all communication of all employees.

  According to the present invention, for example, even if a person is a person who does not speak on the above-mentioned company mailing list (so-called Read Only Member: abbreviated ROM), the person is personally notified by e-mail, telephone / face-to-face, etc. You can evaluate the impact of the mailing list on other organizations if you communicate the information you get on the mailing list to others, forward the mail to other mailing lists, or introduce topics that were published on the mailing list at the conference. it can.

More specifically, according to the present invention, a concept (word, synonym, sentence, semantic network, ontology, etc.) used in a certain mailing list and a concept used outside the mailing list are recorded and analyzed. -By evaluating, it is possible to analyze and evaluate whether the concepts used in the mailing list are used outside the mailing list.
For example, on the mailing list where product x developers and cross-organization sales employees can discuss anything about product x, conversations about "new usage of function y of product x and its promotion method" are active. Let's say.
According to the present invention, regardless of whether or not a participant in the mail list is speaking, if the topic is spoken in a conversation other than the mailing list, the fact can be known as an analysis result.

Further, by analyzing how frequently the words (common words) frequently used in the conversation and the concept thereof are used in e-mails of other organizations, the present invention enables the conversation on the mailing list to be performed as follows. It can be evaluated as an alternative to the conventional evaluation method based on the difference and ratio between the input and the output, how much other organizations are affected and how much other organizations have penetrated.
For example, a conversation about “a new way to use function y of a product x and its sales promotion method” was presented as a unique proposal to a customer by a sales employee, and then another salesperson who was on the same mailing list. Assume that an employee has a conversation that asks his / her boss whether or not he / she wants to use the proposal as his / her proposal.

According to the present invention, such a conversation can be recorded, and its influence / utilization degree can be evaluated as the value of the mailing list.
In addition, other sales employees who were looking at the conversation on the mailing list actually made proposals to their customers. However, in order for the proposals to be approved by the customers in charge of these sales, a new function y was added. Suppose you post a comment on the mailing list that you need to add function and develop function y '.
Therefore, each sales representative has a conversation with his supervisor who wants to ask the development department to develop the function y 'individually. In general, adding such a function may be difficult for one customer. , Often ends with recommendations to the boss, or is proposed through an organizational route for official function updates as a candidate for the next version.

However, in this case, there is a time lag between the time when it is actually necessary and the function y ′ being implemented, and it is often impossible to solve the customer's problem at an appropriate time.
On the other hand, according to the present invention, it is also possible to record a conversation with a boss who does not reach the organizational route for function update, and this is evaluated as the influence level and the influence range of the mailing list.
This evaluation includes a call for function y 'in the semantic network that co-occurs with function y. As a result, managers, development departments, and planning departments can quickly meet market needs from such conversations. It is also possible to grasp and study the development of the function y ′.

[Second Embodiment]
First, the background and outline of the second embodiment of the present invention will be described.
There may be steady fluctuations in the frequency of occurrence of time-series events such as network traffic, email arrival status and car arrival at the toll booth.
As an example of this, there can be mentioned an event where the network traffic in the company increases during the day and decreases during the night.

On the other hand, abnormal fluctuations that are not steady may occur in time-series events.
As an example of this, there can be mentioned an event in which the amount of traffic on the network suddenly increases due to an event such as the occurrence of a computer virus and takes a singular value.
Another example is an event where discussions on the mailing list are excited and the number of emails suddenly increases at one time.
As illustrated here, the events that occur in time series and the data (time series data) indicating the frequency or amount of such events have non-stationary fluctuations (singular values) due to some event. ) May occur.

In this way, singular values that occur in time-series data often have their causes. Conversely, by detecting the occurrence of a singular value, the event (event) that causes this singular value is detected. It is possible to know the possibility that has occurred.
However, if the frequency of occurrence of an event that causes a singular value is low, the frequency of occurrence of a singular value is also low, and it is difficult to know the occurrence.

That is, if the frequency of occurrence of an event is low, the frequency of occurrence of a singular value resulting therefrom is also low, so it is difficult to know an event having a low frequency of occurrence from the occurrence of a singular value resulting therefrom.
In order to detect such an infrequent event, for example, a method may be considered in which statistical test processing is always performed on time-series data for a long period of time, and its singularity is detected. Because of the low occurrence frequency, it is not always easy to always monitor the time series data for a long period of time and perform statistical analysis.
Therefore, it would be convenient to be able to find appropriate singularity candidates appropriately and automatically from time-series data obtained over a certain period of time, and a method that enables such a case would be very difficult. Useful for.

Here, for example, when the amount of event occurrence over time (change in the number of arrivals and arrival rates of e-mails) can be modeled by a known function, statistical processing (for example, chi-square fitness test) is applied. The singular point of the event occurrence amount can be easily found by testing the degree of fit between the event occurrence amount and the estimated amount by the function.
On the other hand, when the function that can model the amount of event occurrence (time-series data) is unknown, in order to detect the singularity by applying the chi-square goodness-of-fit test, the amount of event occurrence (standard It is necessary to estimate a singularity based on the chi-square goodness after estimating a function that can be modeled from this).

Thus, in order to detect the singularity by estimating the modeling function, the distribution of χ square values obtained from the modeling function and the amount of event occurrence (time series data) is calculated. , It is necessary to approximate the chi-square distribution.
For approximation to this chi-square distribution, it is known from experience that it is preferable that the number of samples (time series data) of each category for which the chi-square value is calculated is 10 or more. In general, it is known that when the number of samples is less than 10, the accuracy of the test decreases.
That is, for example, the number of e-mail arrivals at a certain organization is counted over a certain counting period (for example, one month), and this counting period is divided into a plurality of sections (for example, one hour), and the χ square fitness is used. In the case of detecting a specific point, it can be seen that the aggregation period may be divided so that the expected value of the number of emails reached in each section is 10 or more.

On the other hand, for example, the method disclosed in Patent Document 1 (Japanese Patent Laid-Open No. Hei 6-187030) described above performs abnormality diagnosis of a control system with a normal state model obtained in advance from the input / output of the control system. This is done using an abnormal state model.
Therefore, this method cannot be used for an object for which a model of data to be diagnosed cannot be obtained in advance.

Further, for example, the method disclosed in Patent Document 2 (Japanese Patent Laid-Open No. 10-293757) described above identifies an artificial data string or an excessively continuous data string from time series data by a run test.
This identification is made based on whether or not the threshold value is exceeded. Therefore, when the singular point generated in the time series data is excessive or too small due to the balance with the threshold value, this method accurately identifies the singular point. Can not do it.
Since the determination is made based only on the binary value indicating whether or not the threshold value is exceeded, it cannot be applied to determine a value that is too large or too small.

In addition, the method disclosed in Patent Document 3 (Japanese Patent Application Laid-Open No. 2002-344447) described above analyzes time series data indicating network traffic in a plurality of cycles such as one day, one week, or one month. By analyzing the periodic fluctuation factor, the non-periodic fluctuation factor, and the noise, the fluctuation tendency is discovered.
However, this method requires time-series data over a long period of time, and the period actually used for analysis does not necessarily match the period effective for finding a singular value.

The second embodiment of the present invention is made from such a background, and is time-series data indicating events that occur over time (time-series data with periodicity or similar behavior). Whether or not it is time-series data without repetition), a contrivance is made so that a test by statistical processing can be appropriately performed.
When the total period of time-series data is divided into sections of equal length, the number of time-series data in each section is not necessarily limited to a number (10) for suitably performing statistical processing (for example, chi-square test). May not be met.
Accordingly, it is not always appropriate to equally divide the counting period into a section for statistical processing.

On the other hand, in the second embodiment of the present invention, each of the sections obtained as a result of the division is expected to include a suitable number of time series data for statistical processing. Data is divided into variable length sections.
In this way, by appropriately dividing the counting period into sections of variable length, detection of singular values by statistical processing can be suitably performed over the entire counting period.
Note that the second embodiment of the present invention is not limited to the chi-square test, and can be widely used for information processing performed by dividing the totaling period. For example, a case in which statistical processing (χ square test) is performed on time-series data is taken as a specific example.

Hereinafter, a second embodiment of the present invention will be described.
FIG. 37 is a diagram showing the configuration of the second analysis / evaluation program 32 operating on the analysis / evaluation apparatus 3 shown in FIGS.
Of the constituent parts of the second analysis / evaluation program 32, those substantially the same as those of the first analysis / evaluation program 30 shown in FIG.
As shown in FIG. 37, the second analysis / evaluation program 32 employs a configuration in which the first analysis / evaluation unit 306 in the first analysis / evaluation program 30 is replaced with the second analysis / evaluation unit 34. .

FIG. 38 is a diagram illustrating a configuration of the second analysis / evaluation unit 34 illustrated in FIG. 37.
As shown in FIG. 38, the analysis / evaluation unit 34 includes a totaling period determination unit 340, a totaling section determination unit 342 (processing section setting unit), a singular value detection unit 344 (information processing unit), and an output processing unit 346. The
The analysis / evaluation unit 34 detects a singular value in the occurrence frequency of the organization communication information (the time series data; processing target information) by using these components.
FIG. 38 shows components for detecting a singular value by setting a variable-length section as described above and performing a chi-square test on the tissue communication information stored in the log DB 304. However, the analysis / evaluation unit 34 may include a function for realizing the function of the first analysis / evaluation unit 306 (FIG. 7).
In addition, the function of the first analysis / evaluation unit 306 and the function of the second analysis / evaluation unit 34 may be used together or combined unless a problem such as contradiction occurs.

In the analysis / evaluation unit 34, for example, as a totaling period, a period that can be a period (counting period) of statistical processing (chi-square test) for inter-organization communication information by the user via the UI unit 312 (FIG. 37) or the like. Is set.
In the following, a specific example is given of a case where 1 year, 1 month, 1 week, 1 day, 1 hour, and 1 minute are set in the analysis / evaluation unit 34 (aggregation cycle determination unit 340) as the aggregation cycle.
The totaling period determining unit 340 selects a totaling period suitable for the chi-square test from the totaling period of the tissue communication information that is input from the log DB 304 or the like and is subjected to singular value detection.

Hereinafter, the processing of the counting cycle determination unit 340 will be further described.
FIG. 39 is a diagram exemplifying inter-organization communication information input from the log DB 304 shown in FIG. 37 to the aggregation cycle determining unit 340 shown in FIG.
FIG. 40 is a flowchart showing the processing (S22) of the analysis / evaluation unit 34 shown in FIGS.
As shown in FIG. 39, an access log from each organization to the Web server is input from the log DB 304 to the aggregation cycle determining unit 340 of the analysis / evaluation unit 34 as inter-organization communication information.
Each entry of the inter-organization communication information includes the IP address of the reference destination Web server, the reference time, the command, and the URL of the reference source organization.

As shown in FIG. 40, in step (S220), the counting cycle determination unit 340 determines the reference time of the earliest entry from the reference time of each entry of the inter-organization communication information, and the latest summed entry. Find the reference time.
Hereinafter, among the entries of the inter-organization communication information, the reference time of the earliest totalized entry is 8: 8: 55 on Monday, October 1, 2001, and the reference time of the latest totalized entry is 2001. Taking the case of Wednesday, October 31st, 20: 30: 5 as a specific example, the aggregation cycle determination unit 340 first sets the aggregation period of the inter-organization communication information as 30 days 12 hours 21 minutes 10 seconds. calculate.

When dividing the total period into total periods, since the total period must include at least two total periods, the maximum time length of the total period is 15 days 6 hours 10 minutes 35 seconds.
In the analysis / evaluation unit 34, since the set value of the counting cycle is from January to 1 minute as described above, the counting cycle determination unit 340 calculates a counting cycle having a time length of 15 days, 6 hours, 10 minutes, 35 seconds or less, Select from the set values and set it as the aggregation cycle.
Accordingly, in this case, for example, the counting cycle determination unit 340 determines the counting cycle as 1 week to 1 minute.

FIG. 41 is a diagram exemplifying fluctuations in the total period included in the total period and the number of entries in the inter-organization communication information.
However, the change in the number of entries in the inter-organization communication information shown in FIG. 41 and the occurrence time of the entries shown in other figures are not necessarily consistent.
As shown in FIG. 41, the aggregation period (30 days 12 hours 21 minutes 10 seconds) of the inter-organization communication information is not an integral multiple of the aggregation period (1 week). Only entries for 5 days are included.

In step 222 (S222), the totaling section determining unit 342 (FIG. 38) calculates the average number of entries of inter-organization communication information per one totaling period.
For example, when the number of entries of the inter-organization communication information obtained in the aggregation period is 1268, as illustrated in FIG. 41, the aggregation period includes one or four weeks of aggregation period (one week).
Accordingly, the aggregation section determination unit 342 determines the average number of entries included in one aggregation cycle (one week) as 253.6 (= 1268 cases / 5 weeks) to 317.0 (= 1268 cases / 4 weeks). calculate.

Similarly, since the aggregation period includes the aggregation period (one day) for 30 days or 31 days, the aggregation section determination unit 342 calculates the average number of entries included in one aggregation period (one day), It is calculated as 40.9 (≈1268 / 31 days) to 42.3 (≈1268 / 30).
Hereinafter, similarly, the totaling section determination unit 342 calculates the average number of entries for a shorter totaling period (1 hour and 1 minute).

In step 224 (S224), the totaling interval determination unit 342 (FIG. 38) calculates a dividable cycle.
When the splittable period is divided into two or more tabulation intervals, each tabulation section obtained by the division can include a suitable number of entries (10 or more) for the chi-square test as expected values. It is a counting cycle.
Therefore, the totaling interval determination unit 342 selects a totaling cycle that includes 20 or more entries in each cycle as a splittable cycle.

Therefore, in this specific example, the counting section determination unit 342 includes one week including 253.6 to 317.0 entries for each cycle and 40.9 to 42.3 entries for each cycle. One day including is selected as a splittable period.
Accordingly, the totaling section determination unit 342 does not select 1 hour to 1 minute as the division possible period.

FIG. 42 is a diagram illustrating a reference value table that can be used in the splittable cycle selection process (S224) illustrated in FIG.
In the reference value table shown in FIG. 42, the aggregation period (1 year to 1 hour) indicated by each item in the first column of the table includes each of the aggregation periods and each item (per year) in the first row of the table. When there are more than the number of entries corresponding to (per day), it indicates that the period can be divided.
As described above, the processing of S224 by the totaling interval determining unit 342 can be modified so as to select a splittable cycle by referring to the reference value table illustrated in FIG.

In step 226 (S226; FIG. 40), the totaling section determination unit 342 (FIG. 38) determines whether there is a totaling period to be processed in the next steps S226 to S234.
When there is a counting cycle to be processed next, the counting section determination unit 342 proceeds to the processing of S228 with the entry of the inter-organization communication information included in the counting cycle as the processing target, otherwise, the processing of S236 is performed. move on.

In step 228 (S228), the totaling interval determination unit 342 divides the totaling period to be processed into the totaling interval, and each of the obtained totaling intervals is suitable for the chi-square test ( In order to include the number of entries of 10 or more), it is determined how many aggregation sections the division cycle should be divided into.
That is, the totaling section determination unit 342 determines the number of totaling sections (number of sections z) that should be included in the totaling cycle that is the target of processing.
More specifically, for example, when the number of entries is 20 to 29 in the counting period (one day) to be processed, the counting section determination unit 342 sets the number of sections z for this counting period. 2 (z = 2).

As the number of sections z is increased, finer χ square tests are possible, but the amount of calculation and its cost also increase.
In order to avoid such a problem, for example, the analysis / evaluation program 324 subtracts 12 from the average number of entries in the aggregation period (one week) obtained in the process of S222, and rounds down the square root of the subtraction result. Thus, the number of sections z is obtained.
In this specific example, the square root of the average number of entries in the counting period (one week) is 6.23 (= (253.6-12) ^1/2 ) to 6.73 (= (317.0-12) ^1/2 ). The analysis / evaluation program 324 determines the number of sections z to be 6 for the counting cycle (1 week).

However, this method is an example. For example, when the average number of entries in the counting period is 20 to 29, the number of sections z can be guaranteed to be 2, and the average number of entries in each counting section is suitable for the χ square test. It can be changed to an arbitrary method that can guarantee the number (10 or more).
Further, when there may be a plurality of section numbers z that satisfy such conditions, the total section determination unit 342 displays a plurality of section numbers z to the user via the UI unit 312 (FIG. 37), and the user's Depending on the selection, the number of sections z may be determined.

FIG. 43 is a flowchart showing the delimiter time determination process (S26) shown in FIG.
As shown in FIG. 43, in step 260 (S260), the totaling interval determining unit 342 (FIG. 38) divides the totaling interval by the totaling cycle, and calculates the number of cyclic repetitions c (= totaling interval / totaling cycle). To do.

In step 262 (S262), the totaling section determining unit 342 calculates the time of each entry in each totaling period.
This will be specifically described.
The aggregation period from October 1 to October 31 is the aggregation period of one week (October 1 to 7 (first aggregation period), October 8 to 14 (second aggregation period), ... , October 29 to 31 (fifth counting cycle)), attention is paid to the first counting cycle of the inter-organization communication information (FIG. 39).

Here, from the reference time ([01 / Sep / 2001: 08: 48: 06 + 0900] to [01 / Sep / 2001: 09: 55: 37 + 0900]) of each entry included in the first aggregation cycle, The time obtained by subtracting the start date / time (01 / Sep / 2001) of the first aggregation period is the time of each entry ([00: 00: 00: 08: 48: 06 + 0900] to [00:00:00: 09: 55: 37 + 0900]).
That is, for example, when the reference time of an entry included in the first aggregation cycle is 02 / Sep / 2001: 09: 48: 06 + 0900], the time of this entry in the first aggregation cycle is [00: 00: 01: 09: 55: 37 + 0900].

Similarly, for example, when the reference time of an entry included in the second aggregation cycle is [09 / Sep / 2001: 10: 50: 16 + 0900], the start date of the second aggregation period is calculated from this reference time. The time obtained by subtracting the time (08 / Sep / 2001) is the time [00: 00: 01: 10: 50: 16 + 0900] in the second counting period of this entry.
With this concept, the time of each entry in the counting period can be calculated for each counting cycle (1 year to 1 minute).

In step 264 (S264), the counting section determination unit 342 performs relative date ([00:00:00] to [00:00:06] of each entry in each of the first to fifth counting periods; Focusing on Monday to Sunday), entries of the same date are collected from all counting cycles, and entries of the same date are rearranged (sorted) in time order.
More specifically, the tabulation section determination unit 342 collects entries for the same day of the week from all the tabulation cycles regardless of whether the date is October 11th or 18th, and arranges the entries in time order for each day of the week. Change.
In order to simplify the process, the relative date from the beginning of the cycle is converted to time, Tuesday 00:00 to 24:00, Wednesday 00:00 to 40:00 : 00 may be rearranged.
Based on the same idea, the totaling section determining unit 342 sorts such entries even when the totaling period is other than one week.
As a result of this sorting, as shown in FIG. 41, it is possible to know in which order each entry is in each counting period.

In step 266 (S266), the tabulation section determination unit 342 obtains the last entry order e.
For example, a simple average value of 1268 entries (accesses) obtained for about 31 days per week is 1268 ÷ 31 × 7 = 286.32.
However, since the frequency of entries occurring during one week is not constant, it is necessary to correct this average value.
Hereinafter, a specific example will be given in which the order of the last entry (entry obtained at 20: 30: 5 on October 31) shown in FIG. 41 is No. 817 (e = 817) out of 1268 entries.
In other words, in the case illustrated below, it was obtained at 20: 30: 5 on October 31 (Wednesday) (68 hours, 20 minutes, 5 seconds from the beginning of the counting period (0: 0: 0 on Monday)). The number of entries is 817.
Furthermore, in other words, a case where 817 entries are generated between 00:00 on Monday and 20: 20: 5 on Wednesday is illustrated below.

In step 268 (S268), the tabulation section determination unit 342 calculates the expected number of entries (expected frequency) f for each tabulation section.
As shown in FIG. 41, for the 1st to 817th (= e) th entries, the counting cycle is repeated five times, and for the 818th (= e + 1) to 1268th entries, the counting cycle is Repeated 4 times.
The expected frequency f is calculated by [f = (e / r (c) + (se) / i (c)) / total number of sections].
Here, r (c) indicates that the decimal point of the cyclic repetition number c is raised, i (c) indicates that the cyclic repetition number c is rounded down, and s indicates the total number of entries (1268).

In this specific example, the expected frequency f is 46.025 [= (817/5 + (1268−817) / 4) / 6].
That is, in the example shown here, since October 1 is Monday and October 31 is Wednesday, the number of repetitions is 5 from 20:30 to 5 seconds from Monday to Wednesday, and the number of entries in this period is , 1/5, the number of repetitions in other periods is 4, the number of entries in other periods is 1/4, and these are added to obtain the expected frequency f [= (817/5 + (1268− 817) / 4) / 6] are calculated.

FIG. 44 is a diagram exemplifying the total section obtained by the total section determining unit 342 (FIG. 38) by the start time and the end time.
FIG. 45 is a diagram exemplifying a totaling result of the number of entries included in the totaling section illustrated in FIG.
In step 270 (S270), the totaling section determination unit 342 initializes the totaling section number n (n = 1).
The total section number n corresponds to the sections 1 to 6 shown in FIGS.

In step 272 (S272), the tabulation section determination unit 342 calculates the tabulation position v of the tabulation section.
The delimiter position v is calculated by v = n × f × r (c).
In step 274 (S274), the totaling interval determination unit 342 determines whether the delimiter position v is larger than the position e of the last entry.
The counting section determination unit 342 proceeds to the process of S278 when the delimiter position v is larger than the position e of the last entry, and proceeds to the process of S276 otherwise.

In step 276 (S276), the tabulation section determination unit 342 corrects the tabulation position v of the tabulation section.
The division position v is corrected by calculating v = n × f × i (c) + e / r (c).

  In step 278 (S278), the tabulation section determination unit 342 obtains the entry of the end time of the tabulation section shown in FIG. 44 in the tabulation section, and the section P (n ).

In step 280 (S280), the totaling section determination unit 342 increments the totaling section number n.
In step 282 (S282), the tabulation section determination unit 342 determines whether or not the delimiter position v is larger than the number of sections z obtained by the process of S228.
The total section determination unit 342 proceeds to the process of S272 when the delimiter position v is larger than the number of sections z obtained by the process of S228, and proceeds to the process of S230 (FIG. 40) otherwise.

Note that, according to the specific example shown here, the processing of S278 determines the segment P (1) of the first tabulation section because the number of repetitions of this part is 5, and is obtained by the processing of S272, and 230 (= i (46 .025 × 5 × 1)).
Similarly, the delimiters P (2) and P (3) of the second and third total sections are 460, 690 (= i (46.025 × 5 × 2), i (46.025 × 5 × 3). )).

Similarly, the delimiter P (4) of the fourth tabulation interval is once calculated as 920.5, but this value is larger than the position e of the last entry and the number of repetitions is 4, so S276 Is corrected to 900 by the above process.
Similarly, the delimiters P (5) and P (6) of the fifth and sixth summary sections are 1084 and 1268, respectively.

In this specific example, for simplification of processing, for example, 230, 460, 690, 1084, 1268 are adopted as the separation positions P (1) to P (6).
By the process of S26 described above, as illustrated in FIG. 44, a total section in each total period is obtained.
44. As the start time and end time of each aggregation section shown in FIG. 44, for example, even if the reference time (FIG. 39) of the first entry and the last entry of each aggregation section is used, the last entry of the previous aggregation section is used. The intermediate time of the reference time of the first entry in the subsequent aggregation section may be used.

In step 230 (S230; FIG. 40), the singular value detection unit 344 performs aggregation processing on the aggregation interval obtained by the aggregation interval determination unit 342.
That is, for example, the singular value detection unit 344 extracts entries included in one counting cycle, and counts the number of entries in each counting section.
As a result of aggregation by the singular value detection unit 344, for example, the aggregation result shown in FIG. 45 is obtained.

In step 232 (S232), the singular value detection unit 344 performs two significant difference tests by the chi-square test on the aggregation result (FIG. 45) obtained by the process of S230.
One of these two significance tests is used to identify whether or not the aggregation section to be processed shows a unique value with respect to the average aggregation section.
In this test, the number of entries (observed value) in each counting section is X _i , the counting section number z is k, the total number of entries included in the counting period is n, and the probability included in the counting section is p = 1 / k. The following equation 1 is used.

In Equation 1, α indicates a risk factor, and is statistically usually 5% (0.05), or 1% (0.01) depending on the application of the test. is there.
When the inequality shown in Equation 1 is satisfied, it can be seen that there is a significant difference in the counting interval, indicating a singular value.

The other of the above two tests is to test whether or not the aggregation interval m to be processed shows a unique value with respect to the average aggregation interval.
In this test, the number of entries (observed value) in the total section _m is x _m , the total section number is k, the total number of entries included in the total period is n, and the probability included in the total section m is p _m = 1/1 /. The following equation 2 is used, where k is the risk factor and α is the risk factor.

  When the inequality shown in Equation 2 is satisfied, it can be seen that there is a significant difference in the aggregated interval, indicating a singular value.

In step 234 (S234), the singular value detection unit 344 determines whether there is a counting period to be tested next.
The singular value detection unit 344 returns to the process of S232 when there is a counting period to be tested next, and otherwise proceeds to the process of S226.

FIG. 46 is a diagram showing the tabulation results and test results by the singular value detection unit 344 shown in FIG.
In FIG. 46, singular values are shown in bold.
When the singular value detection unit 344 performs the test shown in FIG. 40 and the above formulas 1 and 2 on the aggregation results illustrated in FIG. 45, the test result shown in FIG. Detected.

In step 236 (S236; FIG. 40), the output processing unit 346 outputs the aggregation results and the test results (FIGS. 45 and 46) obtained by the singular value detection unit 344 to the outside via the UI unit 312.
As shown in FIG. 46, the output processing unit 346 does not necessarily output all the total results and the test results, and may output only the total results that have been verified as having a significant difference.

[Modification]
47 shows the configuration of the third analysis / evaluation unit 36 that can be used in place of the second analysis / evaluation unit 34 shown in FIG. 38 in the second analysis / evaluation program 32 shown in FIG. FIG.
As shown in FIG. 47, the third analysis / evaluation unit 36 employs a configuration in which a total period candidate detection unit 360 is added to the second analysis / evaluation unit 34.

In the analysis / evaluation unit 36, the counting cycle detection unit 360 performs, for example, a Fourier transform process on the inter-organization communication information (FIG. 39) input from the log DB 304, and is included in the frequency of occurrence of the inter-organization communication information. The period is detected and the aggregation period is derived.
By using the third analysis / evaluation unit 36 instead of the second analysis / evaluation unit 34 in the analysis / evaluation program 32, the setting of the totaling period by the user becomes unnecessary, and the inter-organization communication information Even if the fluctuation cycle does not coincide with the human activity cycle such as one year or one month, the singular value can be detected more appropriately.

[Effects, etc.]
As described above, according to the second analysis / evaluation program 32 as described above, when a function for modeling steady fluctuations occurring in the frequency of occurrence of inter-organization communication information (time-series data) is unknown. However, a singular value can be detected from time series data.
In addition, according to the analysis / evaluation program 32, a statistical test is performed for a period in which the number of data decreases due to time-series fluctuations by determining the division interval so that the expected frequency becomes equal due to fluctuations in the repeated data and reference data. Can do.

Further, according to the analysis / evaluation program 32, it is possible to perform analysis based on a period in accordance with a human activity cycle such as one day, one week, one month, one year.
Further, by using the third analysis / evaluation unit 36, it is possible to analyze a strong frequency component as a repetition period by Fourier transform or the like.

In addition, according to the analysis / evaluation program 32, a statistically significant singular value can be automatically found from the repetition of the time series data, even if the time series data has a time zone with a small number of data.
Further, according to the analysis / evaluation program 32, a statistical comparison can be made between a plurality of non-repetitive time-series data that behaves similarly in time.
Further, according to the analysis / evaluation program 32, it is possible to perform statistical comparison between data stratified from a set of time-series data.

The information processing method according to the present invention not only detects singular values from repeatedly generated time-series data, but also, for example, examines the movement of the number of access by sales employees in the entire Web access of a certain company. This time series data can also be used to determine whether or not the movement of data classified therein is unique.
Also, the number of failures of different products that show the same failure rate curve is collected as a time series on the relative time axis, and it is determined whether it is fragile compared with the reference data, etc. It is also possible to collect several time-series data that change in this way as reference data and determine whether or not the movement is unique to the data.
Similarly, for example, the present invention can be applied to the case of comparing the arrival status of customers at sales offices nationwide.

In addition, so far, each period of repeated data is taken as one time series and the average of the time series is taken as the reference value, and it is determined whether it is a singular value by whether there is a difference from the reference value However, according to the present invention, even if it is not repetitive data, if it is time series data that can be compared such as log data of an organization of the same period and log data of another organization, those data An average or comparison between them is possible.
In addition, the present invention determines whether or not log data stratified from time-series data is a singular value, such as the log data of laboratory members in the log data of the entire company. Use is also possible.

Also, the day of the week used in the above processing is merely an example, and the day of the week is replaced with a numerical value of 0 to 6, or 0:00:00 on Monday is 00:00:00 (hour: minute: second) ), 00:00 on Tuesday can be set to 24:00:00, and 0:00:00 on Sunday can be set to 114: 00: 00.
That is, in the above-described processing, the occurrence time of each entry may be converted into a relative time from the start of each counting cycle, and the entries may be rearranged.

  The present invention can be used to analyze messages transmitted between organizations and objectively evaluate the value of the organization.

It is a figure which illustrates the structure of the network system with which the evaluation method concerning this invention is applied. It is a figure which shows the hardware constitutions of the client computer, server, and analysis / evaluation apparatus which were shown in FIG. It is a figure which shows the structure of the client program which operate | moves on the client computer shown in FIG. 1, FIG. It is a figure which shows the structure of the server program which operate | moves on the server shown in FIG. 1, FIG. FIG. 5 is a diagram illustrating an e-mail message log recorded in a log DB illustrated in FIG. 4. It is a figure which shows the specific example of the message log shown in FIG. It is a figure which shows the structure of the 1st analysis and evaluation program which operate | moves on the analysis and evaluation apparatus shown in FIG. 1, FIG. It is a figure which shows the organization information which organization / personal DB shown in FIG. 7 memorize | stores. It is a figure which shows the specific example of the organization information shown in FIG. FIG. 8 is a diagram showing personal information stored in the organization / personal DB shown in FIG. 7. FIG. 11 is a diagram showing a specific example of the personal information shown in FIG. It is a figure which shows the organization communication information which a log collection and management part memorize | stores in log DB. It is a figure which shows the specific example of the organization communication information (organization communication information) shown in FIG. It is a flowchart which shows the organization communication information creation process (S24) by the log collection / management part 302 shown in FIG. It is a flowchart which shows the process (S26) which the analysis and evaluation part shown in FIG. 7 extracts and totals the word (common word) used in common as a common concept. It is a figure which shows the separate total process (S28) shown in FIG. FIG. 8 is a first diagram illustrating common word information in an organization stored in an analysis / evaluation result DB by an analysis / evaluation unit illustrated in FIG. 7. (A)-(C) are figures which show the specific example of the common word information shown in FIG. FIG. 8 is a second diagram illustrating common word information in the organization stored in the analysis / evaluation result DB by the analysis / evaluation unit illustrated in FIG. 7. It is a figure which shows the specific example of the common concept shown in FIG. It is a 1st figure which shows the communication information preparation process between organizations by the analysis and evaluation part shown in FIG. It is a figure which shows the specific example of the communication information between organizations shown in FIG. FIG. 8 is a second diagram illustrating inter-organization communication information creation processing by the analysis / evaluation unit illustrated in FIG. 7. It is a flowchart which shows the matching process (S50) in which the analysis and evaluation part shown in FIG. 7 extracts the common concept of organization communication information (FIG. 12, FIG. 13). FIG. 8 is a first diagram illustrating personal common word information stored in the analysis / evaluation result DB by the analysis / evaluation unit illustrated in FIG. 7; FIG. 8 is a second diagram illustrating individual common word information stored in the analysis / evaluation result DB by the analysis / evaluation unit illustrated in FIG. 7. FIG. 8 is a first diagram illustrating an impact evaluation result stored in an analysis / evaluation result DB by an analysis / evaluation unit illustrated in FIG. 7. It is a figure which shows the specific example of the influence range shown in FIG. It is a figure which shows the specific example of an influence degree. FIG. 8 is a second diagram illustrating an impact evaluation result stored in an analysis / evaluation result DB by the analysis / evaluation unit illustrated in FIG. 7. It is a flowchart which shows the process (S52) which calculates the influence range shown to FIG. 27, FIG. It is a flowchart which shows the process (S54) which calculates the influence degree shown in FIG. It is a figure which shows the specific example of the influence range calculated | required for every concept. It is a figure which shows the specific example of the influence degree calculated | required for every concept. It is a figure which shows the analysis and evaluation sequence (S10) in a network system. It is a figure which shows the analysis / evaluation process (FIG. 35) of an analysis / evaluation part (FIG. 7). It is a figure which shows the structure of the 2nd analysis and evaluation program which operate | moves on the analysis and evaluation apparatus shown in FIG. 1, FIG. It is a figure which shows the structure of the 2nd analysis and evaluation part shown in FIG. FIG. 39 is a diagram illustrating inter-organization communication information input from the log DB shown in FIG. 37 to the counting cycle determination unit shown in FIG. 38. It is a flowchart which shows the process (S22) of the analysis and evaluation part shown to FIG. 37, FIG. It is a figure which illustrates the fluctuation | variation of the number of entries of the totaling period contained in a totaling period, and the communication information between organizations. FIG. 42 is a diagram illustrating a reference value table that can be used in the splittable cycle selection process (S224) illustrated in FIG. It is a flowchart which shows the determination process (S26) of the division | segmentation time shown in FIG. It is a figure which illustrates the total area obtained by a total area determination part (FIG. 38) with start time and end time. FIG. 45 is a diagram illustrating a result of counting the number of entries included in the counting section illustrated in FIG. 44. It is a figure which shows the total result and test result by the singular value detection part shown in FIG. FIG. 39 is a diagram illustrating a configuration of a third analysis / evaluation unit that can be used in place of the second analysis / evaluation unit illustrated in FIG. 38 in the analysis / evaluation program illustrated in FIG. 37.

Explanation of symbols

1 ... Network system,
100 ... Network,
2 ... departmental system,
102 ... Departmental LAN,
20: Client computer,
200 ... body,
202... CPU
204: Memory,
206 ... display / input device,
208... Recording device,
210 ... recording medium,
212 ... Communication device,
22 ... Client program,
220 ... UI unit,
222 ... mail program,
224 ... Web browser,
226: LAN communication control unit,
24 ... server,
26: Server program,
260 ... LAN communication control,
262 ... Network communication control,
264 ... Mail server program,
266 ... Web server,
268: Log management unit,
3. Analysis / evaluation equipment,
30, 32 ... Analysis and evaluation program,
300: Network communication control,
302 ... Log collection / management section,
304 ... log DB,
306, 34, 36 ... analysis / evaluation unit,
340 ... Total period determination unit,
342 ... Total section determination unit,
344 ... a singular value detector,
346... Output processing unit,
360 ... counting period candidate detection unit,
308 ... Analysis / evaluation result DB,
310 ... organization / personal database,
312 ... UI unit,

Claims (15)

An information processing apparatus that performs the information processing on a plurality of pieces of processing target information that occurs over time in a predetermined period and can be an object of information processing, wherein the information processing is included in the predetermined period 1 Based on two or more processing intervals,
Depending on the mode of generation of the target information, one or more processing sections, a processing section setting means for setting a variable length,
An information processing apparatus comprising: information processing means for performing the information processing on the processing target information based on the set variable-length processing section. The information processing apparatus according to claim 1, wherein the predetermined period includes one or more periods, and the processing section setting unit sets the processing section based on a period included in the period. A fluctuation period detecting means for detecting a fluctuation period of the processing target information;
The information processing apparatus according to claim 1, wherein the processing section setting unit sets the processing section based on the detected variation period. The information processing is performed based on processing sections each including a predetermined number or more of the processing target information,
The information processing according to any one of claims 1 to 3, wherein the processing section setting means sets the processing section with a variable length so that each of the processing sections includes the predetermined number or more of the processing target information. apparatus. The information processing is performed based on processing sections each including a predetermined number or more of the processing target information,
The information processing according to claim 1, wherein the processing section setting unit sets the processing section with a variable length so that the processing section includes the same number of pieces of processing target information as expected values. apparatus. The processing target is transmitted within or between a plurality of units that can be evaluated,
Processing object information storage means for storing the processing object information;
The information processing means performs the information processing on the recorded processing target information and evaluates an influence of an evaluation target unit, which is an evaluation target, on other units among the plurality of units. The information processing apparatus according to any one of 1 to 5. The information processing apparatus according to claim 1, wherein the information processing is statistical processing. An information processing method for performing the information processing on a plurality of pieces of processing target information that occurs over time in a predetermined period and can be an object of information processing of a predetermined method, wherein the information processing is performed in the predetermined period Based on one or more processing sections included in the
Depending on the mode of occurrence of the target information, one or more of the processing sections are set with a variable length,
An information processing method for performing the information processing on the processing target information based on the set variable-length processing section. An information processing apparatus that performs the information processing on a plurality of pieces of processing target information that is generated over time in a predetermined period and can be an information processing target of a predetermined method, wherein the information processing is performed in the predetermined period Based on one or more processing sections included in the
A processing section setting step for setting one or more of the processing sections in a variable length according to the generation mode of the target information;
A program that causes a computer to execute the information processing for the processing target information based on the set variable-length processing section. The program according to claim 9, wherein the predetermined period includes one or more cycles, and the processing interval setting step sets the processing interval based on a cycle included in the period. Further causing the computer to execute a fluctuation period detecting step of detecting a fluctuation period of the processing target information;
The program according to claim 9, wherein the processing section setting step sets the processing section based on the detected fluctuation period. The information processing is performed based on processing sections each including a predetermined number or more of the processing target information,
The program according to any one of claims 9 to 11, wherein the processing section setting step sets the processing section with a variable length so that each of the processing sections includes the predetermined number or more of the processing target information. The information processing is performed based on processing sections each including a predetermined number or more of the processing target information,
The program according to any one of claims 9 to 11, wherein the processing interval setting step sets the processing interval with a variable length so that the processing interval includes the same number of pieces of processing target information as expected values. The processing target is transmitted within or between a plurality of units that can be evaluated,
A processing object information storing step for storing the processing object information is further executed by a computer;
The information processing step performs the information processing on the recorded processing target information, and evaluates an influence of an evaluation target unit, which is an evaluation target, on other units among the plurality of units. The program in any one of 9-13. The program according to any one of claims 9 to 14, wherein the information processing is statistical processing.

Images