JP2007128163A - System for evaluating relevancy between persons - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To appropriately determine a relevancy present between persons based on description of a text. <P>SOLUTION: The system for evaluating a relevancy between persons comprises a relevancy detection means detecting, based on a text expressing a certain person's evaluation or emotion for the other person, a relevancy to the other person concerned by the person concerned; a storage means storing the detected relevancy; an evaluation means evaluating, for a set of a first person and second person that is an object of evaluation, the relevancy between the first person and the second person based on a plurality of relevancies extending from the first person to the second person through at least one person, including the stored relevancy; and an output means outputting the evaluated relevancy. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a system for evaluating the relationship between persons. In particular, the present invention relates to a system that evaluates relationships between persons based on text descriptions.

  In sales strategies for products and services, there is a case where an acquaintance relationship between people called a personal connection is important. For example, when a person wants to purchase a product or service, it is effective to have the person trusted by that person recommend the product or service. However, the acquaintance relationship is complicated, and it is difficult to know who is effective when introducing someone.

As a reference technique, a technique for graphically displaying a relationship between persons has been proposed (see Patent Document 1). In this technique, association between persons is performed based on whether or not names of a plurality of persons are described in the same electronic data (for example, text). A node indicating a person is connected to a node indicating a person related to the person by an edge. A graph created by repeating this connection is displayed to the user.
JP 2005-108123 A

  However, the graph created in the above-described reference technique is an undirected graph composed of edges having no orientation. Therefore, this graph cannot represent a situation in which a certain person unilaterally respects another person or a situation in which a certain person unilaterally evaluates the personality of another person. For this reason, there has been a problem that it is difficult to use for practical purposes such as sales strategy.

  Furthermore, even when two persons are described in the same text, the two persons may be in competition with each other or may be disgusted with each other. In such a case, using the acquaintance relationship for the sales strategy has an adverse effect. As described above, in the conventional technology, it is difficult to use the acquaintance relationship shown in the graph for practical purposes such as a sales strategy.

  Therefore, an object of the present invention is to provide a system, a method, and a program that can solve the above-described problems. This object is achieved by a combination of features described in the independent claims. The dependent claims define further advantageous specific examples of the present invention.

In order to solve the above-described problem, in the present invention, a system for evaluating the relationship between persons, based on a text expressing an evaluation or feelings for another person by a person, the person concerned Relevance detection means for detecting relevance with orientation to the other person according to the above, a storage means for storing the detected relation, and a set of the first person and the second person to be evaluated, Relevance between the first person and the second person based on a plurality of relations from the first person to the second person via at least one person, including the stored relation A system is provided that includes an evaluation means for evaluating and an output means for outputting the evaluated relevance.
The above summary of the invention does not enumerate all the necessary features of the present invention, and sub-combinations of these feature groups can also be the invention.

  According to the present invention, it is possible to appropriately determine the relationship existing between persons.

  Hereinafter, the present invention will be described through embodiments of the invention. However, the following embodiments do not limit the invention according to the scope of claims, and all combinations of features described in the embodiments are included. It is not necessarily essential for the solution of the invention.

  FIG. 1 shows the overall configuration of the information processing system 10. The information processing system 10 includes a text database 20 and an evaluation system 30. The text database 20 stores a plurality of texts expressing evaluations or feelings by other persons for other persons. The text may further record an expression indicating the fact that a certain person and another person act together or information shared by a certain person and another person. The evaluation system 30 evaluates a relevance with orientation and / or a relevance without orientation based on the text stored in the text database 20 between a certain person and another person.

  FIG. 2 shows a specific example of the text database 20. The text database 20 includes a plurality of texts, for example, texts 200-1 to 200-N. The text stored in the text database 20 may be a fixed text conforming to a predetermined format, or may be an atypical text such as a memo or an e-mail. FIG. 2 illustrates a fixed text. In a fictitious company, a salesperson of the company records information heard from the customer in a text 200-1 according to a predetermined format. The example of FIG. 2 shows that a customer named Koji Yamamoto, the president of XYZ Electric, talked with ABC Kosan President Yamaguchi about a hobby of golf in a magazine in the business world. Furthermore, it is shown that the date when this information was heard was July 21, 2005.

  As in the expression “conversation”, the text 200-1 may represent the fact that a person has acted with another person. In addition, the text 200-1 may express an emotion for another person by a certain person, such as “good friend”. The evaluation system 30 in the present embodiment evaluates the relationship between persons based on these expressions. This makes it possible for a person to have a relevance, such as when a person and another person trust each other, or when a person respects another person, but not vice versa. The purpose of this is to evaluate appropriately. Furthermore, it is an object to appropriately evaluate a state in which the relationship between persons cannot be practically used, such as when a person has a negative feeling toward another person.

  FIG. 3 shows a functional configuration of the evaluation system 30. The evaluation system 30 includes relevance detection means 300, storage means 310, attribute detection means 320, evaluation means 330, and output means 340. Based on the text stored in the text database 20, the relevance detection unit 300 detects a relevance that has a direction with respect to another person and / or a relevance that has no orientation. Specifically, the relevance detection means 300 detects a person name from the text. Then, for each person whose name is detected, another person whose name appears in the same text is detected together with that person. Then, the relevance detection means 300 detects an expression that suggests such relevance from the text in which a person's name appears.

  This expression is, first of all, an evaluation or emotion for a person by another person. For example, emotional expressions include “like”, “trust”, “respect”, “thank you”, or “care for”. The expression of evaluation is not limited to general evaluations such as “I can work” or “I have a good personality”, but also “specific to science and technology” or “detailed to the stock market”. Includes an assessment of the field. In this case, the relevance detection unit 300 may detect the evaluation in association with the specific field.

  Then, the relevance detection unit 300 determines that there is a relevance having a direction from the person to the other person when an expression representing an evaluation or feeling for the other person by the person is detected. . That is, for example, when an expression indicating that the person A respects the person B is detected, it is determined that there is a relevance having a direction from the person A to the person B. The relevance detection unit 300 may generate a directed graph having a person as a node and a relevance as an edge as data indicating such a relevance. A specific example is illustrated in FIG.

  In addition, the relevance detection means 300 is based on the fact that a certain person and another person acted together or the text expressing information shared by a certain person and the other person, and the person and the other person. You may detect the relationship without direction. That is, for example, when an expression indicating that the person A and the person B have talked to each other is detected, it is determined that there is an unoriented relationship between the person A and the person B. The relevance detection unit 300 may generate an undirected graph with a person as a node and a relevance as an edge as data indicating the relevance with no orientation. This undirected graph may be generated by adding an undirected edge to the previously described directed graph.

  The relevance detection unit 300 may further detect a weight indicating the strength of relevance based on the frequency at which the evaluation or emotion is expressed in the text. Specifically, the relevance detection means 300 uses the expression in which a certain evaluation or emotion is expressed when the frequency included in the text database 20 is higher than when the frequency is lower. Greatly evaluate the relevance weight based on. Furthermore, the relevance detection unit 300 may evaluate the weight indicating the strength of relevance based on the time when the evaluation or emotion is expressed in the text. Specifically, the relevance detection means 300, when the time when the evaluation or emotion is expressed in the text is more recent, the relevance of the relevance based on the expression is compared with the case when the time is earlier. The weight may be greatly evaluated.

  The storage unit 310 stores information indicating the relationship detected by the relationship detection unit 300 and its weight, and provides the information to the evaluation unit 330. In addition, the storage unit 310 stores in advance the association between each of a plurality of attribute values and each of the other attribute values for the person's attributes. For example, the attribute of a person indicates an organization to which the person belongs. The attribute value in this case indicates identification information such as the name of the organization. That is, for example, the storage unit 310 stores information such as whether the company ABC Inc. and the company XYZ Electric have a competitive relationship or a partnership relationship as relevance. Furthermore, the storage unit 310 may store the relevance of the attribute value in association with the time when the relevance is evaluated.

  The attribute detection unit 320 detects the attribute of each person expressed in the text stored in the text database 20. The evaluation unit 330 evaluates the relationship between the first person and the second person to be evaluated based on the relationship detected by the relationship detection unit 300 based on the relationship detected by the relationship detection unit 300. Specifically, first, the evaluation unit 330 selects a plurality of relationships from the first person to the second person via at least one other person. The plurality of relationships include relationships detected by the relationship detection unit 300 and stored in the storage unit 310. Then, the evaluation unit 330 evaluates the relationship between the first person and the second person based on the selected plurality of relationships. For example, when the relevance indicates a positive evaluation or emotion and the relevance is indicated as a directed graph, the evaluation means 330 traces the directed graph in the reverse direction and the plurality of paths on the route to be reached Based on the relationship, the relationship between the first person and the second person may be evaluated. When the relevance weight is detected, the evaluation unit 330 may evaluate the relevance based on the relevance weight. Furthermore, the evaluation unit 330 may evaluate the relationship between the persons based on the relationship between the attribute of the person detected for each person and the attribute value.

  Note that instead of the above example, the evaluation unit 330 may evaluate not only the relationship between persons but the relationship between a plurality of organizations. Specifically, the relevance detection unit 300 detects the relevance of the orientation of the organization with respect to the other organization based on the text expressing the evaluation or emotion of the other organization as a target. . This text includes, for example, a statement made by a certain organization, a statement such as a press release, and a statement made by a representative of the organization when the organization is a company. Then, the evaluation unit 330 includes the detected relationship between the first organization and the second organization set to be evaluated from the first organization via at least one organization. Assess the relationship between the first organization and the second organization based on multiple relationships leading to the organization. As described above, according to the evaluation system 30 according to the present embodiment, not only a person but also an entity that can express evaluation or emotion in text, for example, an organization such as a company, can be targeted for evaluation.

  The output unit 340 outputs the relevance evaluated by the evaluation unit 330. The output unit 340 performs output based on the relevance. For example, the output unit 340 may search for a person on a path that reaches the first person from the first person by reversing the relevance in the opposite direction, and the first person as the second person. It may be output as an introducer who can introduce. This path may include an unoriented relationship. As a result, it is possible to support effective sales activities utilizing personal connections. That is, for example, when a product or service is to be sold to a certain person A, the person B trusted by the person A can be selected as an introducer, so that the business activities can be smoothly advanced.

  FIG. 4 shows an example of the data structure of the storage unit 310. The storage unit 310 stores in advance the association between each of the plurality of attribute values and each of the other attribute values. For example, the storage unit 310 stores a relevance weight of +0.5 for a set of ABC Kosan, which is the first attribute value, and XYZ Electric, which is the second attribute value. If this relationship indicates a friendly relationship between these companies, the weight of +0.5 is a positive value, so there is a certain amount of friendly relationship between ABC Kosan and XYZ Electric. . In addition, the storage unit 310 stores the relevance of the attribute value in association with the time when the relevance is evaluated. For example, the example of FIG. 4 shows that the association between ABC Kosan and XYZ Electric was evaluated on April 25, 2005. This evaluation time may be the time when information indicating the relevance is input to the storage unit 310, or may be the creation time of a text in which an expression suggesting the relevance is described.

  FIG. 5 shows an example of processing for evaluating the relationship between persons by the evaluation system 30. The relevance detection unit 300 acquires text from the text database 20 (S500). The relevance detection means 300 detects a person name from the acquired text (S510). The relevance detection means 300 detects, for each person of the detected person name, other persons whose person names appear in the same text together with the person, and further detects expressions that suggest relevance of these persons ( S520). Based on this expression, the relevance detection means 300 detects a relevance with a direction from one person to another and / or a relevance without a direction. Further, the attribute detection unit 320 detects the attribute of each person expressed in the text stored in the text database 20 (S530).

  The evaluation unit 330 evaluates the relationship between the first person and the second person to be evaluated based on the relationship detected by the relationship detection unit 300 (S540). ). A specific example of the evaluated relevance will be described with reference to FIG. FIG. 6 shows an example of the relationship evaluated by the evaluation system 30. In this figure, each rectangular area is a node indicating a person, an edge having an arrow only at one end indicates a directed edge, and an edge having an arrow at both ends indicates an undirected edge.

  For example, first, the evaluation means 330 selects a plurality of relationships from Mr. Abe to Mr. Koji Yamamoto via at least one other person. For example, the evaluation means 330 selects a positive relevance with a weight of 0.2 and a positive relevance with a weight of 0.5 on the path from Mr. Abe to Mr. Koji Yamamoto via President Yamaguchi. . On the other hand, on the route from Mr. Abe to Mr. Koji Yamamoto via Mr. Senichi Hoshino, the evaluation means 330 shows a positive relevance with a weight of 0.8 and a negative relevance with a weight of 0.5. select. These selected relationships include the relationships detected by the relationship detection means 300. That is, for example, as a result of searching for a text that expresses a positive feeling or evaluation of Mr. Koji Yamamoto to President Yamaguchi, a relevance that is directed from Mr. Koji Yamamoto to President Yamaguchi may be detected.

  Next, the evaluation means 330 evaluates the relationship between Director Abe and Mr. Koji Yamamoto based on these selected relationships. Specifically, the evaluation unit 330 may evaluate the total or average value of these selected relevance weights as the relevance between Mr. Abe and Mr. Koji Yamamoto. That is, the evaluation means 330 evaluates the relevance via President Yamaguchi as 0.7 by 0.5 + 0.2, and the relevance via Mr. Senichi Hoshino as 0.8 by (+0.5). It may be evaluated as 3. According to the evaluated relevance, it is clear that Director Abe prefers to contact Mr. Koji Yamamoto via President Yamaguchi rather than Mr. Senichi Hoshino.

  Further, the evaluation unit 330 may select a plurality of relationships including the relationship between the attribute values on the route from Mr. Abe to the attribute value of Mr. Koji Yamamoto via the attribute value. For example, the evaluation unit 330 selects the relevance on the path from GDG communication, which belongs to the manager of Abe, to ABC Kosan, which belongs to Mr. Koji Yamamoto. That is, the evaluation unit 330 selects a positive association of 0.4 weight between GDG communication and XYZ Electric and a negative association of 1.0 weight between XYZ Electric and ABC Kosan. To do. And the evaluation means 330 may evaluate the sum total (namely, -0.6) of these weights as a relationship between Director Abe and Mr. Koji Yamamoto. As described above, the evaluation unit 330 has the first weight when the relevance weight between the attribute of the first person and the attribute value of the second person is larger than that when the weight is smaller. The weight of the relationship between the person and the second person may be evaluated more greatly.

  In the above processing, a plurality of different relevance weights may be evaluated for the same combination of persons. For example, as the weight of the relationship between Mr. Koji Yamamoto with President Yamaguchi and Mr. Abe as the mediator, +0.7 based on the personal relationship and -0.6 based on the affiliated organization are evaluated. Yes. In such a case, for example, the evaluation unit 330 may evaluate the smaller one of the plurality of relevance weights as the relevance weight. As a result, the relevance information can be used conservatively to avoid the risk of failure of sales activities.

Furthermore, when evaluating the relevance using attribute values, it is desirable to use the time when the relevance is evaluated and the nature of the relevance. For example, the evaluation unit 330 may be configured such that when the time when the relationship between the attribute value of the first person and the attribute value of the second person is evaluated is newer than when the time is older, The weight of the relationship between the first person and the second person may be evaluated more greatly. The timing of this evaluation is obtained from the information stored in the storage means 310 shown in FIG. Further, in the case where the attribute indicates the organization to which the person belongs, the evaluation unit 330 determines that the attribute between the person and the other person is greater when the organization to which the person belongs is smaller than when the organization to which the organization belongs is smaller. The relevance based on may be evaluated smaller. The size of the affiliated organization is obtained from the number of links between the attribute value indicating the affiliated organization and the node indicating the person.
According to the above example, the relevance weight between an attribute value and another attribute value can be appropriately evaluated based not only on text data but also on the property of each attribute value.

The output unit 340 performs output based on the relationship evaluated by the evaluation unit 330 (S550). Hereinafter, four application examples will be described using the example of FIG.
(1) Display of introducer The output unit 340 outputs an introducer who introduces a salesperson to a certain person based on the relevance. Generally, in order to effectively perform such sales activities, it is important to select a person who is trusted by the person as an introducer. The output unit 340 can select such an introducer based on the relationship between persons evaluated by the evaluation unit 330.

  Specifically, first, it is assumed that the evaluation unit 330 evaluates a relevance with a direction indicating a positive evaluation or feelings from one person to another. For example, in the example of FIG. 6, it is detected that Mr. Koji Yamamoto has positive evaluation or feelings about President Yamaguchi. Then, the evaluation unit 330 traces the relevance with the direction from the first person (Director Abe) to the second person (Mr. Koji Yamamoto), or the relevance without the direction. Trace and select a person on the route to reach. Then, the evaluation unit 330 evaluates the selected person as an introducer who can introduce the first person to the second person. In other words, in this example, Senichi Hoshino and President Yamaguchi are evaluated as introducers. The output means 340 outputs information indicating these introducers. Preferably, the output unit 340 preferentially outputs each of the plurality of introducers in descending order of the relevance weight in the relevance route passing through the introducer. For example, the output unit 340 may display Mr. Yamaguchi, whose relevance weight total value is 0.7, more prominently than Mr. Senichi Hoshino, whose relevance weight total value is 0.3. However, only President Yamaguchi may be displayed.

(2) Display of group The evaluation unit 330 detects a group composed of a plurality of persons who are strongly related to each other, and the output unit 340 displays the detected group to the user. Specifically, the evaluation unit 330 detects each strongly related component in the graph in which each person is a node and each relevance is an edge and regards the relevance weight as the edge weight, thereby detecting the relevance of each other. A plurality of strong persons may be detected as a group. In this case, the weight of relevance between a certain person and another person having no orientation is the weight of the relevance that has a direction from the person to the other person and the direction from the other person to the person. Assigned to the relevance weight. This makes it possible to apply an existing technique for detecting strongly connected components from a weighted directed graph, and to appropriately detect a plurality of persons who are strongly related to each other.

(3) Display of the central person When the relevance indicates a positive evaluation or emotion from a certain person to another person, the evaluation unit 330 is a value based on the sum of relevance weights from another person to the certain person. For example, a total value or an average value is obtained. Then, the evaluation unit 330 evaluates the person as a central person in the personal relationship on the condition that the value based on the sum is larger than a predetermined reference value. Thereby, the person trusted and evaluated by many people can be evaluated appropriately.

(4) Sales promotion of products etc. When history information of products purchased by each person in the past can be obtained as attribute information of the person, by combining the history information with the relationship between the people, It is considered effective for sales promotion. Specifically, the output unit 340 outputs that the product or service purchased by any person belonging to the group detected in the example of (2) should be recommended to other persons belonging to the group. To do.

  In addition, a product or service of the same type as a product or service purchased by any person belonging to the group may be recommended to another person belonging to the group. Specifically, for example, the storage unit 310 or the like stores a product or service of the same type as the product or service in association with each product or service. Then, the output unit 340 selects a product or service of the same type as the product or service purchased by any person belonging to the group based on the information stored in the storage unit 310 or the like. Then, the output means 340 outputs that the product or service should be recommended to other persons belonging to the group. In this way, if the relationship between persons is utilized in sales activities such as products, a person who has not been assumed as a sales destination can be selected as an effective sales destination.

  FIG. 7 shows an example of the hardware configuration of the information processing apparatus 400 that functions as the evaluation system 30. The information processing apparatus 400 includes a CPU peripheral unit including a CPU 1000, a RAM 1020, and a graphic controller 1075 connected to each other by a host controller 1082, a communication interface 1030, a hard disk drive 1040, and the like connected to the host controller 1082 by an input / output controller 1084. And an input / output unit having a CD-ROM drive 1060, and a legacy input / output unit having a BIOS 1010, a flexible disk drive 1050, and an input / output chip 1070 connected to the input / output controller 1084.

  The host controller 1082 connects the RAM 1020 to the CPU 1000 and the graphic controller 1075 that access the RAM 1020 at a high transfer rate. The CPU 1000 operates based on programs stored in the BIOS 1010 and the RAM 1020 and controls each unit. The graphic controller 1075 acquires image data generated by the CPU 1000 or the like on a frame buffer provided in the RAM 1020 and displays it on the display device 1080. Alternatively, the graphic controller 1075 may include a frame buffer that stores image data generated by the CPU 1000 or the like.

  The input / output controller 1084 connects the host controller 1082 to the communication interface 1030, the hard disk drive 1040, and the CD-ROM drive 1060, which are relatively high-speed input / output devices. The communication interface 1030 communicates with an external device via a network. The hard disk drive 1040 stores programs and data used by the information processing apparatus 400. The CD-ROM drive 1060 reads a program or data from the CD-ROM 1095 and provides it to the RAM 1020 or the hard disk drive 1040.

  The input / output controller 1084 is connected to the BIOS 1010 and relatively low-speed input / output devices such as the flexible disk drive 1050 and the input / output chip 1070. The BIOS 1010 stores a boot program executed by the CPU 1000 when the information processing apparatus 400 is activated, a program depending on the hardware of the information processing apparatus 400, and the like. The flexible disk drive 1050 reads a program or data from the flexible disk 1090 and provides it to the RAM 1020 or the hard disk drive 1040 via the input / output chip 1070. The input / output chip 1070 connects various input / output devices via a flexible disk 1090 and, for example, a parallel port, a serial port, a keyboard port, a mouse port, and the like.

  The program provided to the information processing apparatus 400 is stored in a recording medium such as the flexible disk 1090, the CD-ROM 1095, or an IC card and provided by the user. The program is read from the recording medium via the input / output chip 1070 and / or the input / output controller 1084, installed in the information processing apparatus 400, and executed. The operation that the program causes the information processing apparatus 400 to perform is the same as the operation in the evaluation system 30 described with reference to FIGS.

  The program shown above may be stored in an external storage medium. As the storage medium, in addition to the flexible disk 1090 and the CD-ROM 1095, an optical recording medium such as a DVD or PD, a magneto-optical recording medium such as an MD, a tape medium, a semiconductor memory such as an IC card, or the like can be used. Further, a storage device such as a hard disk or a RAM provided in a server system connected to a dedicated communication network or the Internet may be used as a recording medium, and the program may be provided to the information processing apparatus 400 via the network.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

FIG. 1 shows the overall configuration of the information processing system 10. FIG. 2 shows a specific example of the text database 20. FIG. 3 shows a functional configuration of the evaluation system 30. FIG. 4 shows an example of the data structure of the storage unit 310. FIG. 5 shows an example of processing for evaluating the relationship between persons by the evaluation system 30. FIG. 6 shows an example of the relationship evaluated by the evaluation system 30. FIG. 7 shows an example of the hardware configuration of the information processing apparatus 400 that functions as the evaluation system 30.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Information processing system 20 Text database 30 Evaluation system 200 Text 300 Relevance detection means 310 Storage means 320 Attribute detection means 330 Evaluation means 340 Output means 400 Information processing apparatus

Claims (20)

A program for causing an information processing device to function as a system for evaluating the relationship between persons,
The information processing apparatus;
Relevance detection means for detecting a relevance of the person with respect to the other person based on a text expressing an evaluation or feeling for the other person by the person,
Storage means for storing the detected association;
For the set of the first person and the second person to be evaluated, the second person via the at least one person from the first person including the relationship stored in the storage means Evaluation means for evaluating the relation between the first person and the second person based on a plurality of relations leading to
A program that functions as output means for outputting the evaluated relevance. The relevance detection means indicates the strength of the relevance based on the frequency at which the evaluation or the emotion is expressed in the text and / or the time at which the evaluation or the emotion is expressed in the text. Detect more weights,
The program according to claim 1, wherein the evaluation unit evaluates an association between the first person and the second person based on the detected weight. The relevance detection means detects the strength of the relevance as a positive weight when the evaluation or the emotion is expressed, and the evaluation or the emotion opposite to the evaluation or the emotion is expressed. The program according to claim 2, wherein the strength of the relationship is detected as a negative weight. The relevance detection means further includes the fact that the person and the other person acted together or the text expressing the information shared by the person and the other person and the person and the other person. Detect unrelated relationships,
The program according to claim 2, wherein the evaluation unit evaluates the relationship between the first person and the second person based on the detected relationship without the orientation. In the case where the orientational relevance indicates a positive evaluation or emotion from one person to another,
The evaluation means determines a person on a route to reach the second person from the first person by sequentially tracing the relevance in the opposite direction, and the first person as the second person. The program according to claim 4, wherein the program is evaluated as an introducer who can introduce the person.   The program according to claim 5, wherein the output unit further outputs each of the plurality of introducers with priority in descending order of the relevance weight in the relevance route passing through the introducer. The evaluation means further determines the weight of the non-directional relationship between a person and another person, the weight of the relevance of the direction from the person to the other person, and the weight from the other person. By assigning to the weight of the relevance with orientation to the person, the strongly connected component of the directed graph with each person as the node and the relevance as the weighted edge is detected, and the set of nodes included in the detected strongly connected component is determined. The program according to claim 4, wherein the program is detected as a group composed of a plurality of persons. In the case where the relationship indicates a positive evaluation or feeling from one person to another,
The evaluation means further evaluates the person as a central person in the personal relationship on the condition that a value based on the sum of the relevance weights for the person from another person is larger than a predetermined reference value. The program according to claim 4. The information processing apparatus;
Further functioning as attribute detection means for detecting the attribute of each person represented in the text,
The storage means stores in advance a relationship between each of the plurality of attribute values and each of the other attribute values,
In the case where the orientational relevance indicates a positive evaluation or emotion from one person to another,
In the evaluation means, when the weight of the relationship between the attribute value of the first person and the attribute value of the second person is larger, the weight is smaller than the case where the weight is smaller. The program according to claim 4, wherein the weight of the relationship between one person and the second person is more greatly evaluated. The storage means stores the association of attribute values in association with the time when the association was evaluated,
In the case where the evaluation time of the association between the attribute value of the first person and the attribute value of the second person is newer, the evaluation means compares it with the case where the time is older. The program according to claim 9, wherein the weight of the relationship between the first person and the second person is more greatly evaluated. In the case where the attribute indicates the organization to which the person belongs,
The evaluation means evaluates the relevance of the person and another person based on the attribute smaller when the belonging organization of a person is larger than when the belonging organization is smaller. The program described in. The evaluation means detects a strongly connected component of a graph with each person as a node and a relationship as an edge, and detects a set of nodes included in the detected strongly connected component as a group composed of a plurality of persons. ,
In the case where the attribute indicates a product or service purchased by a person,
The program according to claim 10, wherein the output unit outputs that a product or service purchased by any person belonging to the detected group should be recommended to another person belonging to the group. The storage means stores a product or service of the same type as the product or service in association with the product or service,
13. The output means outputs a message indicating that a product or service of the same type as a product or service purchased by any person belonging to the detected group should be recommended to another person belonging to the group. The program described in. A method for evaluating the relationship between persons,
Detecting a relevance of the person with respect to the other person based on a text expressing an evaluation or feelings of the person for the other person; and
For a set of the first person and the second person to be evaluated, a plurality of the relations from the first person to the second person via at least one person, including the detected association. Evaluating the relevance between the first person and the second person based on relevance. In the step of detecting the relevance, the strength of the relevance based on the frequency at which the evaluation or the emotion is expressed in the text and / or the time at which the evaluation or the emotion is expressed in the text. Further detecting a weight indicating
The method of claim 14, wherein in the step of evaluating, an association between the first person and the second person is further evaluated based on the detected weights. In the step of detecting the relevance, when the evaluation or the emotion is expressed, the strength of the relevance is detected as a positive weight, and the evaluation or the emotion opposite to the evaluation or the emotion is expressed. The method according to claim 15, wherein the strength of relevance is detected as a negative weight. In the step of detecting the relevance, the person and the other person are further determined based on the fact that the person and the other person acted together or the text expressing the information shared by the person and the other person. Detect unrelated relationships between people,
The method of claim 15, wherein in the step of evaluating the relevance, a relevance between the first person and the second person is further evaluated based on the detected non-orientated relevance. In the case where the orientational relevance indicates a positive evaluation or emotion from one person to another,
In the step of evaluating the relevance, a person on the route that reaches the relevance sequentially from the first person to the second person in the reverse direction is represented as the first person. The method according to claim 17, wherein the method is evaluated as an introducer who can introduce the second person. A system for evaluating the relationship between persons,
Relevance detection means for detecting a relevance of the person with respect to the other person based on a text expressing an evaluation or feeling for the other person by the person,
For a set of the first person and the second person to be evaluated, a plurality of the relations from the first person to the second person via at least one person, including the detected association. A system comprising: an evaluation unit that evaluates a relationship between the first person and the second person based on a relationship. A program that allows an information processing device to function as a system for evaluating relationships between organizations,
The information processing apparatus;
Relevance detecting means for detecting a relevance of the orientation of the organization with respect to the other organization based on a text expressing an evaluation or feeling for the other organization by the organization;
A plurality of associations from the first organization through the at least one organization to the second organization, including the detected association for the first organization and the second organization set to be evaluated A program that functions as an evaluation unit that evaluates the relationship between the first organization and the second organization based on sex.

Images