JP2018190017A - TECHNICAL MAP GENERATION DEVICE, TECHNICAL MAP GENERATION METHOD, AND PROGRAM - Google Patents

Abstract

A technical map generation device, a technical map generation method, and a program capable of generating an item generated by collecting items other than the meaning of an item such as “others” are provided. A technology map output device 1 includes a document information storage unit 111 in which a document identifier for identifying a document and document information is stored, and one or more axes constituting a technology map by analyzing two or more pieces of document information. An item acquisition unit 12 that acquires an item term that is a term that can be an item, a classification unit 13 that determines the correspondence between the item term and document information, and two or more items that satisfy a predetermined condition using the result of the classification unit And an item aggregating unit 14 for changing the term into one changed item term. The document information corresponding to the item terms aggregated by the item aggregation unit 14 is associated with the changed item terms. It further includes a technology map configuration unit 15 that configures a technology map, and a technology map output unit 16 that outputs the technology map. [Selection] Figure 1

Description

  The present invention relates to a technical map generation apparatus that outputs a technical map from a document.

  In order to analyze many technical documents such as patents and papers, a bubble chart as shown in FIG. Such a diagram is called a technology map. This bubble chart is a diagram in which two analysis viewpoints are taken on two vertical and horizontal axes, items are arranged on each axis, and bubbles of a size based on the number of documents related to the items are arranged at the intersections of the items. . For example, it is possible to give an overview of what kind of technology is being developed by creating a technology map centered on “issues” and “solving means” with an analysis object as a patent. The technology map is not necessarily a bubble chart, and may be a diagram in which the number of documents is plotted instead of the bubble in FIG. 15A, and is an intersection of items as shown in FIG. It may be a table in which specific document names are plotted in cells. The technical map may be a three-dimensional bubble chart in which three analysis viewpoints are the X axis, the Y axis, and the Z axis.

  Conventionally, a part in which “issue” or “solution” is described from a technical document (for example, [Effect of invention] for a patent document) is specified, and the contents of the specified part of a plurality of technical documents are specified. There is a system that performs clustering and automatically generates a technical map (see, for example, Patent Document 1).

  On the other hand, there has been an apparatus that interactively classifies documents by reflecting correction of classification results performed by a user in a mechanical document classification method in a tabular technical map (see, for example, Patent Document 2).

JP 2007-108867 A (first page, FIG. 1 etc.) JP 2008-305268 A (first page, FIG. 1 etc.)

  However, in the system such as Patent Literature 1 and the device such as Patent Literature 2, there is a problem that the items of the axis cannot be grouped using the result of classification of the technical map (for example, the items of “Others”). was there.

  According to the first aspect of the present invention, there is provided a technical map generation apparatus configured to construct a technical map by analyzing a document information storage unit storing two or more pieces of document information corresponding to a document identifier for identifying a document, and each of the two or more pieces of document information. An item acquisition unit that acquires an item term that can be an item of one or more axes for each axis, and a classification unit that determines a corresponding item term for each axis for two or more pieces of document information An item aggregation unit that changes two or more item terms satisfying a predetermined condition to one item term after change using the result in the classification unit for one or more axes, The document information corresponding to one of the two or more original item terms of the changed item term acquired by the aggregation unit is associated with the changed item term, and is aggregated from the item terms for each axis acquired by the item acquisition unit. Item terms excluding two or more item terms A technical map centered on a set of item terms having changed item terms, a technical map component that constitutes a technical map that clearly indicates the correspondence of each item information of two or more pieces of document information, and a technical map, A technical map generation apparatus further comprising a technical map output unit for outputting.

  With this configuration, it is possible to generate a technical map in which items are aggregated using the classification result.

  Further, the technical map generation device of the second invention further includes a term dictionary in which term groups of two or more groups are stored with respect to the first invention, and the item acquisition unit includes each of two or more terms. Analyze document information, and from each document information, an important term acquisition unit that acquires important terms satisfying a predetermined condition for each axis, and two or more important terms acquired by the important term acquisition unit for each axis, It is a technical map generation device that includes term aggregating means for referring to a term dictionary, aggregating into terms representing groups, and acquiring item terms.

  With this configuration, it is possible to automatically acquire axis items from a technical document and generate a technical map in which those items are collected.

  In the technical map generation device according to the third aspect of the present invention, in the first or second aspect, the item aggregating unit sets a predetermined condition for one or more axes using the result in the classification unit. This is a technical map generation device that changes an item term with a smaller number of corresponding document information to a single item term after change as it is satisfied.

  With this configuration, it is possible to generate a technical map in which items with a small number of corresponding document information are collected in the result of the classification unit.

  In addition, in the technical map generation device according to the fourth aspect of the present invention, for any one of the first to third aspects, the item acquisition unit analyzes two or more pieces of document information, and forms two technical maps. An item term that is a term that can be an item of each axis of the axis is acquired for each axis, and the classification unit determines the item term to which each axis belongs for each of two or more pieces of document information, and a technical map. The configuration unit is a technical map generation device that configures a two-dimensional technical map.

  With this configuration, it is possible to generate a technical map that is compiled using the classification result for each axis in the biaxial technical map.

  Further, in the technical map generation apparatus according to the fifth invention, in contrast to the fourth invention, the item aggregating unit uses two or more conditions that satisfy a predetermined condition using the result of the classification unit for each of the two axes. It is the technical map production | generation apparatus which changes the item term of 1 to the item term after a change.

  With this configuration, it is possible to generate an item in which two or more items are collected on both axes in a two-dimensional technical map.

  Further, the technical map generation device of the sixth invention is a technical map generation device in which the changed item term is a term meaning “others” with respect to any one of the first to fifth inventions.

  With this configuration, it is possible to collect items and generate a technical map including the item “others”.

  According to the technical map generation device of the present invention, it is possible to generate a technical map in which items are aggregated using the classification result.

1 is a block diagram of a technical map generation apparatus according to Embodiment 1 of the present invention. The flowchart which shows operation | movement of the technical map production | generation apparatus in the embodiment The flowchart which shows the operation | movement of the item acquisition process in the embodiment The flowchart which shows the operation | movement of the important term acquisition process in the embodiment Flowchart showing the operation of the aggregate term candidate acquisition process in the same embodiment The flowchart which shows the operation | movement of the collective term determination process in the embodiment The flowchart which shows the operation | movement of the classification process in the embodiment Flow chart showing operation of item aggregation / reclassification processing in the embodiment The flowchart which shows the operation | movement of the technical map structure process in the embodiment The figure which shows an example of the document information management table | surface in the embodiment The figure which shows an example of the term group management table | surface in the embodiment The figure which shows an example of the technical map in the embodiment Overview of the computer system Block diagram of the computer system Example of technology map in the prior art

  Hereinafter, embodiments of a technical map generation device and the like will be described with reference to the drawings. In addition, since the component which attached | subjected the same code | symbol in embodiment performs the same operation | movement, description may be abbreviate | omitted again.

(Embodiment 1)
In the present embodiment, the terms constituting the axis of the technical map are automatically acquired for each axis from a set of document information, the document information belonging to each item is determined, and items satisfying a predetermined condition are The technical map generation apparatus 1 to be combined into one item such as “others” will be described.

  In the present invention, the terms constituting an item represent one or more consecutive words, and the first and last words need not be independent words. That is, the terms constituting the item may be a single word or a phrase.

  FIG. 1 is a block diagram of a technical map generation apparatus 1 in the present embodiment.

  The technical map generation device 1 includes a storage unit 11, an item acquisition unit 12, a classification unit 13, an item aggregation unit 14, a technical map configuration unit 15, and a technical map output unit 16.

  The storage unit 11 includes a document information storage unit 111 and a term dictionary 112.

  The item acquisition unit 12 includes an important term acquisition unit 121 and a term aggregation unit 122.

  The document information storage unit 111 stores two or more pieces of document information corresponding to a document identifier for identifying a document.

  Here, the document may be anything as long as it can be classified on a technical map such as a patent document or a paper. The patent document may be an open patent gazette, a patent publication gazette, or a document at the time of filing a patent, and its format is not limited. In addition to patent documents and papers, for example, product introduction web pages, newspaper articles including product information, company press releases, and the like may be used.

  The document identifier is a symbol string that identifies a document. The document identifier may be any symbol string as long as the document can be uniquely determined in the document information storage unit 111. For example, it is preferable to use a patent application number or a patent publication number if the document is an open patent gazette, or a patent number if the document is a patent publication gazette. Further, the structure of the document information corresponding to the document identifier is not limited. Usually, the document information corresponding to the document identifier includes text. Further, the document information may include information regarding the structure of the document. For example, a markup language such as HTML may be used to represent the text and its structure, and the document structure may be represented by corner brackets as in a patent document. Further, the document information may include meta information of the document. The meta information is, for example, information related to a document other than text, such as an applicant, an inventor, an agent, an application date, a publication date, IPC, FI, and F-term in a patent document.

  The term dictionary 112 stores one or more term groups. Here, the term group represents a set of two or more terms. Each term group forms one group. The group of terms forming the group may be a set of terms having similar or identical meanings, or may be a set of terms having the same superordinate concept.

  In addition, for each group, the term dictionary 112 preferably stores a term representing the group in addition to the term group (hereinafter, the term representing the term group is referred to as a representative term). At this time, the term dictionary 112 preferably stores information indicating the relationship between the representative term and the term group. Here, the information indicating the relationship is usually synonymous or synonymous, or either higher or lower.

  The item acquisition unit 12 analyzes two or more pieces of document information and acquires one or more item terms for each axis. Here, the item term is a term that can be an item of the axis. The item term may be a single word or a phrase as described above. The document information is document information stored in the document information storage unit 111. The axis means a standard for analyzing the technical map. The axis may be referred to as a “viewpoint” corresponding to the vertical axis or horizontal axis of the technology map. For example, in the case of the two-dimensional technical map of FIG. 15A, “issue” and “solution” are the axes. The axis is not limited to “issue” or “purpose” or “solution”, but may be “material”, “content”, “part”, “part”, or the like.

  Further, the axis item is a character string, a symbol, or an image arranged on the axis of the technical map. For example, in the case of the two-dimensional technical map of FIG. 15A, the items are “strong wind countermeasures” and “thunderbolt countermeasures” on the problem axis, and “pitch control” and “variable speed control” on the solution means axis.

  For example, the item acquisition unit 12 extracts an important term for each piece of document information using the important term acquisition unit 121 described later for each piece of document information stored in the document information storage unit 111. Thereafter, the important terms extracted for each piece of document information are aggregated by the term aggregation unit 122 described later, and the item terms are acquired.

  The item acquisition unit 12 may acquire a smaller number of item terms than the total number of document information stored in the document information storage unit 111. That is, the item acquisition unit 12 may acquire item terms so that there is document information that does not have any corresponding item terms. In this case, document information having no corresponding item term is usually not displayed on the technical map. The item acquisition unit 12 may acquire more item terms than the total number of document information stored in the document information storage unit 111. That is, the item acquisition unit 12 may acquire item terms so that two or more item terms correspond to one document information.

  The important term acquisition unit 121 acquires important terms satisfying a predetermined condition from each document information for each axis. Important terms may be simply terminology. An important term is a term that satisfies a predetermined condition. The term here is preferably one word, but may be a phrase including two or more terms.

  Here, the predetermined condition may be any condition as long as the term corresponding to the axis can be acquired from the document information.

  For example, the predetermined condition may be a condition that any term of all term groups stored in the term dictionary 112 included in the document information is acquired.

  For example, the important term acquisition unit 121 may acquire a term from a sentence that matches the sentence pattern. More specifically, for example, in the case of an axis such as “issue” or “effect”, “a word appearing immediately before“ can be ”” in a sentence in [the effect of the invention] of the patent specification may be acquired. . At this time, there may be a plurality of sentence patterns. Further, it is assumed that the sentence pattern information is stored in the storage unit 11, for example.

  For example, the predetermined condition may be a condition obtained by adding a condition for the number of appearances in the document information to the condition. For example, a condition such as “a word that appears more than once among the words that appear immediately before“ can be ”in the patent specification” or “a term that is included in all term groups stored in the term dictionary 112 It is a condition such as “appeared 3 or more times”.

  Further, for example, the predetermined condition may be a condition using machine learning for identifying a sequence, which is used for morphological analysis or specific expression extraction. For example, two or more sets of information of a text and a part (important term) extracted from the text are prepared in advance, and a model such as a hidden Markov model (HMM), a conditional random field (CRF), or a recurrent neural network (RNN) is prepared. Further, learning may be performed using the two or more pieces of information, and important terms may be extracted from the document information using the learned model. Here, since machine learning and morphological analysis, specific expression extraction, HMM, CRF, and RNN are existing technologies, description thereof is omitted.

  Further, for example, the predetermined condition may be a condition using machine learning for solving a classification problem. For example, two or more sets of information on a text and a set of important terms corresponding to the text are prepared in advance, and models such as Support Vector Machine (SVM), Random Forest, and neural network are learned using the two or more information. An important term corresponding to the document information may be obtained using the learned model. In addition, about SVM, Random Forest, and a neural network, since it is an existing technique, description is abbreviate | omitted here.

  The important term acquisition means 121 is the Sakai method (Hiroyuki Sakai, Hiroshi Nonaka, Shigeru Masuyama, extraction of technical problem information from patent specifications, Journal of the Japanese Society for Artificial Intelligence, Vol. 25 No. 6 I, pp.531 -540, 2009.) etc. In addition, for the important terms extracted by the Sakai method, the Nonaka method (Hirofumi Nonaka, Akio Kobayashi, Hiroki Sakaji, Yusuke Suzuki, Hiroyuki Sakai, Shigeru Masuyama, Extraction of the effect and the technology terms from a patent document , Proc. Of 40th International Conference on Computers and Industrial Engineering (CIE40), cie214jp-1, Awaji-shima, Japan, (2010).) It is also good.

  The important term acquisition unit 121 does not need to acquire the important term when the important term corresponding to the axis does not exist in the document information. For example, for a certain document information, when an important term corresponding to the axis “use” does not exist in the document information, the important term acquisition unit 121 may not obtain an important term for the axis “use”. Document information that does not acquire important terms is usually not displayed on the technology map. Moreover, the important term acquisition means 121 may acquire two or more important terms.

  The term aggregating unit 122 aggregates two or more important terms into one term for each axis, and acquires item terms that are items of the axis of the technology map. Here, the term “aggregation” means that two or more important terms are replaced with terms having the same meaning, similar meaning, or superior meaning. The term aggregating unit 122 refers to the term dictionary 112 when aggregating two or more important terms into one term.

  For example, the term aggregating unit 122 first searches each important term acquired by the important term acquiring unit 121 from the term group of the term dictionary 112. When the important term is included in the term group of the term dictionary 112, the term aggregation unit 122 replaces the important term with a representative term corresponding to the term group including the important term in the term dictionary 112. Here, the term aggregation unit 122 may further retrieve the acquired item terms from the term group of the term dictionary 112 and aggregate them into terms having higher meanings. When the important term is not included in the term group of the term dictionary 112, the item acquisition unit 12 acquires the important term as it is as the item term.

  Note that the term aggregation unit 122 may use only some of the important terms when referring to the term dictionary 112. For example, the term aggregation unit 122 may aggregate important terms using only “resource reduction” of the important term “consumption resource reduction”.

  Also, the term aggregation means 122 is a method of Nonaka (Hirofumi Nonaka, Akio Kobayashi, Hiroki Sakaji, Yusuke Suzuki, Hiroyuki Sakai, Shigeru Masuyama, Extraction of the effect and the technology terms from a patent document, Proc. Of 40th International Conference on Computers and Industrial Engineering (CIE40), cie214jp-1, Awaji-shima, Japan, (2010).).

  The item acquisition unit 12 may use the important terms acquired by the important term acquisition unit 121 as the item terms as they are.

  The classification unit 13 associates each document information with an item term. Usually, the classification unit 13 associates the document information with the item terms corresponding to the important terms acquired by the important term acquisition unit 121. An item term corresponding to an important term is an item term that is an important term as it is, or an item term in which the term aggregating unit 122 aggregates the important terms.

  The classification unit 13 may associate the document information with items other than the item terms corresponding to the important terms.

  For example, with respect to certain document information (hereinafter, document information 1), the important term acquisition unit 121 acquires the important term “consumption resource reduction”, and with respect to other document information (hereinafter, document information 2), the important term acquisition unit 121. Has acquired the important term “energy consumption reduction”. Further, it is assumed that “consumption resource reduction” and “consumption energy reduction” are integrated into the term “eco-technology” by the term aggregation means 122. Further, it is assumed that the important term acquisition unit 121 and the term aggregation unit 122 acquire the item term “easy to work” based on another document information (hereinafter, document information 3). Further, it is assumed that the document information 1 includes a text “easy to work”.

  In the above case, the classification unit 13 determines whether each item term acquired by the item acquisition unit 12 and its related term are included in the document information for each document information. Associate document information with item terms. Here, the term related to the item term is a term in the term group in which the representative term matches the item term in the term dictionary 112.

  Further, the classification unit 13 associates the document information corresponding to one of the two or more original item terms of the changed item term acquired by the item aggregation unit 14 described later with the changed item term.

  The item aggregating unit 14 changes two or more item terms that satisfy a predetermined condition to one changed item term using one or more axes and the result in the classification unit 13. Here, the result in the classification unit 13 includes, for example, an item term and document information associated therewith. The predetermined condition is a condition for collecting item terms, for example, a condition such as “when the number of document information associated with item terms is equal to or less than a threshold”. Details of the predetermined conditions will be described later. Further, the post-change item term is a term symbolizing two or more item terms to be aggregated such as “others”. The item term after change may be one of two or more item terms to be aggregated, or a character string in which two or more item terms to be aggregated are connected by a delimiter (for example, “·”). Alternatively, terms other than two or more item terms to be aggregated such as “others” may be used, and terms input by a user using an item term input unit (not shown) may be used.

  Here, details of the predetermined condition will be described.

  For example, the item aggregation unit 14 may be a condition regarding the number of pieces of document information associated with item terms. Specifically, two or more item terms that are associated with only document information that is equal to or less than a predetermined threshold value may be changed item terms (hereinafter referred to as document information associated with the item terms). The number is called "number of documents (in item term)"). Here, the threshold value may be an absolute number of documents of item terms, or may be a relative number of documents represented by “number of documents of item terms / total number of document information”.

  Further, for example, the item aggregation unit 14 may be a condition regarding the number of items. Specifically, when the number of item terms acquired by the item acquisition unit 12 exceeds a predetermined threshold, two or more item terms with a small number of documents are selected so that the number of item terms is equal to or less than the threshold. It may be aggregated. Here, the threshold value may be an absolute number of item terms, or the number of pieces of document information stored in the document information storage unit 111 expressed by “total number of document information × constant (0 to 1)”. It may be a number that depends on. The latter constant means a kind of “compression rate”.

  Further, for example, the item aggregating unit 14 may be a condition including a condition regarding the number of document information associated with the item term and a condition regarding the number of items. For example, the predetermined condition is that the number of items is 10 or less, and that the item term corresponding to the item term is 3 or less is “other”.

  Further, for example, the item aggregating unit 14 may use information relating to past item terms prepared in advance. Here, information related to item terms (hereinafter, item term aggregation information) refers to “results of classification unit 13 (item terms and document information)” and “aggregated item terms and after-changes for a technology map generated in the past. It is information including the information of “item term”. The item aggregation unit 14 uses the item term aggregation information to determine whether or not the item terms that are the result of the classification unit 13 should be aggregated. For example, the probability that each item term is aggregated is derived from the item term aggregation information, and the item term is aggregated when the probability of aggregation exceeds a predetermined value. Here, the probability that item terms are aggregated can be calculated by, for example, “number of item terms changed to item terms after change in item term aggregation information / number of item terms appearing in item term aggregation information”. In addition, for example, using item term aggregation information, an item term to be aggregated or not and all item terms are input, and an item term after change in which the item terms are aggregated is output. Output) and machine terms can be aggregated using the learned model. At this time, models such as Support Vector Machine (SVM), Random Forest, and neural network can be used for machine learning. Since SVM, Random Forest, and neural network are existing technologies, description thereof is omitted. When using item term aggregation information, the technical map generation device 1 includes an item term aggregation information storage unit that stores item term aggregation information (not shown) in the storage unit 11.

  For example, the item aggregating unit 14 may specify the item terms to be aggregated by the user using an aggregated item term designating unit (not shown) and aggregate the designated item terms.

  Furthermore, the item aggregation unit 14 may combine items by combining the above-described conditions. For example, the item aggregating unit 14 proceeds to “Other” when “the number of document of the item term is 3 or less, or when the number of document of the item term is larger than 3 and the number of document of the item term is the top 10 or less”. Conditions such as changing may be used.

  The technical map configuration unit 15 configures a technical map that clearly shows the correspondence of each item of document information to each item term. Here, the item terms associated with each document information are the item terms obtained by excluding two or more item terms aggregated by the item aggregating unit 14 from the item terms for each axis acquired by the item acquiring unit 12, and the changed item terms It is.

  Here, composing a technology map means generating information necessary for outputting the technology map. For example, in the case of a biaxial technical map as shown in FIG. 15A, the technical map configuration unit 15 generates information for configuring a bubble. Specifically, in the above case, for example, information including the bubble radius corresponding to the number of documents is generated. Here, usually, the bubble radius increases in proportion to the number of documents. In the case of the technical map as shown in FIG. 15B, a list of documents described in each cell is generated. Further, for example, in the case of a uniaxial technical map such as a bar graph, the technical map construction unit 15 obtains the height of the bar based on the number of document information corresponding to each item term. The technology map construction unit 15 generates information suitable for the form of the technology map to be output.

  The technology map output unit 16 outputs the technology map configured by the technology map configuration unit 15.

  Here, output refers to display on a display, projection using a projector, printing on a printer, transmission to an external device, storage in a recording medium, processing result to another processing device or other program, etc. It is a concept that includes delivery.

  The storage unit 11, the document information storage unit 111, and the term dictionary 112 are preferably non-volatile recording media, but can also be realized by volatile recording media.

  The process in which information is stored in the storage unit 11, the document information storage unit 111, and the term dictionary 112 does not matter. For example, information may be stored in the storage unit 11, the document information storage unit 111, and the term dictionary 112 via a recording medium, and information transmitted via a communication line or the like is stored in the storage unit 11, document information. The information may be stored in the storage unit 111 and the term dictionary 112, or information input via the input device is stored in the storage unit 11, the document information storage unit 111, and the term dictionary 112. May be.

  The item acquisition unit 12, the classification unit 13, the item aggregation unit 14, and the technology map configuration unit 15 can be usually realized by an MPU, a memory, or the like. The processing procedures of the item acquisition unit 12, the classification unit 13, the item aggregation unit 14, and the technology map configuration unit 15 are usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).

  The technical map output unit 16 may or may not include an output device such as a display. The technology map output unit 16 can be implemented by output device driver software or output device driver software and an output device.

  Next, operation | movement of the technical map production | generation apparatus 1 is demonstrated using the flowchart of FIG. Hereinafter, the i-th information in the predetermined information is referred to as “information [i]”.

  (Step S201) The technical map generation device 1 substitutes 1 for a counter i.

  (Step S202) If the i-th axis to be output to the technical map exists, the technical map generation device 1 proceeds to step S203. Otherwise, the process proceeds to step S207.

  (Step S203) The item acquisition unit 12 acquires an item term that is an item of axis [i] of the technology map from each document information stored in the document information storage unit 111. Such processing is called item acquisition processing. Details of the item acquisition processing will be described with reference to the flowchart of FIG.

  (Step S204) The classification unit 13 associates each document information stored in the document information storage unit 111 with the item term of the axis [i] acquired in step S203. Such a process is called a classification process. Details of the classification process will be described with reference to the flowchart of FIG.

  (Step S205) The item aggregating unit 14 and the classification unit 13 aggregate the item terms of the axis [i] with respect to the result of step S204, and associate the document information with the aggregated items. Such processing is called item aggregation / reclassification processing. Details of the item aggregation / reclassification processing will be described with reference to the flowchart of FIG.

  (Step S206) The technical map generation device 1 increments the counter i by 1, and returns to step S202.

  (Step S207) The technology map configuration unit 15 configures information necessary for outputting the technology map based on the result of the item aggregation / reclassification processing of all axes to be output to the technology map. Such a process is called a technology map construction process. Details of the technology map construction processing will be described with reference to the flowchart of FIG.

  (Step S208) The technical map output unit 16 outputs a technical map based on the content information configured by the technical map configuration unit 15.

  Next, a specific example of details of the item acquisition process in step S203 will be described with reference to the flowchart of FIG.

  (Step S301) The important term acquisition unit 121 acquires important terms from each of the document information stored in the document information storage unit 111. Such a process is called an important term acquisition process. Details of the important term acquisition processing will be described with reference to the flowchart of FIG.

  (Step S302) The term aggregating unit 122 obtains a candidate for a term as an aggregation destination for each important term in order to aggregate the important terms of each document information acquired in Step S301. Such processing is called aggregate term candidate acquisition processing. Details of the aggregate term candidate acquisition process will be described with reference to the flowchart of FIG.

  (Step S303) The term aggregating unit 122 selects one term from the candidate terms to be the aggregation destination obtained in Step S303 for each important term. Such a process is called an aggregate term determination process. Details of the aggregate term determination process will be described with reference to the flowchart of FIG.

  Next, a specific example of the details of the important term acquisition process in step S301 will be described with reference to the flowchart of FIG.

  (Step S401) The important term acquisition unit 121 assigns 1 to the counter i.

  (Step S402) If the i-th document information (hereinafter, document information [i]) is present in the document information storage unit 111, the important term acquisition unit 121 proceeds to step S403. If not, the process returns to the upper process.

  (Step S403) The important term acquisition unit 121 acquires an important term from the document information [i]. As described above, the important term acquisition method may be a pattern method, a TF-IDF method, or a machine learning method.

  (Step S404) The important term acquisition unit 121 increments the counter i by 1, and returns to step S402.

  Next, a specific example of details of the aggregate term candidate acquisition process in step S302 will be described with reference to the flowchart of FIG.

  (Step S501) The term aggregation unit 122 substitutes 1 for a counter i.

  (Step S502) If the document information [i] exists in the document information storage unit 111, the term aggregation unit 122 proceeds to Step S503. If not, the process returns to the upper process.

  (Step S503) The term aggregation unit 122 substitutes 1 for the counter j.

  (Step S504) If the j-th important term exists in the document information [i] for the axis currently being processed (the axis specified by the counter i in the flowchart of FIG. 2), the term aggregation unit 122 proceeds to Step S505. move on. Otherwise, the process proceeds to step S508. Hereinafter, the j-th important term of the document information [i] is expressed as [i] [j].

  (Step S505) If the important term [i] [j] is included in any term group stored in the term dictionary 112, the term aggregation unit 122 proceeds to step S506. Otherwise, the process proceeds to step S507.

  (Step S506) The term aggregation unit 122 associates all the important terms [i] [j] with all the term groups including the important terms [i] [j] among the term groups stored in the term dictionary 112.

  (Step S507) The term aggregation unit 122 increments the counter j by 1 and returns to Step S504.

  (Step S508) The term aggregation unit 122 increments the counter i by 1, and returns to Step S502.

  Next, a specific example of the details of the aggregate term determination process in step S303 will be described using the flowchart of FIG.

  In this flowchart, a term group associated with the most important terms is used as an item term.

  (Step S601) The term aggregation unit 122 substitutes 1 for a counter i.

  (Step S602) When the document information [i] exists in the document information storage unit 111, the term aggregation unit 122 proceeds to Step S603. If not, the process returns to the upper process.

  (Step S603) The term aggregation unit 122 substitutes 1 for the counter j.

  (Step S604) Similar to step S502, the term aggregating unit 122 proceeds to step S605 if there is an important term [i] [j] for the currently processed axis. Otherwise, the process proceeds to step S607.

  (Step S605) If the term group associated with the important term [i] [j] in step S506 is one or more, the term aggregation unit 122 proceeds to step S606. Otherwise, the process proceeds to step S607.

  (Step S606) The term aggregation unit 122 acquires, as item terms, terms corresponding to a term group having the largest number of associated important terms from among a term group associated with the important terms [i] [j]. .

  (Step S607) The term aggregation unit 122 acquires the important terms [i] [j] as item terms.

  (Step S608) The term aggregation unit 122 increments the counter j by 1 and returns to Step S604.

  (Step S609) The term aggregation unit 122 increments the counter i by 1, and returns to step S602.

  In this flowchart, when there are a plurality of term groups associated with a certain important term, step S606 indicates the number of the important terms corresponding to the term group as “1 / number of term groups corresponding to the important term. May be calculated.

  Next, a specific example of details of the classification process in step S204 will be described using the flowchart of FIG.

  (Step S701) The classification unit 13 substitutes 1 for a counter i.

  (Step S702) If the document information [i] exists in the document information storage unit 111, the classification unit 13 proceeds to step S703. If not, the process returns to the upper process.

  (Step S703) The classification unit 13 substitutes 1 for a counter j.

  (Step S704) Similar to step S502, the classification unit 13 proceeds to step S705 if there is an important term [i] [j] for the axis currently being processed. Otherwise, the process proceeds to step S707.

  (Step S705) The classification unit 13 selects an item term corresponding to the term group associated with the important term [i] [j] (step S606) or an item term that is the important term [i] [j] itself (step S607). ) And document information [i].

  (Step S706) The classification unit 13 increments the counter j by 1, and returns to step S704.

  (Step S707) The classification unit 13 increments the counter i by 1, and returns to step S702.

  Next, a specific example of details of the item aggregation / reclassification process in step S205 will be described with reference to the flowchart of FIG.

  (Step S801) The item aggregation unit 14 substitutes 1 for a counter i.

  (Step S802) The item aggregating unit 14 determines that the i-th item term (hereinafter, item term [i]) exists for the axis currently being processed (the axis specified by the counter i in the flowchart of FIG. 2). The process proceeds to step S803. If not, the process returns to the upper process.

  (Step S803) The item aggregating unit 14 proceeds to step S804 when the item term [i] satisfies a predetermined condition. Otherwise, the process proceeds to step S806.

  (Step S804) The classification unit 13 associates the document information (step S705) associated with the item term [i] with the changed item term.

  (Step S805) The item aggregation unit 14 deletes the item term [i].

  (Step S806) The item aggregation unit 14 increments the counter i by 1, and returns to step S802.

  Next, a specific example of details of the technology map configuration processing in step S207 will be described using the flowchart of FIG.

  (Step S901) The technical map construction unit 13 substitutes 1 for a counter i.

  (Step S902) If the i-th item (hereinafter, item [i]) on the first axis exists, the technology map construction unit 13 proceeds to step S903. If not, the process returns to the upper process.

  (Step S903) The technical map construction unit 13 substitutes 1 for a counter j.

  (Step S904) If the j-th item (hereinafter, item [j]) of the second axis exists, the technology map construction unit 13 proceeds to step S905, and otherwise proceeds to step S907.

  (Step S905) The technical map construction unit 13 acquires all the documents associated with both the item [i] and the item “j” and substitutes them for D.

  (Step S906) The technical map construction unit 13 calculates the radius of the bubble based on the size of D. Usually, the radius is determined so that the area of the bubble is proportional to the size of D. Specifically, when the radius when the size of D is 1 is I, the radius when the size of D is n is I × (square root of n).

  (Step S907) The technical map construction unit 13 increments i by 1 and returns to step S902.

  The flowchart in FIG. 9 is for outputting a two-dimensional technical map, but the same applies to outputting a three-dimensional or higher technical map. For example, when outputting a three-dimensional technical map, the first axis item [i], the second axis item [j], and the third axis item [k] (i, j, k are respectively 1 to all the combinations (all combinations up to the last item of each axis) are acquired and substituted for D, and the radius of the bubble is calculated based on the D.

  Hereinafter, a specific operation of the technical map generation device 1 in the present embodiment will be described.

  In this specific example, the document information storage unit 111 stores the document information management table shown in FIG. The document information management table holds records of information regarding two or more documents. Each record has attributes of “ID”, “document identifier”, and “document information”. Here, the document identifier is an identifier for identifying each piece of document information.

  In this specific example, the term dictionary 112 stores the term group management table shown in FIG. The term group management table holds records relating to one or more term groups. Each record has attributes of “ID”, “axis”, “representative term”, and “term group”. Here, the representative term is a term representing the meaning of a term group, and is a term output to the technical map.

  In this specific example, based on each document information of FIG. 10, a bubble chart type technical map as shown in FIG. 15A having “issue” and “solution” as axes is output.

  In this specific example, the extraction of important terms is performed by extracting terms that appear one or more times in the term group included in the term group management table of FIG.

  In this specific example, the item term after change is fixed as “others”, and “item term associated with only one or less document information” is used as a condition for aggregation into the item term after change.

  First, the technical map generation device acquires important item terms (step S301) for the “task” axis. Specifically, the important term acquisition unit 121 assigns an important term including any one of the term groups (ID = 1 to 4) of the subject axis of the term dictionary 112 for the documents 1 to 7 in the document information storage unit 111. Obtained as a term (step S403). As a result, “Reference 1 → Power consumption”, “Reference 2 → Easy operation”, Reference 3 → Production at low cost, “Reference 4 → Power efficiency”, “Reference 5 → Efficient operation”, “Reference 6”. “Standby power” and “Document 7 → Waste power” are acquired.

  Next, the term aggregation unit 122 aggregates the acquired important terms (steps S302 and S303). In the aggregation of important terms, first, the aggregation destination candidates are searched, and the term aggregation unit 122 associates each acquired important term with a term group (step S302). For each important term, the term dictionary 112 is scanned, and the results “power consumption → power consumption”, “easy operation → workability”, “manufacturing at low cost → manufacturing cost”, “power efficiency → efficiency”, “efficiency” Is associated with “operational efficiency → optimization, optimization”, “standby power → power consumption”, and “waste power → power consumption” (respectively “important terms → representative terms”).

  Next, the term aggregation unit 122 aggregates important terms and acquires item terms. In this case, since there is one representative term corresponding to the important term other than the important term “efficient operation”, each important term is collected as it is. Moreover, there are two candidates for the efficient term “efficient operation”, “efficiency” and “optimization”, but the only important term associated with “optimization” is “efficient operation”. On the other hand, since there are two important terms associated with “efficiency”, “power efficiency” and “efficient operation”, they are aggregated into “efficiency” with the most associated important terms ( Step S606).

  As a result of the above processing, four item terms “power consumption”, “workability”, “manufacturing cost”, and “efficiency” are acquired for the “task” axis.

  Next, the classification unit 13 associates the document information with each item term on the “task” axis. The classification unit 13 directly associates the document information from which the important term that is the source of the item term is acquired (step S705). As a result, “power consumption → reference 1, reference 6, reference 7”, “workability → reference 2”, “manufacturing cost → reference 3”, and “efficiency → reference 4, reference 5” are associated (“ Item term → document identifier ”).

  Next, the item aggregating unit 14 further aggregates the aggregated item terms into the changed item terms (step S205). In this specific example, the item aggregating unit 14 determines whether or not “only one or less document information is associated” for each item term (step S803), and if the condition is satisfied, corresponds to the item term The attached document information is associated with the changed item term “others” (step S804). Furthermore, the item aggregation unit 14 deletes the item terms that satisfy the condition (step S805). As a result, the item terms and the document information associated therewith are “power consumption → reference 1, reference 6”, “efficiency → reference 4, reference 5”, “other → reference 2, reference 3”.

  Next, the technology map generation device 1 performs the same processing as the “issue” axis for the “solution means” axis (steps S203 to S205). As a result, in the “solution means” axis, the post-change item terms are not aggregated, but “configuration → reference 1, reference 2”, “material → reference 3, reference 4”, “method → reference 5, reference 6, reference”. 7 ”(respectively“ item term → document identifier ”).

  Next, the technology map configuration unit 15 configures information for outputting a technology map using the results of the “issue” axis and the “solution means” axis (step S207). In the case of this specific example, the technology map configuration unit calculates the radius of the technology map bubble from the item terms on the “issue” axis and the “solution means” axis and the number of document information corresponding thereto. For example, in the case of a bubble located at the intersection of “power consumption” and “configuration”, the document information group {reference 1, reference 6, reference 7} associated with “power consumption” and “configuration” The radius of the bubble is calculated by obtaining the product set {Document 1} of the document information group {Document 1, Document 2}. In the same way, the product set is obtained for all combinations of the problem and the solution and the bubble radius is calculated.

  Next, the technology map output unit 16 outputs a technology map based on the information configured by the technology map configuration unit 15. As a result, the technical map of FIG. 12A is output.

  In addition, based on the same document information management form and term group management table as the above specific example, the changed item term is “character string that concatenates the aggregated item terms with squares (•)”, the changed item terms When the condition of aggregation into “item terms associated with only 1 or less document information” is set, the technical map of FIG. 12B is output.

  Furthermore, based on the same document information management slip and term group management table as in the above specific example, the changed item term is fixed as “Other”, and the condition for aggregation into the changed item term is set to “2 or less”. In the case of “item term associated only with document information”, the technical map of FIG. 12C is output. As shown in FIG. 12 (c), even if the condition is that the “item terms associated with only two or less document information” are aggregated, “other” is associated with only two or less document information. (Problem “Power consumption” × Solution means “Other” and Problem “Other” × Solution means “Method”).

  As described above, according to the present embodiment, it is possible to generate a technical map in which items are aggregated using the classification result.

  Further, according to the present embodiment, items with a small number of document information (number of documents) can be collected into “others” and the like, and the listability of the technical map can be improved.

  Moreover, according to this Embodiment, the number of items which comprise an axis | shaft can be reduced to such an extent that the list property of a technical map can be improved.

  According to the present embodiment, when the axes are hierarchized, the item aggregating unit 14 generates items such as “others” in the hierarchy by applying item aggregation processing in order from the deeper hierarchy. You may do it. For example, when the item terms of main items A, B, C, and D exist and the item B has sub items b1, b2, and b3, the item aggregation unit 14 first sets the sub items b1, b2, and b3. It aggregates into the item term after change, and after that, it aggregates into the item term after change for A, B, C, and D which are main items. At this time, different aggregation conditions may be set for aggregation of sub-items and main items.

  Furthermore, the processing in the present embodiment may be realized by software. Then, this software may be distributed by software download or the like. Further, this software may be recorded and distributed on a recording medium such as a CD-ROM.

The software that realizes the technical map generation apparatus according to the present embodiment is the following program. In other words, the recording medium accessible by the computer is
A document information storage unit for storing two or more pieces of document information corresponding to a document identifier for identifying a document; and a computer that analyzes each of the two or more pieces of document information and includes one or more axes constituting a technical map. An item acquisition unit that acquires an item term that can be an item for each axis, a classification unit that determines a corresponding item term for each axis for each of the two or more pieces of document information, and one or more axes An item aggregating unit that changes two or more item terms that satisfy a predetermined condition to one changed item term using a result in the classification unit, and the classification unit includes the item aggregating unit The document information corresponding to any of the two or more original item terms of the changed item term acquired by is associated with the changed item term and is aggregated from the item terms for each axis acquired by the item acquisition unit Two or more terminology The technical map configuration is a technical map centered on a set of item terms having the item terms and the changed item terms, and constitutes a technical map that clearly shows the correspondence of the two or more pieces of document information to the item terms And a program for functioning as a technical map output unit for outputting the technical map.

  FIG. 13 is a schematic diagram showing an example of the external appearance of a computer that executes the program and realizes the technical map generation apparatus 1 according to the embodiment. The above-described embodiment can be realized by computer hardware and a computer program executed on the computer hardware.

  In FIG. 13, the computer system 3000 includes a computer 3001 including a CD-ROM drive 3005, a keyboard 3002, a mouse 3003, and a monitor 3004.

  FIG. 14 is a diagram showing an internal configuration of the computer system 3000. In FIG. 14, in addition to the CD-ROM drive 3005, a computer 3001 is connected to an MPU (Micro Processing Unit) 3011, a ROM 3012 for storing a program such as a bootup program, and an MPU 3011, and receives instructions of an application program. A RAM 3013 that temporarily stores and provides a temporary storage space, a hard disk 3014 that stores application programs, system programs, and data, and a bus 3015 that interconnects the MPU 3011, the ROM 3012, and the like are provided. The computer 3001 may include a network card (not shown) that provides connection to a LAN, WAN, or the like.

  A program that causes the computer system 3000 to execute the functions of the technical map generation device 1 according to the above-described embodiment may be stored in the CD-ROM 3021, inserted into the CD-ROM drive 3005, and transferred to the hard disk 3014. Instead, the program may be transmitted to the computer 3001 via a network (not shown) and stored in the hard disk 3014. The program is loaded into the RAM 3013 when executed. The program may be loaded directly from the CD-ROM 3021 or the network. Further, the program may be read into the computer system 3000 via another recording medium (for example, a USB memory) instead of the CD-ROM 3021.

  The program does not necessarily include an operating system (OS) or a third-party program that causes the computer 3001 to execute the functions of the technical map generation device 1 according to the above-described embodiment. The program may include only a part of an instruction that calls an appropriate function or module in a controlled manner and obtains a desired result. How the computer system 3000 operates is well known and will not be described in detail.

  In the above program, in a transmission step for transmitting information, a reception step for receiving information, etc., processing performed by hardware, for example, processing performed by a modem or an interface card in the transmission step (only performed by hardware). Not included) is not included.

  Further, the computer that executes the program may be singular or plural. That is, centralized processing may be performed, or distributed processing may be performed.

  In each of the above embodiments, each process (each function) may be realized by centralized processing by a single device (system), or by distributed processing by a plurality of devices. May be.

  The present invention is not limited to the above-described embodiments, and various modifications are possible, and it goes without saying that these are also included in the scope of the present invention.

  As described above, the technical map generation device according to the present invention has an effect of improving the listability of the entire technical map, and is useful as a patent map generation device or the like.

DESCRIPTION OF SYMBOLS 1 Technical map production | generation apparatus 11 Storage part 111 Document information storage part 112 Term dictionary 12 Item acquisition part 121 Important term acquisition means 122 Term aggregation means 13 Classification part 14 Item aggregation part 15 Technical map structure part 16 Technical map output part

Claims (8)

A document information storage unit for storing two or more pieces of document information corresponding to a document identifier for identifying a document;
An item acquisition unit that analyzes each of the two or more pieces of document information and acquires, for each axis, an item term that is a term that can be an item of one or more axes constituting the technical map;
For each of the two or more pieces of document information, a classification unit that determines a corresponding item term for each axis;
An item aggregating unit that changes two or more item terms satisfying a predetermined condition into one changed item term using one or more axes and the result in the classification unit;
The classification unit includes:
Associating document information corresponding to one of the two or more original item terms of the changed item term acquired by the item aggregation unit with the changed item term;
From the item terms for each axis acquired by the item acquisition unit, is a technical map around the item term set having the item terms excluding two or more aggregated item terms and the changed item terms, A technical map component that constitutes a technical map that clearly indicates the correspondence of each item term of two or more pieces of document information;
The technical map production | generation apparatus further provided with the technical map output part which outputs the said technical map. A term dictionary that stores term groups of two or more groups;
The item acquisition unit
An important term acquisition unit that analyzes each of the two or more pieces of document information and acquires, from each document information, an important term that satisfies a predetermined condition for each axis;
2. A term aggregating unit that aggregates two or more important terms acquired by the important term acquisition unit for each axis with reference to the term dictionary into terms representing groups and acquires item terms. The described technical map generator. The item aggregation unit includes:
The item term having a smaller number of corresponding document information is changed to one changed item term as the predetermined condition is satisfied by using the result in the classification unit for one or more axes. Item 3. The technical map generation device according to Item 2. The item acquisition unit
Analyzing each of the two or more pieces of document information, obtaining an item term that is a term that can be an item of each axis of the two axes constituting the technical map, for each axis,
The classification unit includes:
For each of the two or more pieces of document information, determine an item term belonging to each axis,
The technical map component is
The technical map production | generation apparatus as described in any one of Claims 1-3 which comprises a two-dimensional technical map. The item aggregation unit includes:
The technical map generation device according to claim 4, wherein two or more item terms that satisfy a predetermined condition are changed into one changed item term for each of the two axes using the result in the classification unit. The technical map generation device according to any one of claims 1 to 5, wherein the changed item term is a term meaning "others". On the recording medium,
Storing two or more pieces of document information corresponding to a document identifier for identifying a document;
A technology map output method that is processed using an item acquisition unit, a classification unit, an item aggregation unit, a technology map configuration unit, and a technology map output unit,
The item acquisition unit
Analyzing each of the two or more pieces of document information and obtaining, for each axis, an item term that is a term that can be an item of one or more axes constituting the technical map;
The classification unit is
Determining corresponding item terms for each axis for each of the two or more pieces of document information;
The item aggregation unit
For one or more axes, using the result in the classification unit, changing two or more item terms that satisfy a predetermined condition to one changed item term;
Furthermore, the classification unit
Associating document information corresponding to any of the two or more original item terms of the changed item terms acquired by the item aggregation unit with the changed item terms;
The technical map component is
From the item terms for each axis acquired by the item acquisition unit, is a technical map around the item term set having the item terms excluding two or more aggregated item terms and the changed item terms, Constructing a technical map that clearly indicates the correspondence of each item term of two or more pieces of document information;
The technical map output unit
A technical map generating method comprising: outputting the technical map. Computer-accessible recording media
A document information storage unit for storing two or more pieces of document information corresponding to a document identifier for identifying a document;
Computer
An item acquisition unit that analyzes each of the two or more pieces of document information and acquires, for each axis, an item term that is a term that can be an item of one or more axes constituting the technical map;
For each of the two or more pieces of document information, a classification unit that determines a corresponding item term for each axis;
An item aggregating unit that changes two or more item terms satisfying a predetermined condition into one changed item term using one or more axes and the result in the classification unit;
The classification unit includes:
Associating document information corresponding to one of the two or more original item terms of the changed item term acquired by the item aggregation unit with the changed item term;
From the item terms for each axis acquired by the item acquisition unit, is a technical map around the item term set having the item terms excluding two or more aggregated item terms and the changed item terms, A technical map component that constitutes a technical map that clearly indicates the correspondence of each item term of two or more pieces of document information;
A program for functioning as a technical map output unit for outputting the technical map.

Images