JPWO2008075744A1 - Information processing apparatus, method for generating information for selecting partner, and program - Google Patents

Abstract

In the information processing apparatus, using the cluster analysis unit 110 that classifies patent data into classes for each predetermined technical field and attribute information of patent data belonging to each class, the evaluation value for each class is calculated, and For each holder of patent data belonging to the class and for each holder of the patent data, an evaluation value for each class is calculated, an overall distribution tendency indicating a distribution tendency of the evaluation value for each class is obtained, and for each holder, the holder Using the evaluation value calculation unit 120 that creates the holder evaluation information in which the evaluation value for each class of each holder is associated, the obtained overall distribution tendency, and each created holder evaluation information, the name of the analysis target A technical complement degree calculation unit 125 that generates information for selecting a partner of a person (target holder). This makes it possible to efficiently select an optimal alliance destination using technical information such as a patent publication.

Description

  The present invention relates to an information processing technique for analyzing a plurality of technical documents, and more particularly to a technique for selecting an optimal alliance destination between companies using the technical documents.

Patent Document 1 discloses a merger / acquisition support system for searching for an M & A partner. Specifically, the merger / acquisition support system disclosed in Patent Document 1 includes information indicating the contents of the seller company (information on the company size, type of industry, financial status, products handled, etc.) and the conditions of the buyer company that the buyer company wishes to acquire (company). A database for storing information such as scale, type of business, financial status, and handling products). When the above merger / acquisition support system accepts the conditions of a purchase-requested company from a buyer company, the database is searched according to the condition, and information on a seller company that meets the above conditions is output to the buyer company.
JP 2004-318379 A

By the way, companies such as manufacturers sometimes form alliances with other companies in order to strengthen their technical capabilities and competitiveness. In alliances aimed at strengthening technical capabilities, it is hoped that companies that can make use of their strengths and make up for their weaknesses will be selected as partners, and companies that can further strengthen their strengths. It is rare.
However, at present, an effective means for appropriately evaluating and comparing the technologies of the company and other companies has not been established. Also, at present, there is no known effective criterion for selecting an alliance destination. For this reason, it is difficult to efficiently and accurately select partners from many companies that can make use of their strengths and make up for their weaknesses, and that can further strengthen their strengths. It has become.

  The technique of Patent Document 1 is a database search technique for searching for a seller company registered in a database using keywords such as a company scale of a purchase desired company and a product to be handled, and technical characteristics of each company. In consideration of (strengths and weaknesses), the optimal alliance destination is not selected.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to make it possible to efficiently select an optimal alliance destination using technical information such as a patent gazette.

(1) In order to solve the above problem, the first aspect of the present invention is applied to an information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data. Here, the patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Then, the information processing apparatus uses the means for classifying each acquired patent data into classes for each predetermined technical field and the attribute information of the patent data belonging to each class, and assigns an evaluation value for each class to each class. Means for calculating an evaluation value for each class and for each holder of patent data belonging to the class, and calculating an overall distribution tendency indicating a distribution tendency of the evaluation value for each class, Means for creating a name holder evaluation information in which evaluation values for each class name of the name holder are associated with each other, means for accepting designation of a target name holder who is a name of an analysis object, and the total obtained A partner selection information generating means for generating information for selecting the partner of the received target holder using the distribution tendency and each created holder evaluation information, and selecting the partner information The generating means includes, in each candidate name, composite evaluation information obtained by adding, for each class, nominee evaluation information of a candidate nominee who is each nominee other than the subject nominee, and nominee evaluation information of the subject nominee. Obtained for each person, for each of the obtained composite evaluation information, obtain a composite distribution tendency indicating the distribution tendency of the value for each class, and indicate the degree of approximation of the obtained composite distribution tendency with respect to the overall distribution tendency Ask for.

  As described above, in the first aspect of the present invention, the distribution tendency of the value for each class of the combined evaluation information obtained by adding the holder evaluation information of the target holder and the holder evaluation information of the candidate holder is acquired. The degree of approximation (synthetic approximation) with the distribution tendency of the evaluation value for each class of the entire patent data is obtained. The synthetic approximation indicates how similar the distribution tendency of the value for each class of the synthetic evaluation information is with the distribution tendency of the evaluation value for each class of the acquired patent data as a whole. The distribution tendency of the evaluation value for each class of the acquired patent data indicates the social technology distribution of the entire technical area related to the acquired patent data. It is desirable for each company to have a technology portfolio that approximates the above social technology allocation. Therefore, by referring to each synthetic approximation, it is possible to determine which candidate holders can make a technical tie-up to have a technology portfolio that approximates the above social technology distribution. That is, according to the first aspect, it is possible to efficiently select an optimal alliance destination.

  (2) In addition, the partner selection information generation means obtains a right holder distribution tendency indicating a distribution tendency of values for each class of the right holder evaluation information of the target right holder, and the right holder distribution tendency obtained as described above A target holder approximation degree indicating an approximation degree with respect to the overall distribution tendency is obtained, a higher one than the target holder approximation degree is selected from the obtained synthetic approximation degrees, and composite evaluation information having the selected synthetic approximation degree is specified. The candidate holder corresponding to the holder evaluation information used for the calculation of the specified composite evaluation information may be selected as the recommended partner.

  With this configuration, it is possible to select, from among a plurality of candidate holders, only candidate holders who are close to the distribution of the entire patent data as a result of partnering with the target holder. As a result, alliance partners with high technical supplementary effects can be selected as recommended partners.

(3) Further, the second aspect of the present invention for solving the above problem is applied to an information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data. The Here, the patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Then, the information processing apparatus uses the means for classifying each acquired patent data into classes for each predetermined technical field and the attribute information of the patent data belonging to each class, and assigns an evaluation value for each class to each class. Means for calculating an evaluation value for each class and for each holder of patent data belonging to the class, and generating a complementary reference vector having the evaluation value for each class as an element, For each person, means for generating a nominee evaluation vector whose elements are evaluation values for each class of the nominee of the nominee, means for accepting designation of the target nominee who is the nominee to be analyzed, and the generated complement A partner selection information generating means for generating information for selecting a partner of the received target holder using a reference vector and each of the generated right holder evaluation vectors; The partner selection information generating means is configured to add a composite evaluation vector obtained by adding a nominee evaluation vector of a candidate nominee who is each nominee other than the subject nominee and a nominee evaluation vector of the nominee of the subject nominee. Obtained for each person, a composite cosine that is a cosine of the composite evaluation vector and the complementary reference vector is calculated for each composite evaluation vector.

  As described above, in the second aspect of the present invention, a complementary reference vector having the evaluation value for each class of the entire patent data as elements is obtained, and each of the candidate holder's holder evaluation vector and the target holder's holder evaluation. A composite evaluation vector obtained by adding the vectors is obtained. For each composite evaluation vector, a cosine (combined cosine) between the composite evaluation vector and the complementary reference vector is calculated. With this composite cosine, the spatial distance between the composite evaluation vector and the complementary reference vector can be quantitatively grasped. Therefore, the analyst can grasp the tie-up effect for each candidate holder numerically by this synthetic cosine.

  (4) In addition, the partner selection information generation means calculates a target holder cosine that is a cosine of the target holder evaluation vector of the target holder and the complementary reference vector, and selects the target from the composite cosine. Select a higher one than the nominal cosine, identify the composite evaluation vector used to calculate the selected composite cosine, and recommend a candidate nominee corresponding to the nominal evaluation vector used to calculate the specified composite evaluation vector It may be selected as the destination.

  With this configuration, it is possible to select only candidate nominee corresponding to the composite evaluation vector closer to the complementary reference vector as a partner compared to the nominee evaluation vector of the target nominee. As a result, alliance partners with high technical supplementary effects can be selected as recommended partners.

  (5) Moreover, the said evaluation value for every class is a value which corresponds to the value which totaled the evaluation value for every class of the said holder within the applicable class, The said evaluation value for every class of the holder is the said evaluation for every class It is good also as the value normalized using the value.

  With such a configuration, it is possible to select a recommended partner based on a relative comparison between technical fields that are difficult to directly compare due to different absolute value standards.

(6) The third aspect of the present invention for solving the above problem is applied to an information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data. Here, the patent data includes information indicating the owner of the patent data and attribute information of the patent data.
The information processing apparatus uses the means for classifying the acquired patent data into classes for each predetermined technical field and the attribute information of the patent data belonging to each class, and the patent belonging to each class and the class. A means for calculating an evaluation value for each data holder, a means for receiving designation of a target holder who is an analysis target holder, and for each class of the received target holder using the calculated evaluation value A recommended alliance partner selecting means for selecting the alliance partner, and the recommended alliance partner selecting means is a candidate nominee who is each nominee other than the subject nominee for at least one of the classified classes. A total value obtained by adding the evaluation value in the class and the evaluation value in the class of the target holder is obtained for each candidate holder, and the candidate holder obtained using the total value is recommended in the class. Alliance Selected as.

  As described above, according to the third aspect, for each classified technical field, for example, it is possible to select a partner that can improve the technical ability of a predetermined level or more. Therefore, the third aspect is effective when, for example, a technical field in which an alliance is desired can be assumed.

(7) A fourth aspect of the present invention for solving the above problem is applied to an information processing apparatus that acquires a plurality of patent data and performs various types of information processing using the acquired patent data. Here, the patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Then, the information processing apparatus uses the means for classifying each acquired patent data into classes for each predetermined technical field and the attribute information of the patent data belonging to each class, and assigns an evaluation value for each class to each class. And calculating an evaluation value for each class and for each holder of patent data belonging to the class, and using the calculated evaluation value for each class for each class using the evaluation value for each class. And a means for calculating an evaluation value for each standardized holder for each standardized value, and an evaluation value for each class of the standardized holder, for each holder, for each class of the standardized holder of the holder. Means for creating holder evaluation information in which evaluation values are associated, means for receiving designation of a target holder who is a holder of an analysis target, and the received target holder using each of the created holder evaluation information People's partners A partner selection information generation means for generating information for determining, the partner selection information generation means is the holder evaluation information of the target holder and each holder other than the target holder. The candidate evaluation information obtained by adding the candidate evaluation information of each candidate holder for each class is obtained for each candidate holder, and the target holder's evaluation of the target holder among the values of each class of the combination evaluation information A fluctuating value is extracted from the value of the corresponding class of information, and the class to which the maximum value of the extracted value belongs is specified.

  As described above, according to the fourth aspect, when a technical tie-up is made for each holder of the alliance candidate, it is possible to specify a technical field that can most effectively obtain the technical tie-up. Therefore, for example, when nominees (for example, companies) who can form alliances are narrowed down due to competitive relationships with their own companies, etc., using the fourth aspect, technical cooperation is made for each narrowed company. Can understand the technical fields where In addition, by using the fourth aspect, for example, it is possible to obtain a local alliance effect after specifying a technical field where the effect of the technical alliance is large even with a company that cannot obtain a comprehensive alliance effect. it can.

(8) A fifth aspect of the present invention for solving the above problem is applied to an information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data. Here, the patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Then, the information processing apparatus uses the means for classifying each acquired patent data into classes for each predetermined technical field and the sex information of patent data belonging to each class, and obtains an evaluation value for each class for each class. And calculating an evaluation value for each class and for each holder of patent data belonging to the class, and using the calculated evaluation value for each class for each class using the evaluation value for each class. For each standardized name holder of the holder belonging to the class, using the means for calculating the standardized value for each class of the standardized holder and the evaluation value for each class of the standardized name holder Means for generating information associated with the evaluation value for each class, processing the generated information into information that can be visualized, and outputting the information.

  Thus, in the fifth aspect, the standardized evaluation value indicating the technical ability of each holder is visualized and output for each technical field. With this configuration, it is possible to make a relative comparison between technical fields that are difficult to directly compare due to different absolute value criteria. For example, it is possible to grasp the strengths and weaknesses of the company's own technology based on a relative comparison with other companies. Therefore, the analyst can select the alliance destination by grasping the balance of the technical capabilities of the company and other companies by viewing the output information.

  (9, 10) According to another aspect of the present invention, a partner selection information generation method having the same steps as the method executed by each of the above devices and the same processing as the processing executed by each of the above devices are performed on the computer. This is a partner selection information generation program that can be executed. This program may be recorded on a recording medium such as FD, CDROM, or DVD, or may be transmitted / received via a network.

(11) A sixth aspect of the present invention for solving the above problem is applied to an information processing apparatus that acquires a plurality of patent data and performs various types of information processing using the acquired patent data. Here, the patent data includes information indicating the owner of the patent data and attribute information of the patent data.
And the information processing apparatus
Means for classifying each acquired patent data into classes for each predetermined technical field;
Means for accepting designation of a subject holder who is a subject of analysis;
Using the attribute information of the patent data belonging to each class, the evaluation value for each class is calculated for each class, and among the patent data belonging to each class, the candidate nominee who is each nominee other than the subject nominee Means for calculating a composite evaluation value for each class and for each candidate holder, using the attribute information of the patent data and the attribute information of the patent data of the target holder,
A means for obtaining an overall distribution tendency indicating a distribution tendency of the evaluation value for each class, and for each candidate holder, obtaining a composite distribution tendency indicating a distribution tendency of the composite evaluation value for each class;
Using the obtained overall distribution tendency, and the obtained combined distribution tendency, the partner selection information generating means for generating information for selecting the partner of the received target holder,
The partner selection information generating means is
A composite approximation degree indicating the degree of approximation of each of the obtained composite distribution trends with respect to the overall distribution tendency is obtained.

  In this way, according to the sixth aspect, by referring to each synthetic approximation, it is judged which candidate holder can hold a technology portfolio that approximates the above-mentioned social technology distribution by making a technical tie-up. can do. That is, according to the sixth aspect, it is possible to efficiently select an optimal alliance destination.

(12) A seventh aspect of the present invention for solving the above problem is applied to an information processing apparatus that acquires a plurality of patent data and performs various types of information processing using the acquired patent data. Here, the patent data includes information indicating the owner of the patent data and attribute information of the patent data.
And the information processing apparatus
Means for classifying each acquired patent data into classes for each predetermined technical field;
Means for accepting designation of a subject holder who is a subject of analysis;
Using the attribute information of the patent data belonging to each class, the evaluation value for each class is calculated for each class, and among the patent data belonging to each class, the candidate nominee who is each nominee other than the subject nominee Means for calculating a composite evaluation value for each class and for each candidate holder, using the attribute information of the patent data and the attribute information of the patent data of the target holder,
Generating a complementary reference vector having the evaluation value for each class as an element, and generating, for each candidate holder, a composite evaluation vector having a composite evaluation value for each class of the candidate holder as an element;
Using the generated complementary reference vector and the generated composite evaluation vector, the partner selection information generating means for generating information for selecting a partner of the received target holder,
The partner selection information generating means is
For each composite evaluation vector, a composite cosine that is a cosine of the composite evaluation vector and the complementary reference vector is calculated.

  As described above, according to the seventh aspect, the spatial distance between the composite evaluation vector and the complementary reference vector can be quantitatively grasped by referring to each composite cosine. Therefore, the analyst can grasp the tie-up effect for each candidate holder numerically by this synthetic cosine.

(13) The eighth aspect of the present invention for solving the above problem is applied to an information processing apparatus that acquires a plurality of patent data and performs various types of information processing using the acquired patent data. Here, the patent data includes information indicating the owner of the patent data and attribute information of the patent data.
And the information processing apparatus
Means for classifying each acquired patent data into classes for each predetermined technical field;
Means for accepting designation of a subject holder who is a subject of analysis;
For at least one of the classified classes, among the patent data belonging to the class, the attribute information of the patent data of the candidate holder who is each holder other than the target holder and the attributes of the patent data of the target holder Means for using the information to calculate a composite evaluation value for each candidate holder;
And a recommended partner selection unit that selects a candidate holder obtained using the calculated composite evaluation value as a partner for each class of the received target holder.

  As described above, according to the eighth aspect, for each classified technical field, for example, it is possible to select a partner that can improve the technical capability of a predetermined level or more. Therefore, the eighth aspect is effective when, for example, a technical field in which an alliance is desired can be assumed.

(14) A ninth aspect of the present invention for solving the above problem is applied to an information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data. Here, the patent data includes information indicating the owner of the patent data and attribute information of the patent data.
And the information processing apparatus
Means for classifying each acquired patent data into classes for each predetermined technical field;
Means for accepting designation of a subject holder who is a subject of analysis;
Using the attribute information of the patent data belonging to each class, for each class, to calculate the evaluation value for each class and the evaluation value for each class of the subject holder, among the patent data belonging to each class, Using the attribute information of the patent data of the candidate holder who is each holder other than the target holder and the attribute information of the patent data of the target holder, the information is synthesized for each class and for each candidate holder. An evaluation value is calculated, and further, a standardized evaluation value for each class name and a standardized synthesis evaluation value obtained by standardizing the calculated evaluation value for each class name and the combined evaluation value using the evaluation value for each class. Means for calculating;
Using the evaluation value for each standardized holder for each class, creating the holder evaluation information for the target holder, and using the standardized composite evaluation value, for each candidate holder, the candidate holder's Means for creating synthesis evaluation information in association with standardized synthesis evaluation values;
Using the created holder evaluation information and composite evaluation information, the partner selection information generating means for generating information for selecting the partner of the received target holder,
The partner selection information generating means is
The value that fluctuates from the value of the corresponding class of the target person's holder evaluation information from the values of each class of the composite evaluation information, and the class to which the maximum value of the extracted values belongs is extracted. An information processing apparatus characterized by specifying.

  As described above, according to the ninth aspect, when a technical tie-up is made for each holder of the alliance candidate, it is possible to specify a technical field that can most effectively obtain the technical tie-up. Therefore, for example, in the case where a holder (for example, a company) who can form an alliance is narrowed down due to, for example, a competitive relationship with the company, by using the ninth aspect, a technical alliance is made for each narrowed company. Can understand the technical fields where In addition, by using the ninth aspect, for example, it is possible to obtain a local alliance effect after specifying a technical field where the effect of a technical alliance is large even with companies that cannot obtain a comprehensive alliance effect. it can.

(15) For each acquired patent data, means for acquiring a patent score that individually evaluates the value of the patent related to the patent data;
For each of the candidate holders, each patent of the candidate holder and the target holder using the patent score of the patent data in which at least one of the candidate holder and the target holder is the holder. It is good also as a synthetic | combination evaluation value calculation means which calculates the synthetic | combination evaluation value which shows technical evaluation at the time of combining.

  With this configuration, by using a patent score that individually evaluates the value of a patent related to each patent data, a composite evaluation that reflects the value of a patent in which at least one of the candidate holder and the target holder is the holder The value can be calculated. As a result, analysis accuracy can be improved.

(16) Further, the composite evaluation value calculation means includes:
For each class, it is possible to select a patent score that is equal to or higher than a predetermined threshold from among the patent scores of patent data belonging to the class, and calculate a value obtained by tabulating the selected patent scores as the composite evaluation value. Good.

  With this configuration, by dividing the values below the threshold with the values above the predetermined threshold, the number of patents with a small number of important patents is high even if the number of cases is large. Can be prevented. As a result, an appropriate composite evaluation value can be calculated, and the analysis accuracy can be improved.

  (17) The patent score may be a value standardized in a patent group of a population including a class for which the composite evaluation value is calculated.

  With this configuration, it is possible to improve the accuracy of relative comparison between different classes and between different candidate nominal persons by obtaining a standard value in the population and calculating a composite evaluation value. As a result, analysis accuracy can be improved.

  (18) Further, the patent score means that the patent data is classified into groups for each technical field and for each predetermined period, and for each classified group, the progress information of the patent data belonging to the group is used. It may be a calculated value for the patent data.

By classifying into groups for each technical field and every predetermined period, and calculating the patent score using the progress information for each classified group, to correct the bias of the progress information due to the difference in the technical field and the application time, An accurate patent score can be calculated. As a result, an appropriate composite evaluation value can be calculated, and the analysis accuracy can be improved.

1 is a functional block diagram of an information processing apparatus to which an embodiment of the present invention is applied. The hardware block diagram of the information processing apparatus 1 of the said embodiment. The figure which illustrated the data structure of the patent data memorize | stored in the memory | storage device in simulation. The flowchart which shows the partner selection information generation process of the information processing apparatus of the said embodiment. The conceptual diagram explaining the outline of a technical complementation level and a technical deepening level. The conceptual diagram which shows an example of the acquisition process of patent data. The conceptual diagram which shows an example of a cluster analysis result and the classification | category for every applicant. The bar graph which shows the evaluation value for every class for every holder. The bar graph which shows the evaluation value for every standardized holder every class. The flowchart explaining a detail about the calculation process of a technical complementation degree among FIG. The table | surface which illustrated the calculation result of the technical complementation degree calculation process by this embodiment. The flowchart explaining a detail about the calculation process of a technical deep-plowing degree among FIG. The table | surface which illustrated the calculation result of the technical plowing degree calculation process by this embodiment. The table | surface which illustrated the calculation result of the technical deep plowing degree and the technical complementation degree calculation process by this embodiment. An example of the alliance distribution map output by embodiment of this invention. The flowchart explaining the process which produces | generates the graph of FIG. 10 is a flowchart showing partner selection information generation processing of the information processing apparatus according to the modification 8; The distribution of the evaluation value for each class using the patent score of the modified example 8 and the evaluation value for each class using the patent impact index and the progress information index of the above embodiment are shown in relation to the number of publications. Figure. The figure which simulated an example of the data structure of the content information utilized in the modification 8. The figure which simulated an example of the data structure of the progress information utilized in the modification 8. 10 is a flowchart showing the procedure of a patent score calculation process according to Modification 8. 10 is a flowchart showing details of a process for calculating a patent score according to Modification 8;

Explanation of symbols

1: Information processing device, 2: Storage device, 3: Input device, 4: Output device, 100: Control unit, 105: Data acquisition unit, 110: Cluster analysis unit, 120: Evaluation value calculation unit, 125: Technical complement Calculation unit, 130: Technical deep-plow calculation unit, 135: Output unit

<1. Schematic configuration of information processing apparatus>
Hereinafter, embodiments of the present invention will be described with reference to the drawings. First, a schematic configuration of an information processing apparatus to which an embodiment of the present invention is applied will be described.

FIG. 1 is a functional block diagram of an information processing apparatus to which an embodiment of the present invention is applied.
As shown in the figure, the information processing apparatus 1 that generates information for selecting a partner is a patent data 200 of a right holder (target holder) to be analyzed and a right holder (candidate holder) other than the right holder. Are connected to the storage device 2 for storing the various information, the input device 3 for receiving various requests from the analyst, and the output device 4 for outputting the partner selection information generated by the information processing device 1.

  In the following, the information processing device 1 and the storage device 2 are exemplified by a case where they are connected by a network such as a LAN (Local Area Network). Further, the information processing device 1 and the input device 3 are locally connected, and the information processing device 1 and the output device 4 are locally connected.

  The information processing apparatus 1 includes a control unit 100, a data acquisition unit 105, a cluster analysis unit 110, an evaluation value calculation unit 120, a technical complement degree calculation unit 125, a technical deepening level calculation unit 130, and an output unit 135.

  The control unit 100 controls the overall operation of the information processing apparatus 1. In addition, the control unit 100 receives various requests from the analyst via the input device 3. For example, the control unit 100 receives information specifying the technical area of the patent data to be analyzed and information specifying the analysis target holder (target holder) input by the analyst.

  The data acquisition unit 105 accesses the storage device 2, stores various data in the storage device 2, and reads out data stored in the storage device 2. For example, the data acquisition unit 105 receives the technical area of the patent data to be analyzed requested by the user (analyzer) via the control unit 100. Then, the data acquisition unit 105 accesses the storage device 2 and reads patent data belonging to the accepted technical field from the patent data 200 stored in the storage device 2. In addition, the data acquisition unit 105 reads information indicating the owner of the read patent data and attribute information (content information, progress information, etc.). Details of the information to be read will be described later.

  The cluster analysis unit 110 classifies the plurality of patent data acquired by the data acquisition unit 105 into classes for each predetermined technical field. For example, the cluster analysis unit 110 generates a document vector based on the content information of each of the patent data, and performs cluster analysis based on the similarity between the document vectors, so that a plurality of patent data is obtained for each predetermined technical field. Class. When the classification by cluster analysis has a hierarchical structure, it is possible to arbitrarily select which of the lower cluster and the upper cluster is used according to the technical scope to consider the partner.

The evaluation value calculation unit 120 calculates the evaluation value for each class for each class using the attribute information of the patent data belonging to each class classified by the cluster analysis unit 110, and the right holder for each class and each right holder The evaluation value for each class is calculated. Here, it is particularly preferable to use progress information as the attribute information for calculating the evaluation value. Moreover, it is preferable to normalize the evaluation value for each class for each holder using the evaluation value for each class.
Moreover, the evaluation value calculation part 120 produces the holder evaluation information which matched the evaluation value for every class of each holder of the holder for each holder.

  The technical complement degree calculation unit 125 uses the overall distribution tendency indicating the distribution tendency of the evaluation value for each class and the holder evaluation information in which the evaluation value for each class for each holder is associated with each other as the partner selection information. Calculate the degree of technical complementation.

  The technical plow level calculation unit 130 calculates the technical deep plow level as partner selection information, using the right holder evaluation information in which the evaluation value for each standardized right holder for each class is associated with each right holder.

  The output unit 135 outputs the technical complementation level and the technical deepening level as the above-described partner selection information to the output device 4, or generates information indicating the recommended partner as the partner selection information to the output device 4. Output. For example, when the output device 4 is a display, the output unit 135 generates image data indicating partner selection information and outputs the generated image data to the output device 4. Further, when the output device 4 is a printer, the output unit 135 generates print data indicating partner selection information, and outputs the generated print data to the output device 4.

  The storage device 2 is a unit that stores patent data 200 such as a patent gazette, and includes, for example, a DVD-ROM drive, a DVD-ROM, an HDD, and the like.

FIG. 3 is a diagram schematically illustrating the data structure of patent data stored in the storage device.
In the storage device 2, patent data ID 210, bibliographic information 220, content information 230, and progress information 240 are stored in association with each other for a large number of patent data 200.
As the patent data ID 210, for example, a number of a published patent gazette, a number of a patent publication gazette, and the like are stored. Further, the present invention is not limited to this, and an application number, a unique reference number, or the like may be given.
The bibliographic information 220 includes nominee information indicating the applicant or patent owner. In addition, information such as the filing date and the number of claims may be included. In the present embodiment, the right holder information is particularly referred to in the process of classifying each class for each right holder in order to calculate the evaluation value for each right holder class.
The content information 230 preferably includes text data such as a specification, claims, summary, and image data such as drawings. In the present embodiment, in the process of classifying patent data into classes for each predetermined technical field, text data such as a specification and claims is referred to.
The progress information 240 includes accelerated examination request data, invalid trial maintenance decision data, cited count data, and the like. In the present embodiment, in particular, the progress information is referred to in calculating the evaluation value for each class for each holder.

The input device 3 is a unit that accepts instructions for selecting various functions of the information processing device 1 and specifying patent data to be analyzed and a holder of the analysis target. The input device 3 includes hardware switches such as scroll keys and scale change keys, and joystick Is done.
The output device 4 is a unit for receiving and outputting data from the information processing device 1. The output device 4 is composed of, for example, a liquid crystal display or a printer.

<2. Hardware configuration of information processing apparatus>
Next, a hardware configuration of the information processing apparatus 1 according to the present embodiment will be described.
FIG. 2 is a hardware configuration diagram of the information processing apparatus 1 according to the present embodiment.
As illustrated, the information processing apparatus 1 includes a CPU (Central Processing Unit) 10, a main storage device 11 such as a RAM (Random Access Memory), an auxiliary storage device 12 such as an HDD, and a peripheral device (here, an input). Controls data transmission / reception between the IO interface 13 that controls transmission / reception of data between the device 3 and the output device 4) and a device (here, the storage device 2) connected to the network. Network interface 14.

  Here, the auxiliary storage device 12 includes each unit (control unit 100, data acquisition unit 105, cluster analysis unit 110, evaluation value calculation unit 120, technology complementation degree calculation unit 125, and technical deepening degree calculation unit 130 illustrated in FIG. , And a program (partner selection information generation program) for realizing the function of the output unit 135) is stored.

And the function of each part (the control part 100, the data acquisition part 105, the cluster analysis part 110, the evaluation value calculation part 120, the technical complementation degree calculation part 125, the technical deepening degree calculation part 130, and the output part 135) shown in FIG. Is realized by loading the partner selection information generation program stored in the auxiliary storage device 12 into the main storage device 11 and executing it.

<3. Partner selection information generation process>
Subsequently, processing performed by the information processing apparatus of the present embodiment will be described.
FIG. 4 is a flowchart showing the partner selection information generation process of the information processing apparatus of this embodiment.
As shown in the figure, the information processing apparatus according to the present embodiment performs processing for calculating the degree of technical complementation between the company and each other company as the partner selection information, or recommends based on the calculated degree of technical complementation. A process of selecting a partner company is performed (S100 to S160). In addition, the information processing apparatus according to the present embodiment performs processing for calculating the technical depth of the company and each other company as the partner selection information, or based on the calculated technical depth, etc. A process of selecting a recommended partner company is performed (S200 to S260). Moreover, the information processing apparatus of this embodiment produces | generates the alliance distribution map based on a technical complementation degree and a technical deepening degree as cooperation partner selection information (S400).

<3-1. Outline of technical complementation and technical plowing>
An outline of the concept will be described with respect to the above-mentioned technical complementation level and technical deepening level.
FIG. 5 is a conceptual diagram for explaining the outline of the degree of technical complementation and the degree of technical deepening.
In a technical area to be analyzed, a plurality of patent data are classified into classes for each predetermined technical field, in this case, six classes. Each class is further classified into each company, in this case, company α, company β, company γ, and an evaluation value for each class for each holder is calculated based on patent data for each class for each holder. In FIG. 5A, the vertical axis represents each class, and the horizontal axis represents the evaluation value for each class for each holder. However, the values on the horizontal axis are standardized so that the sum of evaluation values for each class in each class (evaluation value for each class) is equal to the sum in other classes (evaluation value for each class). From Fig. 5 (A), you can find out about the strengths and weaknesses of each of α, β, and γ. From the standpoint of α, either β or γ is selected as the partner. It is possible to examine whether to do this.

  When selecting a partner, two different perspectives can be considered. One is to see the overall cooperation effect (technical complement) by mutual complementation in the entire technical area of the analysis target. The other one is to observe the local alliance effect (technical depth) in the specific technical field (class) in the technical area to be analyzed.

When considering the complementary relationship between two companies (for example, α company and β company), if α company has a lack of expertise, β company has a lack of expertise, and α company has a lack of expertise. Complementary relationships are established with each other. The degree of technical complementation quantifies this complementary relationship. Specifically, the evaluation value for each class of the two companies is added for each class to obtain the composite evaluation information, and the technical complement is calculated using this composite evaluation information.
FIG. 5B shows only the evaluation values for each class of the company α and company β in FIG. 5A. Since the evaluation values for each class of holders of company α and company β are displayed side by side, the combined evaluation information of the two companies can be read from the sum of the evaluation values for each class of holders of the two companies. ing. In FIG. 5 (B), the weak fields of company α and company β are mutually supplemented, and it is estimated that the degree of technical complementation is high.

FIG. 5C shows only the evaluation values for each class of the company α and company γ in FIG. 5A. Examining together with FIG. 5 (B), it can be seen that for the company α, the cooperation with the company β has a higher overall complementary effect than that with the company γ. However, when looking at FIG. 5B and FIG. 5C for each class, γ company has an evaluation value higher than β company for class 6. Therefore, if company α is affiliated with company γ, it will be possible to deepen class 6 (that is, it will be possible to enhance the technical capabilities of the technical fields classified as class 6). It has the potential to hope for further breakthroughs. The technical plowing degree is a numerical value of the plowing degree. This technical depth is also calculated from the value obtained by adding the evaluation values for each class of the two companies for each class.

<3-2. Technical complementation (S100-S160)>
Calculation of the degree of technical complementation and selection of a recommended partner company based on the degree of technical complementation will be described with reference to the flowchart of FIG.

<3-2-1. Acquisition of patent data (S100)>
The data acquisition unit 105 of the information processing apparatus acquires specific information of the patent data input by the analyst from the input device 3, and acquires the patent data 200 from the storage device 2 according to the specific information (S100).
As the specification information of patent data, any information can be used as long as it is information for specifying a plurality of patent data, such as an IPC code. However, in this embodiment, assuming that the analysis target holder (target holder) or the analyst who received the request selects his / her partner from a plurality of other companies, the data to be acquired in the following procedure Is specified information of patent data.

FIG. 6 is a conceptual diagram illustrating an example of patent data acquisition processing.
(A) First, the data acquisition unit 105 performs a process of selecting a technology of interest from the patent data of the target holder. Specifically, the data acquisition unit 105 receives the designation of the target holder and acquires the patent data of the target holder from the storage device 2. Further, the cluster analysis unit 110 performs cluster analysis on the acquired patent data of the target holder to obtain a plurality of clusters (target company clusters) (cluster analysis will be described later). Then, the data acquisition unit 105 selects a specific target company cluster from among a plurality of target company clusters as the technology of interest (FIG. 6A). As a method for selecting a target technology from the target company cluster, for example, the target technology may be selected according to the target company cluster selection condition separately input by an analyst, and each target company may be selected by the evaluation value calculation unit 120. An evaluation value of the cluster may be calculated (evaluation value will be described later), and based on this evaluation value, for example, a target company cluster having the largest evaluation value may be selected as the attention technology.
(B) Next, the data acquisition unit 105 changes the target technology and the target technology from all the patent data (self-other patent document group) including publications of target holders and holders other than the target holder (candidate holders). A similar document group (a patent group field document group) is extracted. Specifically, the similarity calculation module included in the cluster analysis unit 110 or the like is used to calculate the similarity between each of the other patent document groups and the noted technology, and a predetermined number of patent data having a higher degree of similarity are automatically acquired. It is extracted as another patent specific field document group (FIG. 6B). In this way, by extracting a group of documents specific to the patent of the subject patent, focusing on the subject's attention subject's attention technology, comprehensive data including patent data related to the subject owner's attention technology and patent data belonging to the same technical area of other companies You can select your own partner using various documents.

<3-2-2. Cluster Analysis (S110)>
After acquiring a plurality of patent data, the cluster analysis unit 110 classifies the patent data into classes for each predetermined technical field (S110).

  The classification method is arbitrary as long as it is a method for classifying patent data for each technical field, such as an IPC code (when the patent data is identified using the IPC code in S100, the lower-level IPC code here). Can be used. However, in the present embodiment, a document such as a specification and claims included in the patent data is expressed as a vector, and cluster analysis is performed.

  The vector representation of the document can be performed by a known method. For example, an index word is extracted by performing a word segmentation process from a specification, a claim, etc. included in the content information 230 of each patent data, and each index word is weighted based on its appearance frequency. As the weighting, for example, an index word frequency (TF: the number of appearances of the index word in the document) and a document frequency (DF: the number of documents in which the index word appears in a predetermined document group) or a document frequency TFIDF obtained by the product of the reciprocal of the logarithm (IDF: inverse document frequency) can be used. A multidimensional vector (the number of dimensions is the number of index words) having this weight as a vector element is a vector expressing each patent data.

  Cluster analysis can also be performed by a known method. For example, the similarity between vectors (inner product value, correlation coefficient, etc.) is calculated, and similar patent data are collected into a cluster. In the hierarchical cluster analysis, not only the lower cluster, but also the upper cluster in which the lower clusters having a high degree of similarity are combined, and the middle cluster between them are generated.

FIG. 7 is a conceptual diagram showing an example of cluster analysis results and classification for each applicant.
FIG. 7A shows an example in which 59 lower clusters, 8 middle clusters, and 4 upper clusters are generated. Here, the lower cluster is a cluster indicated by a solid rectangle in FIG. The middle clusters 711, 721 to 723, 731 and 741 to 743 are clusters shown in each column of FIG. The upper clusters 710, 720, 730, and 740 are clusters generated at the first branch in FIG. Whether to use a lower cluster as a “class”, which is an analysis unit in the present embodiment, or a middle or upper cluster, may be selected according to the purpose of the analysis. In the following, a middle cluster is used. An analysis example targeting individual classes is shown.

  In the present embodiment, the cluster analysis unit 110 further classifies each class for each holder. Specifically, the information on the applicant or the patent owner is extracted from the bibliographic information 220 of the patent data belonging to each class, and based on this, the classification is made for each holder (FIG. 7B).

<3-2-3. Calculation of Patent Impact Index and Progress Information Index (S120)>
After classifying patent data into classes for each predetermined technical field by cluster analysis and classifying each class for each holder, the evaluation value calculation unit 120 calculates the evaluation value for each class for each holder. “Patent Impact Index” and “Progress Information Index” are calculated (S120).

  First, the evaluation value calculation unit 120 determines the number N of patent data documents for each class for each holder. When a published patent gazette and a patent publication gazette are issued for a patent application, it is desirable to count the number of documents for the patent application as two.

Next, the data acquisition unit 105 reads the history information of each patent data for each class for each holder from the storage device 2. As an example of progress information to be read, for each patent application,
“Number of other company citations” (0 or positive integer),
“Number of opposition to patent or request for trial for invalidation of patent” (0 or positive integer),
“Existence request” (1 or 0 (“1” if present, “0” if not present)),
"Patent / non-patent registration" (1 or 0 ("1" if present, "0" if not present)),
“Whether or not there is a request for accelerated examination” (1 or 0 (“1” if yes, “0” if not)),
“Presence or absence of appellate appeal” (1 or 0 (“1” if yes, “0” if not))
However, other information may be used.

Next, the evaluation value calculation unit 120 calculates a plurality of indices based on the read progress information for each class for each holder.
Examples of this indicator include “total number of citations from other companies”, “total number of requests for opposition to patent or patent invalidation trial”, “examination request rate”, “registration assessment rate”, There are “registration rate”, “prompt request rate”, “other companies quoted ratio”, “appraisal appeal trials ratio”, “opposed objection or invalid trial request ratio”, but other indicators may be used . Each definition is as follows.
“Total number of citations from other companies” = “Total number of citations from other companies” in the document group for each class of “No. of citations from other companies”
= Total number of documents for each class of “No. of opposition to patent or request for invalidation of patent” “Request for examination” = Number of requests for examination / Number of patent applications “Rate of patent registration” = Number of patent registrations / Number of patent applications "Registered assessment rate" = Number of patent registrations / Number of requests for examination "Rapid request rate" = Number of requests for accelerated examination / Number of requests for examination "Ratio of citations of other companies" = "Total number of citations of other companies" / Number of patent applications “Rate of Appeals against Appraisal Appraisal” = Number of Appeals against Appraisal / Number of Requests for Examination “Rate of Appeals against Patents or Requests for Invalidation of Patent Invalid”
= "Total number of requests for opposition to patents or requests for invalidation of patented patents" / Number of patent registrations Of these definitions, the number of patent applications is the number of patent applications for each class per holder. The total number of patent registrations in each document class for each class of holder (1 or 0) The total number of patent registrations is the total in the document group for each class of holders (1 or 0). The number of requests is the total number of documents in each class for each holder of “whether or not there is a request for accelerated examination” (1 or 0). It is sufficient to give the total in the document group.

Next, the evaluation value calculation unit 120 calculates a patent impact index by assigning a predetermined weight to the “number of documents” for each class for each holder. The patent impact index is a measure for evaluating the other company's restraining power (the degree to which the rights of other companies are suppressed and the value of the company's patent is improved) for each class of each holder. For example, the “number of documents” is weighted based on the “total number of citations from other companies” and / or the “total number of requests for opposition to a patent or patent invalidation request”,
Patent Impact Index = “Number of documents” + “Total number of citations from other companies” + “Total number of requests for opposition to patent or request for trial for invalidation of patent”
Can be calculated. The weighting for the “number of documents” may be performed by addition as in the above equation or by multiplying by some other ratio.
In addition to these, the predetermined weighting includes, but is not limited to, various things such as patent profitability, patent productivity, patent utilization, and patent competitiveness.
By determining the number of documents as described above, the share of each holder in the class can be grasped. In addition, by giving a predetermined weight to the number of documents, each holder in the class can be evaluated in consideration of economic aspects and technical competitiveness.

Next, the evaluation value calculation unit 120 calculates a progress information index by averaging the indices based on the progress information for each class for each holder. The progress information index seeks to evaluate the value of a patent from the perspective of the company, the patent office and competitors for each class of holder,
Progress information index = √ {Σ _{index No = 1} ^d (index) ² / d}
Can be calculated. That is, d indicators based on the progress information, for example, “examination request rate”, “registered appraisal rate”, “patent registration rate”, “early examination request rate”, “other companies cited number ratio”, “number of appealed trial cases” The progress information index is calculated by taking the positive square root of the value calculated by dividing the sum of squares of a total of seven indicators, "Ratio" and "Ratio of objections to appeal or invalidation request" by the number of indicators d = 7. Can be calculated. Here, the above seven indicators are shown as examples of progress information indicators. However, for example, “in-house citation count ratio”, “domestic priority claim ratio”, “foreign priority claim ratio”, “packet” You may make it use "bag browsing rate" etc. By calculating the evaluation value for each class using each of the above indicators, it is possible to evaluate each name holder in that class taking into account the influence of patents that may be an obstacle to the acquisition of patents and technological development of other companies. It is. It is also possible to make an evaluation that takes into account the applicant's willingness to obtain rights and the evaluation of the examiner.

<3-2-4. Calculation of evaluation value for each class for each holder (S130)>
After calculating the patent impact index and the progress information index, the evaluation value calculation unit 120 calculates the evaluation value for each class for each holder (S130).
Specifically, for each class for each holder, the patent impact index and the progress information index are multiplied to calculate an evaluation value for each class for each holder. By indexing the progress information in this way, for example, quantitative and objective evaluation can be performed.
This evaluation value for each class for each holder has the following properties.
In the case of “examination request rate” = 0, the progress information index is 0 in most cases, and as a result, the evaluation value per class for each holder is also 0.
The progress information index increases as the number of patents registered increases. In addition, if there is a trial against decision of refusal, opposition to objection, etc., it will be taken into account.
Since the patent impact index counts the number of publications, it increases as the number of patent applications increases, and further increases when patent publications are issued. The weights are weighted by “total number of other company citations” and “total number of requests for opposition to patent or requests for invalidation of patent invalid”.
Since the patent document group can be evaluated from the aspect of the historical information by the evaluation value for each class for each holder, it is possible to know the strength of the patent that cannot be measured only by the number of patents.

FIG. 8 is a bar graph showing evaluation values for each class for each holder. The vertical axis represents the class (class 1 to class 8). The horizontal axis shows the evaluation value for each holder of each holder (Company A to J and others), and the length of the bar including all holders for each class is the evaluation value for each holder. Represents the total value of the class.
Thereby, for each technical field, it is possible to grasp the position of each holder (for example, each company) in each class, and to know where the strengths and weaknesses of a specific holder (for example, the company) are. In addition, when any other company is selected as the partner, it is possible to select the partner by reading how much the total evaluation value for each class of each company in the class and the partner is concerned. it can.
For example, if company B is a rival company of company A, it will formulate a strategy such as where to work together in order for company A to continue to maintain a competitive advantage over company B. You can also

図９は、規格化名義人毎クラス毎評価値を示す棒グラフである。縦軸がクラス（クラス１乃至クラス８）を表している。横軸は、各名義人（Ａ社乃至Ｊ社及びその他）の規格化名義人毎クラス毎評価値を表しており、各クラスについて、全名義人を含む棒の長さはすべて１となっている。
これにより、絶対的価値基準が異なるために直接対比することが困難な技術分野間の相対比較も可能となり、その規格化した値に基づいて推奨提携先の選定を行うことができる。すなわち、ある特定の名義人の各クラスでの位置づけを一層容易に把握し、自社の強み、弱みがどこにあるかを把握することができる。また、任意の他社を提携先として選んだ場合に、各クラスにおける自社及び当該提携先の名義人毎クラス毎評価値の合計がどの程度になるかを読み取ることにより、提携先を一層容易に選定することができる。<3-2-5. Standardization (S140)>
Once the evaluation value for each class for each holder is calculated, it is desirable to standardize so that it can be easily compared with the evaluation value for each class for each holder in another class (S140).
In addition, although this embodiment is not specifically limited about the specific procedure of the said normalization, For example, for every class, the sum total of the evaluation value for every class in the said class (evaluation value for every class) And the evaluation value for each class for each holder may be divided by the evaluation value for each class. That is, for example, the evaluation value for each class for each holder may be standardized by (Equation 1) shown below.

FIG. 9 is a bar graph showing the evaluation value for each class for each normalized name holder. The vertical axis represents the class (class 1 to class 8). The horizontal axis represents the evaluation value for each standardized holder for each holder (Company A to J and others), and the length of the bars including all holders is 1 for each class. Yes.
This makes it possible to make a relative comparison between technical fields that are difficult to compare directly because of different absolute value standards, and to select a recommended partner based on the standardized values. In other words, it is possible to more easily grasp the position of a specific holder in each class, and to understand where the company's strengths and weaknesses are. In addition, when any other company is selected as the partner, it is easier to select the partner by reading how much the total evaluation value for each class in each class and each partner's holder can do.

<3-2-6. Calculation of technical complement (S150)>
Next, a technical complement is calculated (S150).
FIG. 10 is a flowchart for explaining the details of the technology complementation degree calculation processing in FIG.

In order to calculate the technical complement degree, the technical supplement degree calculation unit 125 acquires a distribution tendency (overall distribution tendency) of evaluation values for each class (S151). The overall distribution tendency can be expressed by, for example, a vector (complementary reference vector) in which the number of classes is the number of dimensions and the evaluation value for each class is a vector component. In other words, when the distribution of evaluation values for each class is biased, a complementary reference vector with a large variation in component values is obtained, and when it is uniform, a complementary reference vector with a small variation in component values is obtained. The overall distribution tendency is not limited to vector expression, and may be expressed, for example, as a data string of evaluation values for each class.
If the evaluation value for each class for each holder is standardized in S140, since the evaluation values for each class are all 1, the overall distribution tendency is data representing a uniform state. If this is expressed by a vector, all the components are vectors of the same value (for example, 1) other than 0 (see Equation 4 described later).

  The technical complement degree calculation unit 125 acquires, for the holder of the analysis target (target holder), holder evaluation information in which the evaluation value for each class of the holder of the holder is associated with each other (S152). Nominee evaluation information can be expressed, for example, as a vector (nominee evaluation vector) in which the number of classes is the number of dimensions and the evaluation value for each class in each class of the nominee is a vector component (described later). (Refer to Equation 2). The holder evaluation information is not limited to vector expression, and may be expressed, for example, as a data string of evaluation values for each holder's class for each holder.

  The technical complement degree calculation unit 125 also provides, for each holder (candidate holder) other than the target holder, holder evaluation information (name of the candidate holder) that associates the evaluation value for each class of the holder of each holder. As for the vector representation of the evaluation information, the following expression 3 is obtained (S153). If the right holder evaluation information (or right holder evaluation vector) of the target holder and the candidate holder is acquired in S152 and S153, the process of the next S154 is performed using these.

The technical complement degree calculation unit 125 calculates, for each candidate holder, composite evaluation information obtained by adding, for each class, the holder evaluation information of the target holder and the holder evaluation information of each candidate holder. Further, a distribution tendency (composite distribution tendency) of values for each class is calculated for each combination evaluation information (S154).
The composite evaluation information is, for example, a data string of values (composite evaluation values) obtained by adding, for each class, an evaluation value for each class name of the target holder and an evaluation value for each class name of each candidate holder. Expressed.
The composite distribution tendency may be used as it is represented by this data string, or may be expressed by a vector having the data string of the composite evaluation information as a vector element. If the target holder and each candidate holder's holder evaluation information are both expressed as vectors (see nominee evaluation vectors) (see S152 and S153), adding these vectors immediately results in a combined distribution trend (synthesize (Evaluation vector) is calculated (see the following equation 5).
If the target holder and each candidate holder can collaborate and use each other's patents based on this composite evaluation information or composite distribution tendency (or composite evaluation vector), how many patents in each technical field (class) Can be easily predicted for each candidate holder who is a tie-up candidate.

ここで、ベクトルｆ_selfは対象名義人の名義人評価ベクトル、ベクトル成分Ｓ_class1乃至Ｓ_classDは対象名義人の名義人毎クラス毎評価値（規格化したものが好ましい）、次元数Ｄはクラスの数である。

ここで、ベクトルｆ_otherは候補名義人の名義人評価ベクトル、ベクトル成分Ｔ_class1乃至Ｔ_classDは候補名義人の名義人毎クラス毎評価値（規格化したものが好ましい）、次元数Ｄはクラスの数である。

ここで、ベクトルａは補完基準ベクトルである。次元数は、数２及び数３の次元数Ｄである。この式に例示されるベクトルａは全成分が１となっており、これは名義人毎クラス毎評価値を当該クラスでの総計が１となるように規格化した場合に適用される好ましい例の１つである。

ここで、ベクトルｆ_self＋ｆ_otherは合成評価ベクトルである。When the composite distribution tendency is calculated, the technology complementation degree calculation unit 125 calculates the degree of approximation (synthetic approximation degree) of each composite distribution tendency with respect to the overall distribution tendency (S155). This combined approximation is defined as a value indicating whether the combined distribution tendency of the target holder and the candidate holder is close to an ideal state that complements each other's weak fields. As this “ideal state of complementing each other's weak fields”, the overall distribution tendency is used in this embodiment. When the evaluation value for each class for each holder is standardized in S140, the overall distribution tendency indicates a uniform state (when expressed as a vector, all components are vectors having the same value other than 0). In this case, it can be said that the synthetic approximation is an approximation of the synthetic distribution tendency with respect to the uniform state.
Since the combined approximation is an approximation between the combined distribution trend and the overall distribution trend, for example, for each class, each data string indicating the combined distribution trend is divided by each data string indicating the overall distribution trend, and the calculation result It can be assumed that the smaller the variation is, the greater the degree of synthesis approximation is.
When both the composite distribution tendency and the overall distribution tendency are expressed by vectors (composite evaluation vector and complementary reference vector), it is preferable to use the cosine (combined cosine) of these vectors as the composite approximation. In other words, the right holder evaluation vector of the target holder is f _self (see _Equation 2), the _right holder evaluation vector of each candidate holder is f _other (see _Equation 3), and the complementary reference vector is a (for example, see Equation 4). ), The composite evaluation vector is f _self + f _other (see _Equation 5)
Therefore, the synthetic approximation is
{(F _self + f _other ) · a} / {| f _self + f _other || a |} = cos θ. This cos θ (combined cosine) takes a value range from 0 to 1.

Here, the vector f _self is a target person's holder evaluation vector, the vector components S _{class1 to} S _{class D} are target class member evaluation values (preferably normalized), and the dimension number D is a class number. Is a number.

Here, the vector f _other is a nominee evaluation vector of the candidate nominee, the vector components T _{class1 to} T _classD are evaluation values for each nominee of the candidate nominee (preferably standardized), and the number of dimensions D is the class Is a number.

Here, the vector a is a complementary reference vector. The number of dimensions is the number of dimensions D in Equations 2 and 3. The vector a illustrated in this formula has all components of 1, which is a preferable example applied when the evaluation value for each class for each holder is normalized so that the total in the class is 1. One.

Here, the vector f _self + f _other is a composite evaluation vector.

The combined approximation calculated in this way indicates how similar the combined distribution tendency is to the overall distribution tendency. Note that the overall distribution trend indicates social technology distribution in the entire technical area related to the acquired patent data. This social technology allocation shows the current state of equilibrium between the social needs for each technical element included in the technical domain and the corresponding technology supply capability of the entire society. This equilibrium state indicates the optimal technology allocation at the present time. Therefore, it is desirable for each company to have a technology portfolio that approximates the above social technology allocation. Therefore, by referring to each synthetic approximation, it is possible to determine which candidate holders can hold a technology portfolio that approximates the above-mentioned social technology distribution by making a technical tie-up. That is, it becomes possible to efficiently select an optimal alliance destination.
In addition, by calculating the cosine (combined cosine) of the combined evaluation vector and the complementary reference vector as the combined approximation, the spatial distance between the combined evaluation vector and the complementary reference vector can be quantitatively grasped. become. Therefore, the analyst can grasp the tie-up effect for each candidate holder numerically by this synthetic cosine.

The technical complement degree calculation unit 125 acquires a distribution tendency (namer distribution tendency) of values for each class of the right holder evaluation information of the target holder and further approximates the total distribution tendency of the right holder distribution tendency (target name). It is desirable to calculate (person approximation) (S156).
The distribution pattern of the right holders may take the same form of expression as the right holder evaluation information of the target right holder, or may be the form of the right holder evaluation vector of the right holder.
For example, for each class, the degree of approximation to the overall distribution tendency of the right holder distribution tendency is calculated by dividing each data string indicating the right holder distribution tendency by each data string indicating the overall distribution tendency. It can be assumed that the smaller the variation in the calculation results, the greater the degree of approximation of the target holder.
If both the nominal distribution tendency and the overall distribution trend are expressed as vectors (the nominal evaluation vector and the complementary reference vector), the cosine of the vector (the target nominal cosine) is used as the target nominal approximation degree. It is preferable. In other words, when the target person's holder evaluation vector is f _self and the complementary reference vector is a, the target holder approximation degree is
{F _self · a} / {| f _self || a |} = cos θ ₀ This cosθ ₀ (cosine of the target holder) takes a value range from 0 to 1.
If the above-mentioned target holder approximation degree is the same as or higher than the combined approximation degree of S155, even if it ties up with the candidate holder, it will not be close to the ideal state of complementing each other's weak fields. There will be no change, or on the contrary, it will move away from the ideal state, and the effect of the alliance cannot be expected. Therefore, when selecting a recommended partner, it is desirable to select one with a higher synthetic approximation than the target holder approximation. Thereby, among the plurality of candidate holders, the candidate holders who can approach the overall distribution tendency by partnering with the target holders (closer to the complementary reference vector than the target holder evaluation vector of the target holders) Only the candidate nominee corresponding to the composite evaluation vector) can be selected as the partner. As a result, alliance partners with high technical supplementary effects can be selected as recommended partners.
In particular, in order to reflect the target holder approximation degree in the technical complementation degree, it is desirable to calculate the technical complementation degree as follows.

The technical complement degree calculation unit 125 calculates the technical supplement degree as follows (S157).
Subtract the target holder cosine from the composite cosine. If the synthetic cosine cosθ, the target holder cosine was cosθ _0, to calculate the cosθ-cosθ _0. As a result of this subtraction, if the composite cosine and the target holder cosine are equal (a complementary effect cannot be expected even if the partnership is made), the result is 0, and the composite cosine is smaller than the target holder cosine (even if the partnership is made, it is opposite to the complement). ) Is negative.
The result of this subtraction may be used as the technical complement, but in the present embodiment, this is further normalized using the value of the subject nominal cosine.
For example, normalization is performed by dividing the result of the subtraction by (maximum value that the combined approximation can take) − (each target holder approximation). Thereby, the maximum value of the degree of technical complementarity can be made constant regardless of the value of the target holder approximation degree. If the composite cosine is cosθ and the target holder cosine is cosθ ₀ , the technical complement is
_{(Cosθ-cosθ 0) / (} 1-cosθ 0)
Can be calculated. The maximum value of the degree of technical complementation is 1 regardless of the size of the target holder cosine cosθ ₀ .
Further, for example, when the synthetic cosine is cos θ and the target holder cosine is cos θ ₀ , the result of the subtraction is normalized by dividing by cos θ (1-2 cos θ ₀ ) + cos θ ₀ . In other words, the degree of technical complementation,
(Cos θ-cos θ ₀ ) / {cos θ (1-2 cos θ ₀ ) + cos θ ₀ }
Calculate with In this case, the degree of technical complementation is a maximum value of 1 and a minimum value of -1.
When the technical complement degree is calculated, the processing in FIG. Here, FIG. 11 shows an example of the calculation result of the technical complement degree.
FIG. 11 is a table illustrating a calculation result of the technology complementation degree calculation process according to the present embodiment. In addition, in FIG. 11, in order to simplify description, the case where the number of classes to be analyzed is 8 is shown. Only the calculation results are shown. In addition, here, an example in which the evaluation value for each class for each holder is standardized is shown. Technology complements degree here is calculated in the above _{(cosθ-cosθ 0) / (} 1-cosθ 0). In addition, the evaluation value for each class obtained in the calculation process of the degree of technical complementation and the synthetic cosine cosθ are also shown. By referring to the degree of technical complementation, it is possible to know which candidate nominee can obtain a comprehensive supplementary effect.

<3-2-7. Selection of recommended partner companies (S160)>
When the degree of technical complement is calculated, the process returns to FIG. 4, and a recommended partner company is preferably selected (S 160) and output together with the degree of technical complement that is the basis for selection.
Selection of the recommended partner company can be performed, for example, by selecting candidate holders whose technical complement degree is a predetermined value or more, selecting a predetermined number of candidate holders in descending order of technical complement degree, and the like.
Further, not limited to the case of using the technical complementation degree of S157, for example, from the synthesis approximation degree of S155, those having a predetermined threshold value or more are selected, the synthesis evaluation information having the selected synthesis approximation degree is identified, and the identified synthesis is performed. A candidate holder corresponding to the holder evaluation information used for calculating the evaluation information may be a recommended partner.
Further, not only when using the technical complementation degree of S157, for example, a higher one than the target nominal person approximation degree of S156 is selected from the synthesis approximation degrees of S155, and the synthesis evaluation information having the selected synthesis approximation degree is specified. The candidate holder corresponding to the holder evaluation information used for the calculation of the specified composite evaluation information may be set as the recommended partner. Thereby, it becomes possible to select only the candidate holders who are close to the distribution of the entire patent data as a partner from among the plurality of candidate holders. As a result, alliance partners with high technical supplementary effects can be selected as recommended partners.

<3-3. Technological depth (S200-S260)>
Next, calculation of the technical deepening level and selection of a recommended partner company in a specific technical field based on the technical deepening level will be described with reference to the flowchart of FIG. The processes in S200 to S240 in FIG. 4 are the same as the processes in S100 to S140, and a duplicate description is omitted. When calculating both the technical complementation level and the technical deepening level for the same target holder for the same patent data, either the processing of S100 to S140 or the processing of S200 to S240 may be omitted. .

<3-3-1. Calculation of technical deep tillage (S250)>
After calculating the evaluation value for each class for each holder (desired to be standardized) by the processing up to S240, the technical deepening level is calculated (S250).
FIG. 12 is a flowchart for explaining details of the processing for calculating the technical depth in FIG.

  In order to calculate the technical depth, the technical depth calculator 130 acquires, for the holder of the analysis target (target holder), the holder evaluation information that associates the evaluation value for each class of the holder of the holder. (S251). The holder evaluation information can be expressed by, for example, a vector (nominee evaluation vector) in which the number of classes is the number of dimensions and the evaluation value for each class of each holder in each class of the holder is a vector component. The holder evaluation information is not limited to vector expression, and may be expressed, for example, as a data string of evaluation values for each holder's class for each holder. It is desirable that the evaluation value for each class for determining the holder evaluation information is standardized in S240.

  The technical deepening degree calculation unit 130 also acquires, for each holder (candidate holder) other than the target holder, the holder evaluation information in which the evaluation value for each class of the holder of the holder is associated with each other (S252).

  The technical plow level calculation unit 130 identifies a class whose evaluation value for each class per person is a positive number, from among the nominee evaluation information of the candidate nominee (S253). For classes with a non-positive number (for example, 0) per class for each holder, even if the composite evaluation information is calculated in the next S254, it is not different from the target holder's holder evaluation information, and deepening of the technology is expected. It is not possible.

If the right holder evaluation information of the target holder and the candidate holder is acquired in S251 to S253, and the class whose evaluation value for each holder is a positive number is specified from the holder evaluation information of the candidate holder, the technical depth The calculation unit 130 calculates, for each candidate holder, composite evaluation information obtained by adding the target holder's holder evaluation information and the candidate holder's holder evaluation information for each of the specified classes for the specified class. (S254). Thereby, it is possible to calculate how much deep plowing can be expected for each class that can be expected to deepen the technology.
Note that the composite evaluation information may be calculated for all classes in S254 without specifying the classes in S253. In this case, the same result can be obtained by extracting a value that is fluctuating from the value of the corresponding class of the target person's holder evaluation information from the values of each class of the composite evaluation information.

ここで、ベクトルｆ_self＋ｆ_otherは合成評価ベクトルである。ここでは、候補名義人の名義人毎クラス毎評価値のうちＴ_class3及びＴ_class5乃至Ｔ_classDが０であると仮定した場合を示した。

この式は、数６に示す合成により変動があった成分Ｓ_class1＋Ｔ_class1、Ｓ_class2＋Ｔ_class2、及びＳ_class4＋Ｔ_class4のうちの最大値で技術深耕度が求められることを示す。
なお、上記の例では、合成により変動があった成分Ｓ_class1＋Ｔ_class1、Ｓ_class2＋Ｔ_class2、及びＳ_class4＋Ｔ_class4を抽出し、その抽出したもののうちから最大のものを技術深耕度として求めるようにしているが以下のようにしてもよい。
具体的には、技術深耕度算出部１３０は、上記（数６）を用いた合成処理により求めたベクトルから成分（ベクトル要素）を抽出する際、「対象名義人の名義人評価ベクトルｆ_selfの成分（ベクトル要素）が「０」ではないこと」という要件を加えるようにする。すなわち、技術深耕度算出部１３０は、上記（数６）により求めた合成ベクトルから成分（ベクトル要素）を抽出する際、変動があった成分（ベクトル要素）であり、且つ上記合成前の名義人評価ベクトルｆ_selfの成分（ベクトル要素）が「０」ではない成分（ベクトル要素）を抽出する。このようにすることにより、自社が特許を保有している技術分野に絞って、技術深耕という観点からの提携先企業を選定することができるようになる。When the composite evaluation information is calculated, the technical depth calculator 130 calculates, for each of the candidate holders, the total value of the evaluation values for each class with the target holder in the composite evaluation information (limited to the changing values). ) Is identified as the maximum class, and the total value is set as the technical depth (S255).
In the present embodiment, the specific procedure of the process for calculating the technical depth is not particularly limited. For example, when calculating the technical depth using a vector, the following procedure is used. Also good. Here, similarly to the above-described calculation of the degree of technical complement, the target holder evaluation vector of the target holder is set to f _self (see the above formula 2), and the candidate holder evaluation vector of each candidate holder is f _other (the above formula 3). And a vector (see Equation 6) obtained by combining f _self and f _other is calculated. Then, using the calculation formula shown in the following “Equation 7”, the maximum value among the components that have changed due to the synthesis is set as the technical depth.

Here, the vector f _self + f _other is a composite evaluation vector. Here, a case is shown in which T _class3 and T _{class5 to} T _classD are assumed to be 0 among the evaluation values for each class of candidate names.

This equation indicates that the technical depth is obtained at the maximum value among the components S _class1 + T _class1 , S _class2 + T _class2 , and S _class4 + T _class4 that have changed due to the synthesis shown in Equation 6.
In the above example, components S _class1 + T _class1 , S _class2 + T _class2 and S _class4 + T _class4 that have changed due to the synthesis are extracted, and the maximum one of the extracted ones is obtained as the technical depth. However, it may be as follows.
Specifically, when the technical depth calculator 130 extracts the component (vector element) from the vector obtained by the synthesis process using the above (Equation 6), “the target holder's nominal person evaluation vector f _self The requirement that the component (vector element) is not “0” is added. In other words, the technical depth calculation unit 130 is a component (vector element) that has changed when extracting a component (vector element) from the combined vector obtained by (Equation 6) above, and the nominee before the combining. A component (vector element) whose component (vector element) of the evaluation vector f _self is not “0” is extracted. By doing so, it becomes possible to select partner companies from the viewpoint of deepening technology, focusing on the technical fields in which the company owns patents.

Here, an example of the calculation result of the technical depth is shown.
FIG. 13 is a table exemplifying the calculation result of the technical plowing degree calculation process according to this embodiment. FIG. 13 shows a case where the number of classes to be analyzed is 8 in order to simplify the description, and the target holder A company and some candidate holders (candidate holders shown in FIGS. 8 and 9). Only the calculation results with Company B to Company J) are shown. Moreover, the example in the case of standardizing the evaluation value for each class for each holder is shown.
In the table shown in the figure, the highest-level row 1301 shows the value for each class of the holder evaluation information of the target holder A company. Further, in each row after the row 1302, the value for each class of the composite evaluation information of the target holder A company and each candidate holder, and the technical depth when the target holder A company cooperates with each candidate holder Is shown.
For example, the row 1302 shows the synthesis evaluation information of the target holder A company and the candidate holder B company, and the technical depth of the case where the target holder A company has partnered with the candidate holder B company. Here, a procedure for calculating the technical deepening level will be described using the value of the composite evaluation information shown in the row 1302. The technical deep-cultivation degree calculation unit 130 extracts and extracts a value that has fluctuated from the nominee evaluation information of the target holder A company from the value of the composite evaluation information of the target nominee A company and the candidate nominee B company. The value is substituted into Equation 7 described above. Specifically, the technical plow level calculation unit 130 determines the values of classes 1 to 4 and classes 6 to 8 from the values of each class of the composite evaluation information of the target holder A and the candidate holder B. Is extracted, and the extracted value is substituted into Equation 7. That is, “0.3310”, “0.0216”, “0.2641”, “0.2885”, “0.3771”, “0.4627”, and “0.2993” are assigned to Equation 7. Is done. Note that the value of the class 5 of the composite evaluation information of the target holder A company and the candidate holder B company is not extracted because it does not vary from the value of the class 5 of the holder evaluation information of the target holder A company.
Then, the technical plowing degree calculation unit 130 sets “0.4627”, which is the maximum value among the values assigned to Equation 7, as the technical plowing degree (class 7 value). In addition, regarding the rows 1303 to 1310, the technical depth is obtained in the same procedure.

  As described above, according to the present embodiment, when a technical tie-up is made for each candidate candidate for an alliance candidate, it is possible to specify a technical field in which the effect of the technical tie-up is most obtained. Therefore, for example, when companies that can form an alliance are narrowed down based on competitive relationships with the company, for example, it is possible to grasp the technical field where the effect of the technical alliance is large for each narrowed down company. Further, for example, it is possible to obtain a local alliance effect after specifying a technical field where the effect of the technical alliance is large even with a company that cannot obtain a comprehensive alliance effect.

Focusing only on a certain class from the beginning, not only the technical depth determined in this way, the evaluation value of the target holder in that class (the evaluation value for each class for each holder) and the evaluation value of each candidate holder ( The sum (synthetic evaluation value) of the evaluation value for each class for each holder may be obtained for each candidate holder, and this may be used as the technical depth. Thereby, the alliance effect can be judged only by the specific class focused on. Therefore, for example, this is effective when a technical field in which an alliance is desired can be assumed.
In this case, since comparison between different classes is unnecessary, standardization of the evaluation value for each class for each holder (S240) is not necessary.

<3-3-2. Selection of recommended partner companies (S260)>
After calculating the technology deepening level, the process returns to FIG. 4, and a recommended partner company is preferably selected (S260), and is output together with the technical deepening level as a basis for selection.
Selection of the recommended partner company can be performed, for example, by selecting candidate holders whose technical plowing degree is equal to or higher than a predetermined value, selecting a predetermined number of candidate holders in descending order of the technical deepening degree, and the like.

<3-4. Alliance distribution map (S400)>
After calculating the technical complementation level and the technical deepening level, it is desirable to create and output an alliance distribution map that summarizes both (S400). The alliance distribution map is created in the following procedure, for example. That is, the technical complement degree calculation unit 125 outputs the calculation result to the output unit 135 after calculating the technical supplement degree by the processing of S150 (see FIG. 4). After calculating the technology plowing degree by the process of S250 (see FIG. 4), the technical plowing degree calculation unit 130 outputs the calculation result to the output unit 135. When the output unit 135 receives the calculation result from the technical complementation degree calculation unit 125 and the calculation result from the technical deepening degree calculation unit 130, the alliance information in which the technical complementation degree and the technical deepening degree are associated with each candidate holder. Is generated. Here, FIG. 14 shows an example of the alliance information in which the technical complementation level and the technical deepening level are associated with each candidate holder. In addition, FIG. 14 shows the case where the number of classes to be analyzed is 8 as in FIGS. 11 and 13, and shows only the calculation results between the target holder A and some candidate holders. Yes. Moreover, the example in the case of standardizing the evaluation value for each class for each holder is shown. Moreover, the class (corresponding class corresponding to the maximum value of the fluctuation component) specified for calculating the technical depth is also shown.
Then, the output unit 135 generates image information (visualized information) representing an alliance distribution diagram illustrated in FIG. 15 by using the illustrated alliance information. The output unit 135 outputs the generated image information to the output device 4. As a result, for example, when a liquid crystal display is used as the output device 4, an alliance distribution map is displayed on the liquid crystal display.
FIG. 15 is an example of an alliance distribution diagram output according to the embodiment of the present invention.
In this distribution chart, each candidate holder is plotted with the technical complementarity on the horizontal axis and the technical deepening level on the vertical axis. Among the candidate holders, it can be seen that the companies B, E, F, G, and I and the company A who is the target holder have a high degree of technical complementation, and a comprehensive supplementary effect can be expected from the alliance. In addition, it can be seen that the companies C, H, and J have a high level of technological deepening for Company A and can be expected to be deeply cultivated through alliances in specific fields. Although FIG. 15 shows an example in which the degree of technical complementation is normalized and plotted with an average of 0 and a variance of 1, the calculated values may be plotted as they are without being standardized.

<4. Another aspect of partner selection information generation processing>
FIG. 16 is a flowchart illustrating a process for generating the graph of FIG. 16, S300 to S340 are the same as S100 to S140 in FIG. 4, and S352 and S353 are the same as S152 and S153 in FIG.
The evaluation value calculation unit 120 outputs the calculation result to the output unit 135 after calculating an evaluation value (standardized) for each class in S352 and S353. Upon receiving the calculation result from the evaluation value calculation unit 120, the output unit 135 associates, for each class, a standardized evaluation value for each standardized holder of the holder belonging to the class, and each class is represented by the vertical axis. Image information indicating a bar graph with the horizontal axis representing the evaluation value for each name holder is generated (visualized information is generated), and the generated image information is output to the output device 4 (S360). As a result, for example, when a liquid crystal display is used for the output device 4, the graph illustrated in FIG. 9 is displayed on the liquid crystal display.
In this way, in the present embodiment, the graph illustrated in FIG. 9 is generated and output, so that relative comparisons between technical fields that are difficult to directly compare with each other because the absolute value criteria are different are possible. It becomes. By looking at the bar graph output to the output device 4, the analyst can grasp the strengths and weaknesses of the entire proprietary technology based on a relative comparison with other companies. As a result, the analyst can select an alliance destination after grasping the balance of technical capabilities between the company and other companies.

<5. Modification>
The present invention is not limited to the embodiment described above, and various modifications can be made within the scope of the gist of the present invention.
<5-1. Modification 1>
In the above embodiment, each functional unit (the control unit 100, the data acquisition unit 105, the cluster analysis unit 110, the evaluation value calculation unit 120, the technical complementation degree calculation unit 125, the technical deepening level calculation unit 130, and the output unit of the processing device 1 However, the present invention is not limited to this example. Each functional unit of the processing device 1 may be realized by a circuit (ASIC (Application Specific Integrated Circuit) or the like) designed exclusively for executing each functional unit.
<5-2. Modification 2>
Moreover, in the said embodiment, although the information processing apparatus 1 took the case where patent data, such as patent publications etc. of evaluation object, are acquired from the memory | storage device 2, it does not specifically limit to this. For example, the information processing apparatus 1 may communicate with an external information providing server via a network such as the Internet and acquire patent data from the external information providing server.
<5-3. Modification 3>
In the above embodiment, the information processing apparatus 1 that has received a request from the analyst via the input device 3 generates information for selecting a partner, and outputs the generated information to the output device 4. However, this is merely an example. For example, in a client / server model computer system, the function of the information processing apparatus 1 may be provided as a server. In this case, the information processing apparatus 1 is provided with a communication function for communicating with a client terminal connected to the network. When the information processing apparatus 1 receives a request from the client terminal, the information processing apparatus 1 provides the function of the information processing apparatus 1 to the client terminal. The network may be wired or wireless. For example, a telephone line or the Internet may be used for the network.
<5-4. Modification 4>
In the above embodiment, a cluster analysis technique is used for classification of patent data (S110 in FIG. 4), but the invention is not particularly limited to this. For example, instead of the cluster analysis method, patent data may be classified using existing statistical methods (factor analysis, principal component analysis, etc.).
<5-5. Modification 5>
In the above embodiment, an example is shown in which each class is standardized so that the sum of evaluation values for each class for each holder is 1. However, the present invention is not particularly limited to this. You may make it represent evaluation value for every class for every holder by another method. For example, the evaluation value for each class for each holder may be standardized with an average of 0 and a variance of 1, and the value may be applied to this embodiment. Also in this case, the technical complementation level and the technical deepening level can be calculated as in the above-described example.
<5-6. Modification 6>
In the above-described embodiment, the distribution tendency of the evaluation value for each class is regarded as the overall distribution tendency, and the technical supplement degree is obtained by the approximate degree of the combined distribution tendency with respect to the overall distribution tendency. However, the present invention is not limited to this. . For example, the distribution tendency of the value obtained by assigning a predetermined weight to each evaluation value for each class may be used as the overall distribution tendency (a vector having the elements obtained by assigning a predetermined weight to each evaluation value for each class as an element is a complementary reference vector. As well). This weighting is, for example, when a class that is important for some reason or especially a class that is to be strengthened is specified by an alliance, and the evaluation value for each class of such a class is larger than the evaluation value for another class. And Thereby, the so-called ideal distribution tendency can be set as an overall distribution tendency (or a complementary reference vector). The overall cooperation effect is also observed under the above weighting based on the degree of technical complementation obtained from the degree of approximation (or cosine) of the synthetic distribution tendency (or synthetic evaluation vector) with respect to the overall distribution tendency (or complementary reference vector). be able to.
<5-7. Modification 7>
Further, in the above embodiment, the case where the target holder and the candidate holder are each a company unit has been described, but the present invention is not particularly limited thereto. For example, a corporate group that gathers multiple companies (for example, a group of companies that belong to the same group, a group that combines affiliated companies, a parent company, a subsidiary, an affiliated company, etc. subject to consolidated financial statements) It is also possible to create a group of companies or the like arbitrarily combined by a group or a user and treat the company group as a target holder or a candidate holder. In this case, for example, a part of the processing of FIG. 4 described above is changed as follows.
Specifically, prior to the start of the processing of S100, the control unit 100 receives an instruction from the user as to whether or not to accept the target holder as a corporate group via the input device 3. When the control unit 100 receives an instruction from the user to accept the target holder as a company group, the control unit 100 further receives designation of a company group forming the company group. In addition, the control unit 100 receives an instruction as to whether or not there is a company group that is accepted as a company group among the candidate holders. When the control unit 100 receives an instruction indicating that there is a company group to be accepted as a company group among the candidate holders, the control unit 100 further receives designation of the company group forming the company group. Then, the control unit 100 notifies the cluster analysis unit 110 of information indicating the group of companies that form the received company group, and instructs the data acquisition unit 105 to start the process of S100.
Next, the data acquisition unit 105 acquires patent data according to the same procedure as in the above-described embodiment, and transitions to the processing of S110.
In S110, the cluster analysis unit 110 classifies the acquired patent data into classes for each predetermined technical field according to the same procedure as in the above-described embodiment. Then, when classifying each class for each holder, the cluster analysis unit 110 classifies the company group forming the company group as one holder for the company group forming the company group that has received the above notification. . Further, the cluster analysis unit 110 classifies the company group other than the company group forming the company group by the company unit as in the above-described embodiment.
Then, the subsequent processing (S130 to 160, S400) is performed by the same procedure as in the above-described embodiment. Here, since the group of companies forming the company group is classified as one holder, the evaluation value for each class for each holder is calculated for the company group received from the user. Thereafter, a technology complementation degree calculation process (S150) and a recommended partner company selection process (S160) are performed.
As for the process for calculating the technical depth (S200 to S260), prior to the process of S200, the designation of the company group forming the company group is accepted, and in S210, the company group forming the company group is set to 1 You should classify as nominee.
By configuring in this way, the user can consider the alliance destination with the company in units of company groups, not in units of companies. In addition, alliance companies can be considered based on the corporate group to which the company belongs. Alternatively, alliance destinations can be examined not in units of companies but in units of companies, based on the company group to which the company belongs.

<5-8. Modification 8>
Moreover, in the said embodiment, the example which calculates the evaluation value for every class for every holder using the patent impact and progress information index is demonstrated for every class for every holder in S120-S130 (or S220-S230). However, it is not particularly limited to this. For example, an individual evaluation value for each patent gazette belonging to each class may be calculated, and the sum of the calculated values may be used as the evaluation value for each class for each holder. For example, S120 to S130 (or S220 to S230) in FIG. 4 may be as follows.
That is, prior to the processing of FIG. 4, the evaluation value calculation unit 120 uses the individual evaluation values (hereinafter referred to as “patent score PS”) for each patent data 200 using the progress information 240 and the content information 230 included in each patent data 200. ”). And the evaluation value calculation part 120 calculates | requires what calculated the calculated patent score PS for every class for every holder as an evaluation value for every class for every holder in S120-S130 (S220-S230). Hereinafter, Modification 8 of the present embodiment using the patent score PS will be described.

  In the present modification 8, the evaluation value calculation process for each class per holder performed by the evaluation value calculation unit 120 is changed among the processing steps of FIG. 4 described above. In the present modification 8, only the function of the evaluation value calculation unit 120 is partially changed, and other configurations are the same as those in the above embodiment. In addition, this modification 8 will be described using the same reference numerals as in the above embodiment for the convenience of description. Moreover, in this modification 8, only a different part from the said embodiment is demonstrated.

Specifically, the evaluation value calculation unit 120 of the present modification 8 performs processing for generating partner selection information according to the processing steps shown in FIG. 17 in which a part of the processing in FIG. 4 is changed.
Further, the evaluation value calculation unit 120 of the present modification 8 calculates the patent score PS for each patent data 200 before performing the processing of FIG. 17, and stores the calculated patent score PS in the memory (auxiliary of the information processing apparatus 1). It is stored in the storage device 12 or the main storage device 11) (or stored in the storage device 2). Then, the evaluation value calculation unit 120 stores the patent score PS of each patent data 200 in the memory, and then performs the process of FIG. The processing procedure for calculating the patent score PS will be described in detail later.

Next, processing for generating partner selection information performed using the patent score PS will be described with reference to FIG.
FIG. 17 is a flowchart showing partner selection information generation processing of the information processing apparatus according to the eighth modification. In addition, the flowchart of FIG. 17 changes S120-S130 (S220-S230) of the flowchart of FIG. In FIG. 17, the same processes as those in FIG. 4 are denoted by the same reference numerals.
In the following description, processing steps different from those in FIG. 4 will be described, and description of the same processing will be omitted.

As shown in the figure, when the cluster analysis (S110) is completed and classified for each nominee, the evaluation value calculation unit 120 displays all the patent data subject to the cluster analysis (for all classes generated by the cluster analysis). The patent score PS is acquired for the patent data 200) (S121).
Specifically, the evaluation value calculation unit 120 reads the patent scores PS of the patent data 200 of all classes from the memory of the information processing apparatus 1.

この式の分子は各特許データｉのパテントスコアＰＳｉから全クラスでの平均値を減算したものであるから、標準値Ｐ_0,iの全クラスでの平均値は０となる。分母はパテントスコアＰＳｉの全クラスでの標準偏差であるから、標準値Ｐ_0,iの全クラスでの分散及び標準偏差は１となる。なお、式中のｎは全クラスの特許データ数であるが、パテントスコアＰＳが０であるもの（後述のように失効した特許出願又は消滅した特許権）は除いた特許データ数とするのが好ましい。
この標準値Ｐ_0,iを各特許データの個別評価値としても良いが、本実施形態では更に以下のようにして各特許データの個別評価値を算出する。
すなわち、評価値算出部１２０は、上記求めた標準値Ｐ_0,iおよび以下に示す（数９）により、各特許データ２００の偏差値（全クラスにおける偏差値）を求め、これを個別評価値とする。

この（数９）では、上記（数８）で求めた標準値Ｐ_0,iに１０を乗算し５０を加算しているので、この（数９）により求められた個別評価値（偏差値）Ｐｉは、全クラスにおける平均が５０、標準偏差が１０となる。Next, when the patent value PS of the patent data 200 of all classes is read, the evaluation value calculation unit 120 calculates an individual evaluation value of each patent data (S122). The calculation method of the individual evaluation value is not particularly limited, and the obtained patent score PS can be used as it is as the individual evaluation value of the patent data. For example, according to the procedure shown below, the patent data The individual evaluation value may be obtained.
First, the evaluation value calculation unit 120 calculates the standard value (standard value in all classes) P _{0, i} of each patent data according to the following (Equation 8).

Since the numerator of this formula is obtained by subtracting the average value in all classes from the patent score PSi of each patent data i, the average value in all classes of the standard value P _{0, i} is 0. Since the denominator is the standard deviation in all classes of the patent score PSi, the variance and standard deviation in all classes of the standard value P _{0, i} is 1. Note that n in the formula is the number of patent data in all classes, but the number of patent data excluding those with a patent score PS of 0 (expired patent application or expired patent right as described below). preferable.
The standard value P _{0, i} may be used as the individual evaluation value of each patent data, but in the present embodiment, the individual evaluation value of each patent data is further calculated as follows.
That is, the evaluation value calculation unit 120 obtains a deviation value (deviation value in all classes) of each patent data 200 from the obtained standard value P _{0, i} and (Equation 9) shown below, and obtains the individual evaluation value And

In this (Equation 9), the standard value P _{0, i} obtained in (Equation 8) is multiplied by 10 and 50 is added, so the individual evaluation value (deviation value) obtained by this (Equation 9). Pi has an average of 50 and a standard deviation of 10 in all classes.

この（数１０）では、上記（数８）で求めた標準値Ｐ_0,iがしきい値Ｐｓｔｄ以上である場合は、標準値Ｐ_0,iのしきい値Ｐｓｔｄとの差分を１０倍したものを個別評価値Ｐｉとし、標準値Ｐ_0,iがしきい値Ｐｓｔｄ未満である場合は、０を個別評価値Ｐｉとしている。
しきい値Ｐｓｔｄとしては、ここでは標準値Ｐ_0,iの全クラスでの平均値（上記（数８）のＰ_0,iによれば０となる）を用いている。従ってこの場合、個別評価値Ｐｉが負数となることはない。従って、後述のＳ１３１の（数１２）を用いた名義人毎クラス毎評価値Ｓ_k,jが負数となることはない。In S122 (S222), the evaluation value calculation unit 120 may obtain the individual evaluation value of the patent data 200 by using the following (Equation 10) instead of the above (Equation 9).

In (Equation 10), when the standard value P _{0, i} obtained in (Equation 8) is equal to or greater than the threshold value Pstd, the difference between the standard value P _{0, i and} the threshold value Pstd is multiplied by 10 When the standard value P _{0, i} is less than the threshold value Pstd, 0 is set as the individual evaluation value Pi.
As the threshold value Pstd, here, an average value in all classes of the standard value P _{0, i} (it becomes 0 according to P _{0, i} in the above (Equation 8)) is used. Therefore, in this case, the individual evaluation value Pi does not become a negative number. Therefore, the evaluation value S _{k, j for} each class using the expression (Equation 12) of S131 described later does not become a negative number.

There is the following difference between the case where (Equation 9) is used and the case where (Equation 10) is used.
The individual evaluation value of the patent data 200 is used to calculate an evaluation value for each class for each holder in (Equation 12) described later. When the holder is a large company or the like (for example, a major electronics manufacturer), the number of patent data 200 is enormous (because the number of patents held is enormous). Even if the patent score PS is a low value, by summing the low value patent score PS (or the deviation value obtained from the patent score PS), the evaluation value of the holder (evaluation value for each holder) is obtained. Get higher. That is, when the deviation value of (Equation 9) is used for aggregation as it is, the evaluation value for each class for each holder is an analysis result that is strongly influenced by the number of patent data being applied.
On the other hand, in the case of the above (Equation 10), only the patent score PS that is a standard value equal to or higher than the threshold value Pstd is used as the calculation target of the evaluation value for each class for each holder, and the standard that is less than the threshold value is used. By not adding the patent score PS as a value, the influence of the number of patents for each holder can be reduced, and the “quality” of the patent data possessed by the holder can be reflected. Further, since the individual evaluation value Pi is calculated from the difference between the standard value P _{0, i and} the threshold value Pstd, even if there are a large number of patents near the threshold value Pstd, it is given to the evaluation value for each class per holder. The impact is small, and high quality patents that exceed the threshold greatly affect the evaluation value for each class per holder.
In the above (Equation 10), the average value in all classes of the standard value P _{0, i} is used as the threshold value Pstd, but this is merely an example. The user may arbitrarily set the value of the threshold value Pstd, such as using the average value of the target holders.

すなわち、個別評価値を算出しようとする特許データ２００が共同出願に係る特許出願又は共有名義の特許権に係るものである場合には、記憶装置２に記憶されている当該特許の共有者の数（共同出願人の数又は特許権共有者の数）を読み取る。そして、上記（数９）又は（数１０）の右辺で算出される値を、更に当該特許の共有者の数で除算して、個別評価値Ｐｉとする。
こうして名義人毎の適正な個別評価値Ｐｉを算出し、後述の（数１２）により名義人毎クラス毎評価値を算出すれば、Ｓ１４０やＳ２４０の規格化処理等において用いるクラス毎評価値の算出にあたって、共同出願の特許データに係る個別評価値の二重計上を避けることができる。また、Ｓ１５０の技術補完度算出処理やＳ２５０の技術深耕度算出処理において用いる合成評価情報の算出にあたっても、同様に共同出願の特許データに係る個別評価値の二重計上を避けることができる。
また（数１１）では共有者の数で除算することにより、（数９）又は（数１０）の右辺で求められる値を共有者間で均等に按分したが、これに限らず個々の共有者の貢献度や持分比率で重み付けをして按分しても良い。In S122 (S222), the evaluation value calculation unit 120 uses the value obtained on the right side of the above (Equation 9) or (Equation 10), and further uses (Equation 11) shown below. The individual evaluation value of the patent data 200 may be obtained.

That is, when the patent data 200 for which the individual evaluation value is to be calculated is related to a patent application related to a joint application or a patent right under a common name, the number of sharers of the patent stored in the storage device 2 Read (number of joint applicants or patent sharers). Then, the value calculated on the right side of (Equation 9) or (Equation 10) is further divided by the number of persons sharing the patent to obtain an individual evaluation value Pi.
Thus, by calculating the appropriate individual evaluation value Pi for each holder and calculating the evaluation value for each class according to (Equation 12) described later, calculation of the evaluation value for each class used in the normalization processing of S140 and S240 and the like. In doing so, it is possible to avoid double counting of individual evaluation values relating to patent data of joint applications. In addition, double calculation of individual evaluation values related to patent data of a joint application can be similarly avoided in calculating composite evaluation information used in the technical complementation degree calculation process of S150 and the technical deepening degree calculation process of S250.
Further, in (Equation 11), the value obtained on the right side of (Equation 9) or (Equation 10) is divided equally among the sharers by dividing by the number of sharers. You may apportion by weighting with the degree of contribution or equity ratio.

すなわち、名義人毎クラス毎評価値Ｓ_k,jは、或るクラスｊに属する名義人ｋの特許データ２００の個別評価値Ｐｉの総和で算出される。
なお、上記（数９）又は（数１０）の右辺で求められる値を、（数１１）により共有者間で按分せずにそのまま個別評価値Ｐｉとする場合、Ｓ１４０やＳ２４０の規格化処理等において用いるクラス毎評価値の算出にあたっては、上記（数１）の分母に示されるように名義人毎クラス毎評価値を当該クラス内で単に総計すると、共同出願の特許データに係る個別評価値が二重計上されることになる。そこで、上記（数９）又は（数１０）の右辺で求められる値をそのまま個別評価値Ｐｉとする場合、例えば、クラス毎評価値の算出は、当該クラスｊに属する全名義人の特許データ２００の個別評価値Ｐｉの総和で算出することが望ましい。これにより、複数名義人の共同出願に係る特許データが例えば１件あれば、その特許データの個別評価値は１件分としてクラス毎評価値にカウントされるため、二重計上を避けることができる。
また、Ｓ１５０の技術補完度算出処理やＳ２５０の技術深耕度算出処理において用いる合成評価情報の算出にあたっても、同様に共同出願の特許データに係る個別評価値の二重計上を避けることが望ましい。そこで、上記（数９）又は（数１０）の右辺で求められる値をそのまま個別評価値Ｐｉとする場合、例えば、合成評価情報は、名義人毎クラス毎評価値を加算して求めるのではなく、対象名義人と或る候補名義人の一方又は両方が名義人に含まれる特許データ２００を抽出し、この抽出された特許データの個別評価値Ｐｉをクラス毎に総和することにより算出することが望ましい。これにより、対象名義人と当該候補名義人の共同出願に係る特許データが例えば１件あれば、その特許データの個別評価値は１件分として合成評価情報にカウントされるため、二重計上を避けることができる。なお、こうして算出された合成評価情報は、Ｓ１４０やＳ２４０の名義人毎クラス毎評価値の規格化と同様に、クラス毎に、クラス毎評価値で除算することにより規格化することが望ましい。Next, the evaluation value calculation unit 120 obtains the individual evaluation value for each patent data 200 obtained in S122 (for example, the individual evaluation value obtained in (Equation 9), (Equation 10), more preferably (Equation 11)) Pi. Is used to obtain an evaluation value for each class for each class (S131). Specifically, the evaluation value calculation unit 120 obtains an evaluation value for each class for each class by the following (Equation 12).

That is, the evaluation value S _{k, j for} each class per holder is calculated as the sum of the individual evaluation values Pi of the patent data 200 of the holder k belonging to a certain class j.
When the value obtained from the right side of (Equation 9) or (Equation 10) is used as it is as the individual evaluation value Pi without apportioning among the sharers according to (Equation 11), the normalization processing of S140 and S240, etc. In calculating the evaluation value for each class used in the above, if the evaluation value for each class for each holder is simply totaled within the class as shown in the denominator of (Equation 1), the individual evaluation value related to the patent data of the joint application is obtained. It will be doubled. Therefore, when the value obtained on the right side of (Equation 9) or (Equation 10) is used as it is as the individual evaluation value Pi, for example, the evaluation value for each class is calculated by the patent data 200 of all the holders belonging to the class j. It is desirable to calculate the sum of the individual evaluation values Pi. As a result, if there is, for example, one piece of patent data related to a joint application by a plurality of holders, the individual evaluation value of that patent data is counted as an evaluation value for each class, so double counting can be avoided. .
In addition, it is desirable to avoid double counting of individual evaluation values relating to patent data of joint applications in the same way in calculating composite evaluation information used in the technical complementation degree calculation process of S150 and the technical deepening degree calculation process of S250. Therefore, when the value obtained on the right side of (Equation 9) or (Equation 10) is used as it is as the individual evaluation value Pi, for example, the composite evaluation information is not obtained by adding the evaluation value for each class for each holder. It is possible to calculate by extracting the patent data 200 in which one or both of the target holder and the candidate holder are included in the holder and summing the individual evaluation values Pi of the extracted patent data for each class. desirable. As a result, if there is one patent data related to the joint application of the target holder and the candidate holder, for example, the individual evaluation value of the patent data is counted as one case in the composite evaluation information. Can be avoided. It should be noted that the composite evaluation information calculated in this way is preferably normalized by dividing the evaluation value for each class for each class, similarly to the standardization of the evaluation value for each class in S140 and S240.

  Then, when the processing of S131 is completed, the same processing as in FIG. 4 described above is performed, and processing for generating partner selection information is performed. Note that “S221, S222, and S231” are the same as the processes of “S121, S122, and S131”.

In the present modification 8, a patent score PS (details will be described later) obtained in consideration of the technical field, the application time, etc. is used as information used for the alliance analysis. The patent score PS is weighted according to the type of the progress information 240, which will be described later.
Therefore, according to the present modification 8, a more accurate evaluation value for each class for each holder is calculated, and therefore, by using the evaluation value for each class for each holder, an optimum technical partner can be selected. Will be able to.
Moreover, since the individual evaluation value of each patent data 200 is used, as described above, it is possible to avoid double counting of the patent data of the joint application by a relatively simple process.

Further, when the evaluation value for each class for each holder is calculated using the above (Equation 10), the influence of the number of patents for each holder is reduced as described above, and the “quality” of the patent data possessed by the holder is reduced. It can be reflected.
FIG. 18 shows the distribution of the evaluation value for each class using the patent score of the modified example 8 and the evaluation value for each class using the patent impact index and the progress information index of the above embodiment, and the number of publications. It is the figure shown in the relationship. Specifically, multiple classes obtained by clustering patent groups in a specific field are further classified for each holder, and the evaluation value for each class is calculated by multiplying the patent impact index and the historical information index. And standardized to mean 0 and variance 1. In addition, for each classification for each class of the same holder, the evaluation value for each class for each holder is calculated by counting the values below the threshold as 0 by the above (Equation 10) using the patent score. Standardized to 1. The standardized evaluation value for each class for each holder is taken on the vertical axis, and the number of publications of each classification for each class for each holder (also standardized) is taken on the horizontal axis.
As shown in the figure, the evaluation value for each class using the patent impact index and the progress information index of the above embodiment has a distribution close to a straight line indicating a direct proportional relationship with the number of publications. It is greatly influenced by the number of cases. On the other hand, the evaluation value for each class using the patent score of the modified example 8 is distributed in a region far away from a straight line indicating a direct proportional relationship although it is not irrelevant to the number of publications. It can be seen that the impact has been mitigated.

  Next, in the present modification 8, a patent score PS used for calculation of evaluation values for each class for each holder will be described.

The patent score PS is a score indicating the evaluation value of each patent data. The patent score PS is calculated using the content information 230 and the progress information 240 of the patent data.
The data structure of the content information 230 and the progress information 240 used for calculating the patent score PS is not particularly limited. In the present modification 8, the content information 230 and the progress information 240 shown in FIGS. 19 and 20 are used. Hereinafter, the data structure of the content information 230 and the progress information 240 will be described.

FIG. 19 is a diagram schematically illustrating an example of the data configuration of the content information used in Modification 8 of the present embodiment.
As shown in the figure, the content information 230 includes a field 2301 for registering “patent data ID (gazette number, etc.)”, a field 2302 for registering “number of claims” of the patent data, and “claims” A record 230 includes a field 2303 for registering “average number of characters” and a field 2304 for registering “number of specifications” of the patent data. The content information 230 includes a plurality of records.
Here, the “number of claims” is information indicating the number of claims of the patent application, and the “average number of characters of the claim” is the average number of characters (or the number of words) per claim of the patent application. It is information to show. The “number of specification pages” is information indicating the number of specification pages or publication pages of the patent application. Such information is extracted from published patent gazettes and other patent data of each patent application.

FIG. 20 is a diagram schematically illustrating an example of the data configuration of the progress information used in Modification 8 of the present embodiment.
As shown in the figure, the progress information 240 includes a field 2401 for registering “patent data ID (gazette number, etc.)”, a field 2402 for registering “elapsed days from the filing date”, and “examination request date”. A field 2403 for registering “elapsed days since”, a field 2404 for registering “elapsed days since registration”, a field 2405 for registering information indicating the presence / absence of “divisional application”, “ Register a field 2406 for registering information indicating the presence or absence of “early examination”, a field 2407 for registering information indicating the presence or absence of “trial decision of appeal”, and information indicating the presence or absence of “opposition to maintain opposition” Field 2408 for registering information, a field 2409 for registering information indicating the presence or absence of “invalidation trial maintenance decision”, and the presence of “priority claim” A field 2410 for registering information indicating the presence of “PCT application”, a field 2411 for registering information indicating the presence / absence of “PCT application”, a field 2412 for registering information indicating the presence / absence of “packaging browsing”, And a field 2413 for registering information indicating the “number of times cited” constitutes one record. The progress information 240 includes a plurality of records.

  Here, “elapsed days from application”, “elapsed days from examination request”, and “elapsed days from registration date” are information relating to the period of the corresponding patent data. “Elapsed days from application” is the application date, “Elapsed days from examination request” is the application examination request date, and “Elapsed days from registration date” is the evaluation date (calculation of patent score). The number of elapsed days up to a predetermined date close to the evaluation date is calculated and stored in the storage device 2. “Elapsed days from examination request” for a patent application that has not yet been requested for examination of application is NULL, and “elapsed days from registration date” for a patent application that has not yet been set and registered is NULL.

  Among the progress information 240, “divisional application”, “accelerated examination”, “approval trial decision”, “opposition maintenance decision”, “invalid trial maintenance decision”, “priority”, “PCT application”, “packaging” “Browsing” is information indicating the presence or absence of a predetermined action on the patent data. "Divisional application" is whether the divisional application has been filed based on the patent application, "Rapid examination" is whether the patent application has been expedited, and Whether an appeal against a decision to reject the application has been requested and whether a patent trial decision has been made in that trial, whether or not the opposition maintenance decision has been made, whether or not a patent opposition to the patent has been made and a maintenance decision has been made, The “invalidation trial maintenance decision” is whether the patent invalidation trial has been requested for the patent, and whether the appeal has been rejected in the trial, “priority” is the priority claim based on the previous patent application etc. “PCT application” is whether the patent application is an international application based on the Patent Cooperation Treaty, and “Bag browsing” is for the patent application. Based on whether list claims have been made, each given 1 for example if the predetermined action has been performed, if not been given a 0, for example.

Next, the patent score PS calculation process performed by the evaluation value calculation unit 120 will be described with reference to FIG.
FIG. 21 is a flowchart showing the procedure of the patent score calculation process of the modification 8 of the present embodiment.

First, the evaluation value calculation unit 120 receives input of an IPC code from the user, and acquires patent data 200 (electronic data indicating a patent publication) (S400).
Specifically, when receiving the input of the IPC code from the user, the evaluation value calculation unit 120 accesses the storage device 2 and acquires the patent data 200 classified into the IPC code. The patent data acquired here is the analysis target population for calculating the patent score. Note that the bibliographic information 220 of the patent data 200 includes information on the filing date of the patent application, information on priority dates (only when priority is claimed), and the like.

Next, the evaluation value calculation unit 120 uses the application date information or the priority date information among the bibliographic information of the acquired patent data 200 to obtain the analysis target population patent data 200 for each predetermined period (this modification). In FIG. 8, the data is classified into groups t by application year, year to which the priority date belongs, etc. (S500).
Next, the evaluation value calculation unit 120 calculates the evaluation value of each patent data 200 (S600). Details of this processing will be described with reference to FIG.

  FIG. 22 is a flowchart showing details of processing for calculating a patent score of patent data.

The evaluation value calculation unit 120 acquires the progress information 240 and the content information 230 for the patent data 200 belonging to the group t classified in S500 (S610).
In FIG. 22, it is assumed that the acquired one group is composed of J patent data, and a subscript j (j = 1, 2,..., J) is used to distinguish each of the J cases.
When the evaluation value calculation unit 120 acquires the J patent data 200, the evaluation value calculation unit 120 uses the progress information 240 and the content information 230 of the J patent data, and uses “evaluation item corresponding presence / absence data” in S6302 to S6304 to be described later. “Total value for J cases” and the like are obtained in advance.

  Next, the evaluation value calculation unit 120 sets the variable j to 1 (S620), and calculates the evaluation raw score of the patent data j as follows.

  The evaluation value calculation unit 120 uses information registered in each field of the progress information 240 as an evaluation item, and I evaluation items i (i = 1, 2,..., I) are previously determined for each evaluation item. The set evaluation point calculation method is selected (S6301).

  There are the following three evaluation point calculation methods in Modification 8 of the present embodiment. That is, for information registered in the fields 2405, 2406, 2407, 2408, 2409, 2410, 2411, and 2412, S6302 [presence / absence type] is selected as information indicating the presence / absence of a predetermined action on the patent data. In the fields 2402, 2403, and 2404, S6303 [Time decay type] is selected as information related to the period of the patent data. In the field 2413, S6304 [number type] is selected as information indicating the number of times the patent data is cited.

  When the evaluation score calculation method is selected, the evaluation score of the patent data j is calculated for each of the I evaluation items i (S6302, S6303, S6304).

<Presence type>
In S6302, the evaluation value calculation unit 120 calculates an evaluation score from the following [Equation 13] for the evaluation item i for which [presence / absence type] is selected.

  Here, the “existence / non-existence data of the evaluation item i” arranged in the numerator is, for example, “1” if the divisional application has been filed as described above, and “0” if it has not been made.

  In the denominator, the positive square root of the in-group total value of the above “existence data of evaluation item i” is arranged. Accordingly, when there are a large number of patent data 200 corresponding to the evaluation items in the group, the denominator is large, and when there are only a small number of patent data 200 corresponding to the evaluation items in the group, the denominator is small. Patents with fewer evaluation items (such as “Invalidation Trial Maintenance Decision”) than patents with a higher number of evaluation items (such as “Bag Viewing”) will be maintained after patent registration (In general, a high maintenance rate is considered to indicate a high economic value commensurate with the maintenance cost (patent fee)), and thus each evaluation item is automatically weighted. In addition, since it is tabulated in groups for each predetermined period, for example, older patents have more progress information added, and new patents that have just been published often do not yet have progress information added. It can alleviate the tendency of new patents to be given low ratings for that reason.

<Time decay type>
In S6303, the evaluation value calculation unit 120 calculates an evaluation score according to the following [Equation 14] for the evaluation item i for which [Time decay type] is selected.

  “Exp (− (Min (elapsed time, year limit)) / year limit)” placed in the numerator here is the “number of days elapsed since the request for examination”. ”Or“ year ”, which is the value obtained by dividing the smaller value by“ year ”and multiplying by −1 and the power of the Napier number e. The “year” is the maximum number of years from the filing date until the expiration of the patent right (20 years under the current Japanese law). The same formula is used for “elapsed days from registration date”, and “year” is the maximum number of years from the filing date to the expiration of the patent term (20 years under the current Japanese law). The same formula is used for the “elapsed days from the filing date”, but the “year” is the number of years from the filing date to the application examination request deadline (3 years in the current Japanese law). According to this, while the elapsed time is short, the value of the numerator is a value close to Exp (0) = 1. However, when the elapsed time ≧ years, the value of Exp (−1) = 1 / e. To fall. The advantage of using an exponential function is that a depreciation effect on the value can be introduced and that the evaluation value distribution can be eliminated and a smooth distribution can be achieved. “Elapsed days from request for examination”, “Elapsed days from application date”, and “Elapsed days from registration date” are basic evaluation items applicable to many patents. Tying can be avoided.

  In the denominator of [Formula 14], the same formula as the [presence / absence type] used in the above S6302 is arranged, but regarding “the number of days elapsed from the request for examination”, the application examination request for the patent application has not been requested. For example, the value is 1, and if not, for example, 0 is summed up in the group to obtain a positive square root. For the “elapsed days from the date of registration”, the denominator is a value obtained by adding a value of 1 within the group by taking the positive square root by adding 1 if the patent application has been registered for patent right setting and not being registered. . As for “elapsed days since filing”, all patent data is applicable, so if the evaluation data of the evaluation item is 1, the value of the denominator is equal to the positive square root of the number of patent data 200 in the group. Become. In any case, the denominator is large when there are a large number of patent data 200 corresponding to the evaluation items in the group, and the denominator is small when there are only a few patent data 200 corresponding to the evaluation items in the group. As described above, “Elapsed days from request for examination”, “Elapsed days from application date”, and “Elapsed days from registration date” are basic evaluation items applicable to many patents. Tends to be small.

The evaluation score calculated in [time decay type] in S6303 is further corrected by the content information 230 (content information in FIG. 17).
When evaluating only with the progress information 240, there is a possibility that the patent application or patent right shortly after the application is published or the patent right setting registration does not have the progress information 240 expected to be granted in the future and the evaluation cannot be performed correctly. is there. Therefore, in order to correct this, content information is added to the evaluation based on the progress information 240. However, the content information 230 tends not to have a high correlation with the maintenance rate as the progress information 240. If the content information 230 is inadvertently included, the accuracy of the evaluation may decrease.
Therefore, in order to effectively reflect the content information 230 in the evaluation of a patent for which the progress information 240 is insufficient, the influence of the content information 230 is kept small in the evaluation of the patent to which the progress information 240 is sufficiently given. Only the evaluation score calculated by “attenuation type” is multiplied by the correction coefficient based on the content information 230.
As described above, according to the present modification 8, the content information 230 of each patent data 200 is added to the information related to the period having characteristics that are easily given to any patent data 200 regardless of whether the application is old or new. be able to. As a result, it is possible to appropriately evaluate patent data consisting of a new application to which the progress information 240 is not given much.

Specifically, for each evaluation point in [Expression 14],
a ₁ × a ₂ × a ₃
here,
a ₁ = 2 ^1/3 (when the average number of characters per claim is below average) or 2 ^-1/3 (when the average number of characters per claim is above average)
a ₂ = 2 ^1/3 (when the total number of pages is above average) or 2 ^-1/3 (when the total number of pages is below average)
a ₃ = 2 ^1/3 (when the number of claims is within an average value ± 1 standard deviation) or 2 ^−1/3 (when the number of claims is outside the above range)
Multiply By setting the maximum values of a ₁ , a ₂ , and a ₃ to 2 ^1/3 , the correction is limited to a ₁ × a ₂ × a ₃ as the maximum value. In the present modification 8, the value of a ₁ × a ₂ × a ₃ is set to 2 at the maximum.

<Number of times type>
In S6304, the evaluation value calculation unit 120 calculates an evaluation score by the following [Equation 15] for the evaluation item i for which [Number-of-times] is selected.

Here, “f (quotation) × log (n _j +1)” arranged in the numerator is the weight of the logarithm of the value obtained by adding 1 to the “cited count n _j ” for the “cited count”. Quoting). According to the verification by the present inventors, it has been found that the maintenance rate of the patent right changes depending on the number of citations as well as the presence or absence of citations. Since the increase gradually shows a tendency to peak, the logarithm is taken.

In the denominator, the positive square root of the total value in the group of “f (quotation) × log (n _j +1)” is arranged. Accordingly, when there are many patent data cited in other applications in the group, the denominator is large, and when there are only a few patent data 200 cited in other applications in the group, the denominator is small. .

具体的な重みとしては、他社引用の場合のｆ（引用_other）と、自社引用の場合のｆ（引用_self）との比を、１：２とした。In the numerator and denominator of the Equation 15, the weight f (reference) any may be used a positive number, cited times (third reference count) patent application company n _Jother and other patents of its It is distinguished from the number of times cited in the application (in-house citation number) n _jself, and a different weight is _assigned to each logarithm. In this case, instead of the above [Equation 15], the following [Equation 16] is used.

As a specific weight, the ratio of f (quoting _other ) in the case of _other company citations and f (quoting _self ) in the case of company citations was set to 1: 2.

The number of times cited is highly correlated with the value of a patent. Furthermore, according to the verification by the present inventors, the number of times cited in the examination of patent applications of other companies (citation of other companies) and the number of times cited (in-house quotation) in examinations of other patent applications of the company are the latter. Was found to be significantly higher in correlation with patent value. The invention cited in the examination of other patent applications of the company is presumed to be due to the fact that it is often the basic invention that is the core in the implementation technology of the company. And while recognizing that the company has already applied for such a basic invention, there is a high possibility that the company has applied for the improved technology and built a strong patent portfolio.
According to the present modification 8, it is possible to appropriately evaluate a patent application or a patent right by dividing the number of citations into other company citations and company citations and reflecting the latter number in the evaluation value. .

この式に示されるように、評価素点は、Ｉ個の評価点の二乗和の正の平方根、又は０となる。評価素点が０となるのは、審査請求期限までに出願審査請求しなかった場合、出願を取下げ又は放棄した場合、拒絶査定が確定した場合、その他特許出願が失効した場合と、異議申立による取消決定や無効審判による無効審決が確定した場合、特許権を放棄した場合、特許権の存続期間が満了した場合、その他の特許権が消滅した場合である。これらの情報も各特許データの経過情報から読み取り、該当する場合は評価素点を０とする。
なお、上述したように、Ｓ６３０３において、〔時間減衰型〕で算出された評価点に対しては、内容情報２３０による補正を行う。When the evaluation score of the patent data j is calculated for all the evaluation items i (i = 1, 2,..., I), the evaluation value calculation unit 120 evaluates the evaluation score of the patent data j based on this. Is calculated by the following [Equation 17] (S640).

As shown in this equation, the evaluation raw score is a positive square root of the sum of squares of I evaluation points, or 0. The evaluation score is 0 because the application request is not requested by the deadline for requesting examination, the application is withdrawn or abandoned, the decision of refusal is finalized, other patent applications have expired, The case where the decision of revocation or the trial for invalidation by the trial for invalidation is finalized, the patent right is abandoned, the duration of the patent right expires, or the other patent right is extinguished. These pieces of information are also read from the progress information of each patent data, and the evaluation raw score is set to 0 when applicable.
As described above, in step S6303, the evaluation score calculated by the [time decay type] is corrected by the content information 230.

As described above, the reason why the patent evaluation raw score is obtained by [Equation 17] in S640 is as follows.
As a method of calculating an evaluation raw score from an evaluation point i based on a plurality of evaluation items, there is a method of calculating a sum of each evaluation point i (simple sum method). However, according to this calculation method, since the evaluation of a patent to which a lot of historical information having a correlation with the patent maintenance rate (economic value) is given is calculated high, the sum of the evaluation points i should be used as an evaluation raw score. Is reasonable at first glance, but the evaluation score of a patent that has been granted a lot of historical information that does not have a high correlation with the maintenance rate (a lot of low evaluation points are added) has a very high correlation with the maintenance rate Care should be taken because it may exceed the evaluation score of a patent to which a small amount of progress information is granted.
As one method for solving this problem, there is a method in which the maximum value among the evaluation points i is used as an evaluation raw score (maximum value method). However, according to this calculation method, especially when investigating the correlation between certain historical information and the maintenance rate of a group of patents, when investigating the correlation regardless of what other historical information is given, maintaining a certain patent Since the rate is expected to be best expressed by the maintenance rate of the historical information having the highest maintenance rate, it is reasonable to use the maximum value of the evaluation point i as an evaluation raw score. If the maximum value is the same in the two patents, no superiority or inferiority is given. Furthermore, when the maximum value method is used, it is not possible to carry out an evaluation that takes into account the viewpoints of three different entities of the applicant, the JPO, and competitors, and only the viewpoints of any one of those entities The viewpoint of the remaining subject cannot be reflected in the evaluation of patent data.
The above-described method for taking the square root of the sum of squares can be said to be a method that combines the advantages of the simple sum method and the maximum value method. That is, by taking the square root of the sum of squares, when there is a high evaluation point i in I evaluation items i related to a certain patent data j, the high evaluation point i greatly affects the evaluation raw score. The evaluation points other than the evaluation item having a high evaluation point i are also evaluation raw points that are somewhat considered. Therefore, a high evaluation score is given to patent data j that corresponds to multiple items such as “early examination”, “opposition to maintain opposition”, and “invalidation maintenance decision” that tend to be high. be able to.
Thus, in this modification 8, patent evaluation is performed in consideration of all evaluation points calculated according to the type of patent attribute information (S630, S640). As a result, it is possible to evaluate the value of patent data from multiple aspects.

  When the evaluation score is calculated, the evaluation value calculation unit 120 calculates the logarithm thereof to obtain the evaluation value of the patent data j (S650).

Next, it is determined whether or not evaluation values have been calculated for all patent data j (S660). If not calculated (S660: NO), the process proceeds to S670, the variable j is set to j + 1, and the process returns to S630. The evaluation value is calculated for the following patent data.
When the evaluation values are calculated for all the patent data j (S660: YES), the evaluation value calculation processing for the patent data 200 belonging to the group ends.
As described above, in the present modification 8, a plurality of patent data having different characteristics are evaluated in consideration of the characteristics for each technical field and each filing time. As a result, the value of patent data can be more appropriately evaluated.

  The evaluation value calculation processing from S610 to S670 is executed for all the groups t obtained by classifying the patent data acquired in S400 in S500. When the evaluation values are calculated for all the groups t, the process returns to S700 in FIG.

  In S700, the evaluation value calculation unit 120 calculates the deviation value in the analysis target population acquired in S400 as the patent score PS based on the evaluation value obtained in the above processing (S700). The evaluation value calculation unit 120 stores the calculated patent score PS in the memory or the storage device 2 of the information processing apparatus 1 (the patent score PS is stored in association with a patent data ID (patent publication number, etc.)). .

  Thus, in the present modification 8, the analysis target population is classified into groups for each technical field and for each period, and the patent score is calculated using the value obtained for each classified group as the denominator. . Therefore, according to the present modification 8, an appropriate relative evaluation can be performed within an analysis target population including patent applications or patent rights at various times in various technical fields.

Claims (18)

An information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Means for classifying each acquired patent data into classes for each predetermined technical field;
Using the attribute information of patent data belonging to each class, the evaluation value for each class is calculated for each class, and the evaluation value for each class and for each holder of patent data belonging to the class is calculated. Means for calculating;
A means for creating an overall distribution tendency indicating a distribution tendency of the evaluation value for each class, and for each holder, creating nominee evaluation information that associates an evaluation value for each class of the holder of the holder,
Means for accepting designation of a subject holder who is a subject of analysis;
Using the obtained overall distribution tendency, and each created right holder evaluation information, the partner selection information generating means for generating information for selecting a partner of the received target holder,
The partner selection information generating means is
For each candidate holder, composite candidate evaluation information obtained by adding the holder evaluation information of the candidate holder who is each holder other than the target holder and the holder evaluation information of the target holder for each class is obtained. Determining a composite distribution tendency indicating a distribution tendency of the value for each class for each of the determined composite evaluation information, and determining a composite approximation indicating a degree of approximation of each of the determined composite distribution trends with respect to the overall distribution trend. A characteristic information processing apparatus. The information processing apparatus according to claim 1,
The partner selection information generating means is
Obtaining a holder distribution tendency indicating the distribution tendency of the value of each class of the holder evaluation information of the target holder, and a target holder approximation degree indicating the degree of approximation of the obtained holder distribution trend to the overall distribution trend Obtaining the selected synthetic approximation from the obtained synthetic approximations, specifying the synthetic evaluation information having the selected synthetic approximation, and using the nominal used for calculating the specified synthetic evaluation information An information processing apparatus that selects a candidate holder corresponding to human evaluation information as a recommended partner. An information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Means for classifying each acquired patent data into classes for each predetermined technical field;
Using the attribute information of patent data belonging to each class, the evaluation value for each class is calculated for each class, and the evaluation value for each class and for each holder of patent data belonging to the class is calculated. Means for calculating;
Means for generating a complementary reference vector having the evaluation value for each class as an element, and generating a nominal person evaluation vector having an evaluation value for each class of the holder for each holder as an element;
Means for accepting designation of a subject holder who is a subject of analysis;
Using the generated complementary reference vector, and each generated right holder evaluation vector, the partner selection information generating means for generating information for selecting a partner of the received target holder,
The partner selection information generating means is
A combined evaluation vector obtained by adding a candidate evaluation vector of a candidate holder who is each holder other than the target holder and a holder evaluation vector of the target holder is obtained for each candidate holder, and the combined evaluation An information processing apparatus that calculates a composite cosine that is a cosine of the composite evaluation vector and the complementary reference vector for each vector. The information processing apparatus according to claim 3,
The partner selection information generating means is
A target holder cosine that is a cosine of the target holder evaluation vector of the target holder and the complementary reference vector is calculated, a higher one than the target holder cosine is selected from the composite cosines, and the selected composite cosine is selected. An information processing apparatus characterized by specifying a composite evaluation vector used for calculating the candidate evaluation person and selecting a candidate holder corresponding to the nominee evaluation vector used for calculating the specified composite evaluation vector as a recommended partner. An information processing apparatus according to any one of claims 1 to 4,
The evaluation value for each class is a value that matches the value obtained by totaling the evaluation values for each class in the corresponding class,
The evaluation value for each class for each holder is a value normalized using the evaluation value for each class. An information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Means for classifying each acquired patent data into classes for each predetermined technical field;
Means for calculating an evaluation value for each class and for each holder of patent data belonging to the class, using attribute information of the patent data belonging to each class;
Means for accepting designation of a subject holder who is a subject of analysis;
Using the calculated evaluation value, having a recommended partner selection means for selecting a partner for each class of the received target holder,
The recommended partner selection means is:
For at least one of the classified classes, a total value obtained by adding an evaluation value in the class of a candidate holder who is each holder other than the target holder and an evaluation value in the class of the target holder For each candidate holder, and selecting the candidate holder determined using the total value as a recommended partner in the class. An information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Means for classifying each acquired patent data into classes for each predetermined technical field;
Using the attribute information of patent data belonging to each class, the evaluation value for each class is calculated for each class, and the evaluation value for each class and for each holder of patent data belonging to the class is calculated. Means for calculating a standardized evaluation value for each class name holder, wherein the evaluation value for each class of the standard name is normalized using the evaluation value for each class;
Using the evaluation value for each class for each standardized holder, means for creating, for each holder, nominee evaluation information that correlates the evaluation value for each class for each standardized name of the holder;
Means for accepting designation of a subject holder who is a subject of analysis;
Using each created right holder evaluation information, and having a partner selection information generating means for generating information for selecting a partner of the received target holder,
The partner selection information generating means is
For each candidate holder, composite evaluation information is obtained by adding, for each class, the holder evaluation information of the target holder and the holder evaluation information of the candidate holder who is each holder other than the target holder. , Extracting a fluctuating value from the corresponding class value of the target person's holder evaluation information from the values of each class of the composite evaluation information, and specifying the class to which the maximum value of the extracted values belongs An information processing apparatus characterized by: An information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Means for classifying each acquired patent data into classes for each predetermined technical field;
Using the attribute information of patent data belonging to each class, the evaluation value for each class is calculated for each class, and the evaluation value for each class and for each holder of patent data belonging to the class is calculated. Means for calculating a standardized evaluation value for each class name holder, wherein the evaluation value for each class of the standard name is normalized using the evaluation value for each class;
Using the evaluation value for each standardized holder for each class, for each class, generate information that associates the evaluation value for each standardized holder for each standardized holder and visualizes the generated information An information processing apparatus comprising: means for processing and outputting information that can be output. A method of generating a plurality of patent data and generating information for selecting a partner to be performed by an information processing apparatus that performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
The information processing apparatus includes:
Classifying each acquired patent data into classes for each predetermined technical field;
Using the attribute information of patent data belonging to each class, the evaluation value for each class is calculated for each class, and the evaluation value for each class and for each holder of patent data belonging to the class is calculated. A calculating step;
Obtaining an overall distribution tendency indicating a distribution tendency of the evaluation value for each class, and creating, for each holder, nominee evaluation information that associates an evaluation value for each class of the holder of the holder;
Receiving the designation of the subject owner who is the subject of the analysis;
Generating information for selecting a partner of the received target holder, using the obtained overall distribution tendency, and each created holder evaluation information,
The step of generating information for selecting the partner is as follows:
For each candidate holder, composite candidate evaluation information obtained by adding the holder evaluation information of the candidate holder who is each holder other than the target holder and the holder evaluation information of the target holder for each class is obtained. Determining a composite distribution tendency indicating a distribution tendency of the value for each class for each of the determined composite evaluation information, and determining a composite approximation indicating a degree of approximation of each of the determined composite distribution trends with respect to the overall distribution trend. A method of generating information for selecting a feature partner. A program for acquiring a plurality of patent data and generating information for selecting a partner in an information processing apparatus that performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
The program is
A process of classifying each acquired patent data into a class for each predetermined technical field;
Using the attribute information of patent data belonging to each class, the evaluation value for each class is calculated for each class, and the evaluation value for each class and for each holder of patent data belonging to the class is calculated. Processing to calculate,
A process of obtaining a general distribution tendency indicating a distribution tendency of the evaluation value for each class, and creating, for each holder, nominee evaluation information in which an evaluation value for each class of the holder of the holder is associated.
A process of accepting the designation of the subject owner who is the subject of analysis,
Using the obtained overall distribution tendency and the created individual holder evaluation information, the information processing apparatus is caused to execute processing for generating information for selecting a partner of the received target holder,
The process of generating information for selecting the partner is as follows:
For each candidate holder, composite candidate evaluation information obtained by adding the holder evaluation information of the candidate holder who is each holder other than the target holder and the holder evaluation information of the target holder for each class is obtained. Determining a composite distribution tendency indicating the distribution tendency of the value for each class for each of the determined composite evaluation information, and determining a composite approximation indicating the degree of approximation of each of the determined composite distribution trends with respect to the overall distribution trend. A featured program. An information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Means for classifying each acquired patent data into classes for each predetermined technical field;
Means for accepting designation of a subject holder who is a subject of analysis;
Using the attribute information of the patent data belonging to each class, the evaluation value for each class is calculated for each class, and among the patent data belonging to each class, the candidate nominee who is each nominee other than the subject nominee Means for calculating a composite evaluation value for each class and for each candidate holder, using the attribute information of the patent data and the attribute information of the patent data of the target holder,
A means for obtaining an overall distribution tendency indicating a distribution tendency of the evaluation value for each class, and for each candidate holder, obtaining a composite distribution tendency indicating a distribution tendency of the composite evaluation value for each class;
Using the obtained overall distribution tendency, and the obtained combined distribution tendency, the partner selection information generating means for generating information for selecting the partner of the received target holder,
The partner selection information generating means is
An information processing apparatus that obtains a composite approximation degree indicating an approximation degree of each of the obtained composite distribution trends with respect to the overall distribution tendency. An information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Means for classifying each acquired patent data into classes for each predetermined technical field;
Means for accepting designation of a subject holder who is a subject of analysis;
Using the attribute information of the patent data belonging to each class, the evaluation value for each class is calculated for each class, and among the patent data belonging to each class, the candidate nominee who is each nominee other than the subject nominee Means for calculating a composite evaluation value for each class and for each candidate holder, using the attribute information of the patent data and the attribute information of the patent data of the target holder,
Generating a complementary reference vector having the evaluation value for each class as an element, and generating, for each candidate holder, a composite evaluation vector having a composite evaluation value for each class of the candidate holder as an element;
Using the generated complementary reference vector and the generated composite evaluation vector, the partner selection information generating means for generating information for selecting a partner of the received target holder,
The partner selection information generating means is
An information processing apparatus that calculates a composite cosine that is a cosine of the composite evaluation vector and the complementary reference vector for each composite evaluation vector. An information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Means for classifying each acquired patent data into classes for each predetermined technical field;
Means for accepting designation of a subject holder who is a subject of analysis;
For at least one of the classified classes, among the patent data belonging to the class, the attribute information of the patent data of the candidate holder who is each holder other than the target holder and the attributes of the patent data of the target holder Means for using the information to calculate a composite evaluation value for each candidate holder;
An information processing apparatus comprising: a recommended partner selection unit that selects a candidate holder obtained using the calculated composite evaluation value as a partner for each class of the received target holder. An information processing apparatus that acquires a plurality of patent data and performs various information processing using the acquired patent data,
The patent data includes information indicating the owner of the patent data and attribute information of the patent data.
Means for classifying each acquired patent data into classes for each predetermined technical field;
Means for accepting designation of a subject holder who is a subject of analysis;
Using the attribute information of the patent data belonging to each class, for each class, to calculate the evaluation value for each class and the evaluation value for each class of the subject holder, among the patent data belonging to each class, Using the attribute information of the patent data of the candidate holder who is each holder other than the target holder and the attribute information of the patent data of the target holder, the information is synthesized for each class and for each candidate holder. An evaluation value is calculated, and further, a standardized evaluation value for each class name and a standardized synthesis evaluation value obtained by standardizing the calculated evaluation value for each class name and the combined evaluation value using the evaluation value for each class. Means for calculating;
Using the evaluation value for each standardized holder for each class, creating the holder evaluation information for the target holder, and using the standardized composite evaluation value, for each candidate holder, the candidate holder's Means for creating synthesis evaluation information in association with standardized synthesis evaluation values;
Using the created holder evaluation information and composite evaluation information, the partner selection information generating means for generating information for selecting the partner of the received target holder,
The partner selection information generating means is
From the values of each class of the composite evaluation information, a value that is fluctuating from the value of the corresponding class of the target person's holder evaluation information is extracted, and the class to which the maximum value of the extracted values belongs is specified. An information processing apparatus characterized by that. The information processing apparatus according to any one of claims 11 to 14,
For each acquired patent data, means for acquiring a patent score that individually evaluates the value of the patent related to the patent data;
For each of the candidate holders, the patent possessed by the candidate holder and the target holder using the patent score of the patent data in which at least one of the candidate holder and the target holder is the holder. An information processing apparatus comprising: a composite evaluation value calculating unit that calculates a composite evaluation value indicating a technical evaluation when combining the two. The information processing apparatus according to claim 15,
The composite evaluation value calculation means includes
For each of the classes, a patent score that is equal to or higher than a predetermined threshold is selected from among the patent scores of patent data belonging to the class, and a value obtained by totaling the selected patent scores is calculated as the composite evaluation value. Information processing apparatus. The information processing apparatus according to claim 16,
The information processing apparatus according to claim 1, wherein the patent score is a value standardized in a patent group of a population including a class for which the composite evaluation value is calculated. An information processing apparatus according to any one of claims 15 to 17,
The patent score means that the patent data is classified into groups for each technical field and every predetermined period, and for each classified group, the progress information of the patent data belonging to the group is used, An information processing apparatus characterized by being a calculated value.

Images