Information processing apparatus, information processing method, and program

Abstract

To allow the type of an individual or a group to be obtained using document information.SOLUTION: An information processing apparatus 1 is configured to obtain the type of an individual or a group using document information, the information processing apparatus including: a type determination unit 134 which acquires a thesis code concentration degree based on the maximum value of the number of codes that appear the most in three kinds of codes, such as a thesis code, a patent code, a competitive funds code, and thesis codes paired with two or more pieces of document information that are paired with one or more attribute values, acquires a patent code concentration degree based on the maximum value of the number of codes that appear the most in patent codes paired with two or more pieces of document information, acquires a competitive funds code concentration degree based on the maximum value of the number of codes that appear the most in competitive funds codes paired with two or more pieces of document information, and determines the type corresponding to a thesis set which is two or more pieces of document information, using the thesis code concentration degree, the patent code concentration degree, and the competitive funds code concentration degree; and a type output unit 141 which outputs the type.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing device and the like that determines the type of author, inventor, creator, or the like of a document corresponding to two or more pieces of document information. [Background technology]

[0002] In recent years, innovation creation from universities, starting from interdisciplinary research and industry-academia collaborative research, has become an important mission for reviving Japan's industrial power. A number of university researchers have begun to take the excellent research results that are the starting point of this research and turn them into startups. However, on the other hand, there are also many researchers who remain in academia and devote themselves to research. These differences in research stances are thought to be due to the individual personalities of each researcher, and Stokes (1997) classifies "researchers who emphasize the pursuit of truth" as the Bohr type, named after the founder of quantum mechanics, "researchers who emphasize applications" as the Edison type, named after the genius inventor, and "researchers who emphasize both" as the Pasteur type, named after the founder of bacteriology and inventor of vaccines (see Figure 35). In the following, these types of researchers will be referred to as "Stokes' BEP types" or simply "BEP types" as appropriate. In addition, Takada Jin (2020) states in “Considerations on the relationship between Pasteur-type researchers and university-based ventures” (Non-Patent Document 1) that Pasteur-type researchers play an important role in turning academic achievements into innovation.

[0003] However, the methods for determining BEP types generally involve time-consuming interviews with the parties involved or subjective surveys, and although quantitative methods using citation data on both papers and patents have also been proposed, Japanese university researchers originally filed small numbers of patent applications, so these methods have only been applicable to a limited extent.

[0004] Therefore, although, as Takada indicates, it is an important factor in innovation, an objective and comprehensive method for determining BEP types has yet to be found.

[0005] Furthermore, there is a technique for associating academic papers with the International Patent Classification (IPC) (see, for example, Non-Patent Document 2). Similarly, there is a technique for associating academic papers with Grant-in-Aid for Scientific Research classifications. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] Jin Takada, "Considerations on the relationship between 'Pasteur-type' researchers and university-based ventures" (Research and Technology Planning, Vol. 35, No. 3, 2020) [Non-Patent Document 2] Hidetsugu Nanba and 3 others, "Automatic classification of academic papers into the International Patent Classification", [online], Japio 2008 YEAR BOOK, [Retrieved October 4, 2019], Internet [URL:http: / / www.japio.or.jp / 00yearbook / files / 2008book / 08_4_04.pdf#search=%27%E9%9B%A3%E6%B3%A2%E8%8B%B1%E5%97%A3+%E8%AB%96%E6%96%87+%EF%BC%A9%EF%BC%B0%EF%BC%A3%E4%BB%98%E4%B8%8E%27] Summary of the Invention [Problem to be solved by the invention]

[0007] However, in the prior art, the type of an individual or group cannot be obtained using bibliographic information. [Means for solving the problem]

[0008] The information processing device of the first invention is an information processing device that acquires a paper code concentration degree based on αmax, which is the maximum number of codes that appear most frequently among the paper codes that are paired with two or more pieces of literature information and are paired with one or more attribute values ​​related to the literature information, and acquires a patent code concentration degree based on βmax, which is the maximum number of codes that appear most frequently among the patent codes that are paired with two or more pieces of literature information, and acquires a competitive funding code concentration degree based on γmax, which is the maximum number of codes that appear most frequently among the competitive funding codes that are paired with two or more pieces of literature information, and is equipped with a type determination unit that determines a type corresponding to a paper set that is two or more pieces of literature information using the paper code concentration degree, the patent code concentration degree, and the competitive funding code concentration degree, and a type output unit that outputs the type.

[0009] With this configuration, the type of individual or group who is the author, inventor, creator, etc. of the document information can be obtained by using three types of codes paired with the document information.

[0010] Moreover, the information processing device of the second invention, compared to the first invention, further comprises a coordinate storage unit which acquires, for each of two or more pieces of literature information, coordinate information (x, y, z) in which a value corresponding to the paper code paired with the literature information is x, a value corresponding to the patent code paired with the literature information is y, and a value corresponding to the competitive funding code paired with the literature information, and which associates the coordinate information (x, y, z) with each of the two or more pieces of literature information and accumulates the coordinate information (x, y, z) in the coordinate storage unit, and the type determination unit acquires from the coordinate storage unit one or more pieces of coordinate information corresponding to each of the two or more pieces of literature information, and the number f(x, y, z) of each of the one or more pieces of coordinate information acquired by the coordinate acquisition unit, and calculates αmax, which is the maximum value of ∫∫f(x, y, z)dydz≡α(x), as the paper code concentration. the maximum value of ∫∫f(x,y,z)dxdz≡β(y) is the patent code concentration degree, and the maximum value of ∫∫f(x,y,z)dydz≡γ(z) is the competitive funding code concentration degree, and quasi-Stokes type determination means for determining the type to be quasi-Bohr type if MAX(paper code concentration degree, patent code concentration degree, competitive funding code concentration degree)=paper code concentration degree, the type to be quasi-Edison type if MAX(paper code concentration degree, patent code concentration degree, competitive funding code concentration degree)=patent code concentration degree, and the type to be quasi-Pasteur type if MAX(paper code concentration degree, patent code concentration degree, competitive funding code concentration degree)=competitive funding code concentration degree. It is estimated that even if Stokes' BEP type and the "quasi" BEP type of the present invention are considered to be equivalent, the validity will not be lost for the following reasons. That is, the highest paper code concentration means that the researcher is devoting efforts to a specific classification of papers, as shown in Figure 11(a), and therefore the research stance is considered to emphasize the search for truth. Also, the highest patent code concentration means that the researcher is devoting efforts to a specific classification of patents, as shown in Figure 11(b), and therefore the research stance is considered to emphasize the use. Furthermore, the highest competitive fund code concentration means that the researcher is widely distributed in a horizontal plane rather than a vertical plane, as shown in Figure 11(c), and therefore the research stance is considered to emphasize both the search for truth and the use. However, the word "quasi" is added to each type because it cannot be proven that it is completely consistent with Stokes' BEP type.

[0011] With this configuration, the type of individual or group who is the author, inventor, creator, etc. of the document information can be appropriately obtained by using three types of codes paired with the document information.

[0012] The information processing device of the third invention, compared to the second invention, further comprises a coordinate storage unit for acquiring, for each of the one or more document information out of the two or more document information, first coordinate information which is coordinate information using only the first classification code of the three types of codes and one or more sub coordinate information which is coordinate information using at least one sub coordinate code, for each of the one or more document information out of the two or more document information, and for storing the first coordinate information and the one or more sub coordinate information in the coordinate storage unit in association with each of the one or more document information, and the type determination unit further comprises a coordinate acquisition means for acquiring two or more coordinate information which are the first coordinate information and the one or more sub coordinate information corresponding to each of the one or more document information, and a code type not associated with a score, for each of the one or more document information, and a degree acquisition means for acquiring a total number of top classification codes and a total number of sub-classification codes corresponding to the codes, acquiring a code concentration degree based on both totals, determining a minimum score to be adopted so that a difference between the total number of codes of types with corresponding scores and the total number of codes of types with no corresponding scores satisfies an allowable condition, acquiring the total number of top classification codes and the total number of sub-classification codes using only coordinate information corresponding to codes with corresponding scores equal to or greater than the minimum value, and acquiring the code concentration degree based on both totals; and a quasi-Stokes type determination means for determining the type to be a quasi-Bohr type if MAX(paper code concentration degree, patent code concentration degree, competitive funding code concentration degree)=paper code concentration degree, the type to be a quasi-Edison type if MAX(paper code concentration degree, patent code concentration degree, competitive funding code concentration degree)=patent code concentration degree, and the type to be a quasi-Pasteur type if MAX(paper code concentration degree, patent code concentration degree, competitive funding code concentration degree)=competitive funding code concentration degree.

[0013] With this configuration, the type of individual or group who is the author, inventor, creator, etc. of the document information can be appropriately obtained by using three types of codes paired with the document information.

[0014] Moreover, the information processing device of the fourth invention, compared to the second invention, further comprises a condition receiving unit for receiving selection conditions which are conditions related to one or more attribute values ​​related to literature information and which specify one individual or one group, and a coordinate storage unit for acquiring, for each of two or more pieces of literature information, coordinate information (x, y, z) in which a value corresponding to a paper code paired with the literature information is x, a value corresponding to a patent code paired with the literature information is y, and a value corresponding to a competitive fund code paired with the literature information is z, and storing the coordinate information (x, y, z) in a coordinate storage unit in association with each of the two or more pieces of literature information, and the type determination unit further comprises a coordinate acquisition means for acquiring, from the coordinate storage unit, one or more pieces of coordinate information corresponding to each of two or more pieces of literature information of an individual or one group which satisfy the selection conditions, and a MAX(α) of the maximum share rate of the paper code, patent code, and competitive fund code of each of two or more individuals or each of two or more groups including the individual or one group. and a degree acquisition means for calculating a standard deviation of the maximum share rate of each of the three codes of an individual or a group using MAX(paper code concentration, patent code concentration, competitive funding code concentration)=paper code concentration, determining the type to be a quasi-Bohr type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=paper code concentration, determining the type to be a quasi-Edison type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=patent code concentration, and determining the type to be a quasi-Pasteur type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=competitive funding code concentration.

[0015] With this configuration, the type of individual or group who is the author, inventor, creator, etc. of the document information can be appropriately obtained by using three types of codes paired with the document information.

[0016] Furthermore, an information processing device of the fifth invention is an information processing device according to any one of the first to fourth inventions, wherein the type determination unit determines the type of each of two or more individuals constituting a group, and is equipped with an evaluation unit that obtains an evaluation value of the group using the types of each of the two or more individuals, and an evaluation value output unit that outputs the evaluation value, and the evaluation unit is an information processing device that obtains a lower evaluation value when the types of the two or more individuals are the same compared to when they are different.

[0017] With this configuration, an evaluation value regarding innovation in one group can be obtained.

[0018] In addition, the information processing device of the sixth invention is an information processing device that, for any one of the second to fourth inventions, determines an intersection line ab between a plane a corresponding to a first type determined by the MAX (paper code concentration, patent code concentration, competitive funding code concentration) of one group and a plane b corresponding to a second type determined by the MAX (paper code concentration, patent code concentration, competitive funding code concentration) of another group, obtains coordinate information on the intersection line ab, which is a number, share rate, or standard deviation for each piece of coordinate information corresponding to both the one group and the other group, and is equipped with an evaluation unit that obtains a larger evaluation value the larger the number, share rate, or standard deviation, and an evaluation value output unit that outputs the evaluation value.

[0019] With this configuration, it is possible to obtain an evaluation value regarding innovation for two or more groups of authors, inventors, creators, etc. of the document information.

[0020] Moreover, in comparison with the fourth invention, the information processing device of the seventh invention further comprises a three-dimensional vector calculation unit in addition to or instead of the type determination unit, and a BEP vector output unit in addition to or instead of the type output unit, and the three-dimensional vector calculation unit further comprises a coordinate acquisition means for acquiring one or more coordinate information (x, y, z) corresponding to two or more pieces of document information of one individual or one group that satisfy a selection condition from the coordinate storage unit, and a coordinate acquisition means for acquiring one or more pieces of coordinate information (x, y, z) corresponding to two or more pieces of document information of one individual or one group that satisfy a selection condition from the coordinate storage unit, and a coordinate acquisition means for acquiring one or more pieces of coordinate information (x, y, z) corresponding to two or more pieces of document information of one individual or one group that satisfy a selection condition, and a coordinate acquisition means for acquiring one or more pieces of coordinate information (x, y, z) corresponding to two or more pieces of document information of one individual or one group, and The information processing device is equipped with a BEP vector acquisition means that calculates (x, y, z) dx dz / Σ, and γ(z) / Σ = ∫∫ f(x, y, z) dy dx / Σ (where f is a function that acquires the number of document information corresponding to the coordinate information (x, y, z)), calculates normalized values ​​b(x), e(y), and p(z) that are normalized values ​​of α(x) / Σ, β(y) / Σ, and γ(z) / Σ, and acquires a BEP vector (B, E, P) that is a three-dimensional vector whose starting point is (x, y, z) and whose B component is b(x), whose E component is e(y), and whose P component is p(z), and the BEP vector output unit outputs the BEP vector acquired by the BEP vector acquisition means.

[0021] Such a configuration allows vectors to be obtained for types of individuals or groups.

[0022] Moreover, in an information processing device of the eighth invention, compared to the seventh invention, the condition receiving unit receives a first selection condition which is a condition for identifying a first subject who is a first individual or a first group, and a second selection condition which is a condition for identifying a second subject who is a second individual or a second group, the coordinate acquisition means acquires from the coordinate storage unit a first coordinate set which is one or more pieces of coordinate information corresponding to two or more pieces of literature information of the first subject who satisfies the first selection condition, and acquires a second coordinate set which is one or more pieces of coordinate information corresponding to two or more pieces of literature information of a second subject who satisfies the second selection condition, and the BEP vector acquisition means acquires a BEP vector from the first coordinate set. The information processing device obtains one BEP vector (B1, E1, P1) and obtains a second BEP vector (B2, E2, P2) using a second coordinate set, and the three-dimensional vector calculation unit is equipped with a heterogeneity calculation means for obtaining a heterogeneity which is the degree of difference between the first BEP vector (B1, E1, P1) and the second BEP vector (B2, E2, P2) whose starting points are at the same coordinates, or a homogeneity calculation means for obtaining a homogeneity which is the degree of similarity between the first BEP vector (B1, E1, P1) and the second BEP vector (B2, E2, P2) whose starting points are at the same coordinates, and further equipped with a relationship output unit for outputting the heterogeneity or homogeneity.

[0023] With this configuration, the vectors of two subjects, either individuals or groups, can be used to obtain the degree of heterogeneity or homogeneity of the two subjects.

[0024] Furthermore, in the information processing device of the ninth invention, in relation to the eighth invention, the three-dimensional vector calculation unit includes a document information prediction means, and the document information prediction means performs a calculation of multiplying HT(x,y,z), which is the square of sine theta, by √f1(x,y,z), which is the square root of f1(x,y,z) of the first subject, and √f2(x,y,z), which is the square root of f2(x,y,z) of the second subject, or by half the sum of f1(x,y,z) of the first subject and f2(x,y,z) of the second subject, where HT(x,y,z) is the square of sine theta, and This information processing device obtains measured values, or performs a calculation to multiply HM(x,y,z), which is the square of cosθ, as the angle θ between the first BEP vector and the second BEP vector, whose starting points are at the same coordinates, by √f1(x,y,z), which is the square root of f1(x,y,z) of the first subject, and √f2(x,y,z), which is the square root of f2(x,y,z) of the second subject, or by half the sum of f1(x,y,z) of the first subject and f2(x,y,z) of the second subject, to obtain a predicted value of the literature information based on the homogeneity, and the relation output unit outputs the predicted value of the literature information obtained by the literature information prediction means.

[0025] With this configuration, it is possible to obtain a predicted value of literature information based on the heterogeneity or the homogeneity of the two subjects by using vectors of two subjects, which are individuals or groups.

[0026] Moreover, in the information processing device of the tenth invention, in comparison with the ninth invention, the document information prediction means obtains f1(x,y,z) of the first subject and f2(x,y,z) of the second subject for HT(x,y,z) which is the square of sine theta as the angle ta between the first BEP vector and the second BEP vector, the starting points of which are at the same coordinates, and further obtains the product and sum of these, and if the product is zero and the sum is positive, multiplies by the square root of the sum, otherwise multiplies by the square root of the product to obtain a document information prediction means based on the heterogeneity. This information processing device obtains a predicted value of document information, or obtains f1(x,y,z) for a first subject and f2(x,y,z) for a second subject for HM(x,y,z), which is the square of cosθ as the angle θ between a first BEP vector and a second BEP vector whose starting points are at the same coordinates, and further calculates the product and sum of these. If the product is zero and the sum is positive, multiply by the square root of the sum, or if not, multiply by the square root of the product to obtain a predicted value of document information based on homogeneity.

[0027] With this configuration, it is possible to obtain a predicted value of literature information based on an appropriate heterogeneity or a predicted value of literature information based on an appropriate homogeneity of the two subjects, using vectors of two subjects who are individuals or groups. Effect of the Invention

[0028] According to the information processing device of the present invention, it is possible to use document information to obtain the type of individual or group who is the author, inventor, creator, etc. of the document information. Also, according to the information processing device of the present invention, it is possible to obtain the degree of heterogeneity or homogeneity of two subjects, which are individuals or groups, using vectors of the two subjects, and obtain a predicted value of the document information based on the obtained value, in other words, the degree of possibility of intellectual property creation through joint research, etc., between the two subjects. [Brief description of the drawings]

[0029] [Figure 1] Block diagram of information processing device 1 according to embodiment 1. [Diagram 2] A flowchart illustrating an example of the operation of the information processing device 1. [Diagram 3]A flowchart illustrating an example of the coordinate acquisition process [Figure 4] Flowchart illustrating an example of a same-type determination process [Diagram 5] A flowchart for explaining an example of the paper code concentration acquisition process [Figure 6] A flowchart for explaining an example of the patent code concentration acquisition process [Figure 7] A flowchart for explaining an example of the competitive fund code concentration acquisition process [Figure 8] Flowchart for explaining an example of the same type acquisition process [Figure 9] A flowchart illustrating an example of the second evaluation process. [Figure 10] A diagram showing the image of the three-dimensional space [Figure 11] A diagram showing the three-dimensional space corresponding to each researcher. [Figure 12] Image of the network structure for each research team [Figure 13] A graph showing the number of papers published per person in each of the six research teams over a five-year period [Figure 14] FIG. 2 is a diagram illustrating the second evaluation process. [Figure 15] Block diagram of information processing device 2 according to embodiment 2. [Figure 16] A flowchart for explaining an example of the operation of the information processing device 2. [Figure 17] A flowchart for explaining an example of the BEP vector acquisition process [Figure 18] A flowchart for explaining an example of homogeneity degree acquisition processing [Figure 19] A flowchart for explaining an example of homogeneity acquisition processing [Figure 20] A figure showing the number of documents for each candidate coordinate [Figure 21] A diagram showing an example of the output of the same BEP vector [Figure 22] A diagram showing an example of the output of the same BEP vector [Diagram 23]A figure showing the number of documents for each candidate coordinate [Figure 24] A diagram showing an example of the output of the same BEP vector [Diagram 25] Example of output of homogeneity index [Figure 26] Example of output of homogeneity index [Figure 27] Example of output of homogeneity index [Figure 28] A figure showing the number of documents for each candidate coordinate [Figure 29] A diagram showing an example of the output of the same BEP vector [Diagram 30] An example of homogeneity output [Diagram 31] A graph showing the predicted value of bibliographic information based on homogeneity and the actual situation of collaborative papers [Diagram 32] A graph showing the predicted value of literature information based on homogeneity and the actual situation of collaborative papers [Diagram 33] Overview of the computer system according to the above embodiment. [Diagram 34] Block diagram of the computer system [Diagram 35] Diagram showing types of researchers in the prior art DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] Hereinafter, an embodiment of an information processing device and the like will be described with reference to the drawings. Note that, in the embodiments, components with the same reference numerals perform similar operations, and therefore repeated description may be omitted.

[0031] (Embodiment 1) In this embodiment, we will describe an information processing device that determines and outputs a type corresponding to each of two or more pieces of literature information using three types of codes, namely, a paper code, a patent code, and a competitive funding code (hereinafter referred to as "funding codes" as appropriate), that are assigned to each of two or more pieces of literature information.

[0032] In this embodiment, an information processing device that obtains and outputs an evaluation value regarding innovation in a group will be described.

[0033] Furthermore, in this embodiment, an information processing device that obtains and outputs an evaluation value regarding innovation in two groups conducting joint research will be described.

[0034] In this specification, information X being associated with information Y means that information Y can be obtained from information X, or information X can be obtained from information Y, and the method of association is not important. Information X and information Y may be linked, may exist in the same buffer, information X may be included in information Y, or information Y may be included in information X, etc.

[0035] Furthermore, in this specification, selecting or determining information Z means obtaining information Z, obtaining a pointer to information Z, obtaining the ID of information Z, setting a flag for information Z, etc., and is sufficient as long as information Z can be accessed.

[0036] 1 is a block diagram of an information processing device 1 according to the present embodiment. The information processing device 1 is, for example, a so-called personal computer, a smartphone, or a tablet terminal. The information processing device 1 may be, for example, a server capable of communicating with a terminal not shown. When the information processing device 1 is a server, the information processing device 1 may be, for example, a cloud server, an ASP server, or any other type of device.

[0037] The information processing device 1 includes a storage unit 11, a reception unit 12, a processing unit 13, and an output unit 14. The storage unit 11 includes a document storage unit 111 and a coordinate storage unit 112. The reception unit 12 includes a condition reception unit 121. The processing unit 13 includes a document acquisition unit 131, a code assignment unit 132, a coordinate accumulation unit 133, a type determination unit 134, and an evaluation unit 135. The type determination unit 134 includes a coordinate acquisition means 1341, a degree acquisition means 1342, and a Quasi-Stokes type determination means 1343. The output unit 14 includes a type output unit 141, and an evaluation output unit 142.

[0038] Various types of information are stored in the storage unit 11 constituting the information processing device 1. The various types of information include, for example, a document set, a learning device, and coordinate information, which will be described later.

[0039] The storage unit 11 stores a value (x) corresponding to each of two or more paper codes. In other words, it may be considered that each of the two or more paper codes has an order. The storage unit 11 stores a value (y) corresponding to each of two or more patent codes. In other words, it may be considered that each of the two or more patent codes has an order. The storage unit 11 stores a value (z) corresponding to each of two or more funding codes. In other words, it may be considered that each of the two or more funding codes has an order.

[0040] The learning device is information created by a learning unit (not shown). The learning device is information for inputting information based on literature information and outputting one or more codes. Information based on literature information is all or a part of literature information, or information obtained by processing literature information. The learning device is information created by learning, for example, teacher data, which is two or more pieces of literature information associated with one or more codes of one type, using a machine learning algorithm. The learning device may also be called a learning model, a classification model, a classifier, or the like.

[0041] The learning unit performs a learning process, acquires a learning device, and accumulates the learning device in the storage unit 11. The learning unit configures a learning device by a machine learning algorithm, for example, using two or more pieces of document information corresponding to one or more codes. This learning device is information used in an assignment process that outputs at least one or more codes when information based on literature information is given as input. It is preferable that the document information is a patent document and the code is an IPC (International Patent Classification). It is also preferable that the document information is an academic paper and the code is a JST classification code.

[0042] In addition, the machine learning used here can be, for example, deep learning, SVM, decision tree, random forest, etc. However, the type of machine learning is not limited.

[0043] The learning unit provides two or more pairs of a part or all of two or more pieces of document information and one or more codes corresponding to each piece of document information to a machine learning function (e.g., fastText, tensorflow library, TinySVM), runs the function, and performs a learning process to configure a learning device, which is then stored in the storage unit 11. In the present invention, since a paper code, a patent code, and a funding code are used, the learning device can perform a process of assigning, for example, three types of codes. Note that such learning processes are well known techniques, and therefore detailed explanations will be omitted.

[0044] A set of documents is stored in the document storage unit 111. Note that the document storage unit 111 may exist in an external device (not shown).

[0045] A document set is a collection of two or more document information. One or more attribute values ​​are usually associated with a document information. At least three types of codes are associated with a document information. The process by which the three types of codes are associated with the document information is not important.

[0046] Literature information is usually information on documents related to science or technology. Literature information is, for example, academic paper information (hereinafter, appropriately referred to as "paper information") or patent information. Literature information is, for example, a file or a database record, but the data structure is not important.

[0047] The three types of codes are paper codes, patent codes, and funding codes. It is preferable that any one or more of the three types of codes corresponding to the document information have one or more subclassification codes in addition to the first classification code. The patent code may include, for example, the first classification code and one or more subclassification codes.

[0048] The paper code is a code for classifying paper information. The paper code is, for example, the JST classification code. For information on the JST classification code, see, for example, the explanation at the URL "http: / / t21help.nikkei.co.jp / reference / docs / JD3_sousa.pdf."

[0049] A patent code is a code for classifying patent information. Types of patent codes include, for example, IPC, FI, F-term, and CPC.

[0050] The funding code is a code that is specified when obtaining competitive funding. The funding code is a code for classifying the field of application when obtaining competitive funding. The funding code may also be called the competitive funding code. The funding code is, for example, the Grant-in-Aid for Scientific Research classification code.<h2 style=";text-align:left;direction:ltr">https: / / www.jsps.go.jp / file / storage / gr ants / j-grantsinaid / 03_keikaku / data / h29 / h29_koubo_06.pdf#:~:t ext=%E7%92%B0%E5%A2%83%E7%94%9F%E7%90%86%E5%AD%A6%28%E5%90%AB%E4%BD%93%E5%8A%9B%E5%8C%BB%E5%AD%A6%E3%83%BB%E6%A0%84%E9%A4 %8A%E7%94%9F%E7%90%86%E5%AD%A6%29%207903%20%E5%8B%95%E7%89%A9%E7%94%9F%E7%90%86%E3%83%BB%E8%A1%8C%E5%8B%956803%E8%96%AC%E7%90%86%E5%AD%A6%E4%B8%80%E8%88%AC7904%20%E9%81%BA%E4%BC%9D%E3%83%BB%E6%9F%93%E8%89%B2%E4%BD%93%E5%8B%95%E6%85%8B6804%E5 %8C%BB%E5%8C%96%E5%AD%A6%E4%B8%80%E8%88%AC7905%20%E9%80%B2%E5%8C%96%E7%94%9F%E7%89%A9%E5%AD%A66805%E7%97%85%E6%85%8B%E5%8C%BB%E5%8C%96%E5%AD%A67906,%E7%94%9F%E7%89%A9%E5%A4%9A%E6%A7%98%E6%80%A7%E3%83%BB%E5%88%86%E9%A1%9E6806%E4%BA%BA%E9%A1%9 E%E9%81%BA%E4%BC%9D%E5%AD%A67907%20%E5%8C%BB%E6%AD%AF%20%E5%9F%BA%E7%A4%8E%E5%8C%BB%E5%AD%A6%20%E7%94%9F%E6%85%8B%E3%83% BB%E7%92%B0%E5%A2%836807%E4%BA%BA%E4%BD%93%E7%97%85%E7%90%86 %E5%AD%A67908%E2%80%BB%20%E8%96%AC%E5%AD%A6<h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr">

[0051] <h2 style=";text-align:left;direction:ltr"> In a situation where a paper code is associated with paper information, it is preferable that the patent code is automatically associated by the code assignment unit 132 described later. The patent code that is automatically associated is, for example, IPC. In such a case, the code assignment unit 132 acquires, for example, the patent code and the score, and associates the patent code and the score with the paper information.

[0052] It is also preferable that the code assigning unit 132, which will be described later, automatically associates the funding code. The funding code that is automatically associated is, for example, a Grant-in-Aid for Scientific Research classification code. In this case, the code assigning unit 132 acquires, for example, a Grant-in-Aid for Scientific Research classification code and a score, and associates the Grant-in-Aid for Scientific Research classification code and the score with the paper information. Here, the score is an evaluation index of the results in machine learning, and the higher the score, the more likely it is.

[0053] The same is true for situations where patent information is associated with a patent code.

[0054] The patent information may include, for example, one or more of the following: a patent specification, claims, and an abstract. The patent information may be a published patent publication or a registered patent publication. The patent information may be only a patent identifier and three codes that identify the patent.

[0055] The one or more attribute values ​​corresponding to the document information are, for example, one or more personal identifiers. A personal identifier is information that identifies an individual. A personal identifier is, for example, a name or an ID. An individual here is, for example, an author of a paper or an inventor of a patent.

[0056] Two or more pieces of three-dimensional coordinate information (x, y, z) are stored in the coordinate storage unit 112. The coordinate storage unit 112 stores one or more pieces of coordinate information corresponding to two or more pieces of document information.

[0057] The coordinate information is information in which a value corresponding to the paper code paired with the literature information is x, a value corresponding to the patent code paired with the literature information is y, and a value corresponding to the funding code paired with the literature information is z. Note that x, y, and z may be interchanged.

[0058] The receiving unit 12 receives various instructions or information, such as a selection condition, a coordinate accumulation instruction, a first evaluation instruction, and a second evaluation instruction, which will be described later.

[0059] The coordinate storage instruction is an instruction to store coordinate information.

[0060] The first evaluation instruction is an instruction to perform a first evaluation process. The first evaluation process is a process to obtain an evaluation value for a group. Such an evaluation value can be said to be information indicating the degree of innovation occurring in a group, or information regarding the activity of research in a group. The first evaluation process includes, for example, two or more personal identifiers. The first evaluation process includes, for example, a group identifier corresponding to two or more personal identifiers.

[0061] The second evaluation instruction is an instruction to perform a second evaluation process. The second evaluation process is a process to obtain an evaluation value when two groups conduct joint research. Such evaluation value can be said to be information indicating the degree of innovation when two groups conduct joint research, and information regarding the activity level when two groups conduct joint research. The second evaluation instruction includes one or more personal identifiers for each of the two or more groups. The second evaluation instruction includes, for example, a group identifier for each of the two or more groups, and such group identifiers correspond to one or more personal identifiers.

[0062] The input means for various instructions or information may be anything, such as a touch panel, a keyboard, a mouse, a menu screen, etc. The reception unit 12 may be realized by a device driver for an input means such as a touch panel or a keyboard, or control software for a menu screen, etc.

[0063] The condition receiving unit 121 receives a selection condition. The selection condition is a condition for selecting two or more pieces of literature information. The selection condition is a condition regarding one or more attribute values ​​related to the literature information. The selection condition is usually a condition for identifying one individual or one group. The selection condition has, for example, one or two or more individual identifiers.

[0064] Here, reception is a concept that includes reception of information input from input devices such as a keyboard, mouse, or touch panel, reception of information transmitted via a wired or wireless communication line, and reception of information read from recording media such as an optical disk, magnetic disk, or semiconductor memory.

[0065] The processing unit 13 performs various types of processing. The various types of processing are, for example, processing performed by a document acquisition unit 131, a code assignment unit 132, a coordinate accumulation unit 133, a type determination unit 134, and an evaluation unit 135.

[0066] The document acquisition section 131 acquires two or more pieces of document information from the document storage section 111. The document acquisition section 131 acquires, from the document storage section 111, two or more pieces of document information that satisfy the selection condition accepted by the condition acceptance section 121, for example.

[0067] The code assigning unit 132 performs an assignment process using a machine learning algorithm with the learning device of the storage unit 111 for each of two or more pieces of document information in the document storage unit 111, assigns one or more classification codes of one type, and associates them with each of the two or more pieces of document information. The assignment process may be called a prediction process. Assigning a classification code may be considered as obtaining a classification code.

[0068] It is preferable that the code assignment unit 132 uses a learning device to perform an assignment process using a machine learning algorithm for each of two or more pieces of document information in the document storage unit 111, assigns one or more patent classification codes (e.g., IPC) to each piece of document information, and stores the one or more IPCs in association with each piece of document information.

[0069] It is also preferable to assign one or more paper codes (for example, JST classification codes) and one or more funding codes (for example, Grant-in-Aid for Scientific Research classification codes) to each piece of literature information and store them.

[0070] The machine learning algorithm used in the assignment process does not matter. Examples of machine learning algorithms that can be used include deep learning, SVM, decision trees, random forests, etc.

[0071] The assignment process using a machine learning algorithm means, for example, providing two or more pieces of literature information and a learning device to a machine learning function (e.g., a library of fastText, TinySVM, or tensorflow) and running the function to obtain one or more classification codes of one type for each piece of literature information. Note that when the code assignment unit 132 obtains two or more classification codes and scores corresponding to each classification code using the machine learning function, the code assignment unit 132 may select and obtain one or more classification codes whose scores are high enough to satisfy a predetermined condition. A score that is high enough to satisfy a predetermined condition means, for example, that the score is the maximum value, that the score is equal to or greater than a threshold value, or the like.

[0072] The code assigning unit 132 assigns, for example, to each of two or more pieces of document information in the document storage unit 111, one or more Grant-in-Aid for Scientific Research classification codes corresponding to each piece of document information.

[0073] The coordinate accumulation unit 133 acquires, for each of two or more pieces of literature information in the literature storage unit 111, coordinate information (x, y, z) from the storage unit 11, where x is a value corresponding to the paper code paired with the literature information, y is a value corresponding to the patent code paired with the literature information, and z is a value corresponding to the competitive funding code paired with the literature information, and accumulates one or more pieces of coordinate information (x, y, z) in the coordinate storage unit 112 in association with each of the two or more pieces of literature information.

[0074] The coordinate accumulation unit 133 may acquire, for each of one or more pieces of literature information among the two or more pieces of literature information, primary coordinate information, which is coordinate information using only the primary classification code of three types of classification, and one or more pieces of secondary coordinate information, which is coordinate information using at least one secondary classification code, and accumulate the primary coordinate information and the one or more pieces of secondary coordinate information in the coordinate storage unit 112 in association with each of the one or more pieces of literature information.

[0075] The type determination unit 134 acquires the paper code concentration based on αmax, which is the maximum value of the number of codes that appear most frequently among the paper codes paired with each of the two or more pieces of literature information to be processed. The two or more pieces of literature information to be processed are, for example, two or more pieces of literature information that satisfy the selection conditions accepted by the condition acceptance unit 121, or two or more pieces of literature information in the literature storage unit 111. The two or more pieces of literature information to be processed are, for example, a collection of literature information of one individual, or a collection of literature information of one group.

[0076] The type determination unit 134 obtains a patent code concentration degree based on βmax, which is the maximum value of the number of most frequently appearing codes among patent codes paired with each of the two or more document information to be processed. The type determination unit 134 obtains a funding code concentration degree based on γmax, which is the maximum value of the number of most frequently appearing codes among competitive funding codes paired with each of the two or more document information to be processed. Then, the type determination unit 134 determines a type corresponding to each of the two or more document information using the paper code concentration degree, the patent code concentration degree, and the funding code concentration degree.

[0077] The paper code concentration is information regarding the degree to which each of two or more pieces of document information to be processed is biased toward a classification indicated by a specific paper code.

[0078] The patent code concentration is information regarding the degree to which each of two or more pieces of document information being processed is biased toward a classification indicated by a specific patent code.

[0079] The fund code concentration is information regarding the degree to which each of two or more pieces of document information to be processed is biased toward a classification indicated by a specific fund code.

[0080] Each of the two or more pieces of document information to be processed is paired with three types of codes, namely, a paper code, a patent code, and a funding code, and one or more attribute values ​​related to the document information.

[0081] The type determination unit 134 may determine the type of each of two or more individuals that make up one group.

[0082] A type is information for classifying two or more pieces of document information to be processed. A type may also be called a class, a classification, etc. A type may also be called, for example, a quasi-Stokes type. Types include, for example, a quasi-Bohr type, a quasi-Edison type, and a quasi-Pasteur type. The quasi-Bohr type, the quasi-Edison type, and the quasi-Pasteur type may also be called the first type, the second type, and the third type, respectively. In other words, the names of the types are not important.

[0083] The reason why "quasi" is added to the type of type output by the information processing device 1 is to ensure accuracy, since, as described above, it may not completely match the Bohr type, Edison type, and Pasteur type described in Stokes' paper. The types of types may be, for example, nine types, such as [quasi-Bohr-single type, quasi-Bohr-following Edison type, quasi-Bohr-following Pasteur type], [quasi-Edison-single type, quasi-Edison-following Bohr type, quasi-Edison-following Pasteur type], [quasi-Paseur-single type, quasi-Paseur-following Bohr type, quasi-Paseur-following Edison type], by the processing of the flowchart in FIG. 8 described later. Even if the types are the nine types mentioned above, they are considered to include the quasi-Bohr type, quasi-Edison type, and quasi-Paseur type. Also, even if there are nine types, for example, each type may be called the first type, the second type, ..., and the ninth type. In other words, the names of the types are not limited.

[0084] The coordinate acquiring means 1341 acquires, from the coordinate storage section 112, two or more pieces of coordinate information corresponding to two or more pieces of document information to be processed.

[0085] The coordinate acquiring means 1341 may acquire two or more pieces of coordinate information, that is, head coordinate information and one or more pieces of sub-coordinate information, corresponding to one or more pieces of document information.

[0086] The degree acquiring means 1342 acquires the paper code concentration degree, the patent code concentration degree, and the fund code concentration degree using three types of codes corresponding to each of two or more pieces of document information to be processed.

[0087] The paper code concentration degree is usually a value based on αmax, which will be described later. The paper code concentration degree is, for example, αmax, the maximum share rate of the paper code, which will be described later, and the maximum deviation value of the paper code, which will be described later. The patent code concentration degree is usually a value based on βmax, which will be described later. The patent code concentration degree is, for example, βmax, the maximum share rate of the patent code, which will be described later, and the maximum deviation value of the patent code, which will be described later. The fund code concentration degree is usually a value based on γmax, which will be described later. The fund code concentration degree is, for example, γmax, the maximum share rate of the fund code, which will be described later, and the maximum deviation value of the fund code, which will be described later.

[0088] The degree acquisition means 1342 acquires a paper code concentration degree based on αmax, which is the maximum value of the number of most frequently appearing codes among paper codes paired with each of two or more document information to be processed. The degree acquisition means 1342 acquires a patent code concentration degree based on βmax, which is the maximum value of the most frequently appearing codes among patent codes paired with each of two or more document information to be processed. The degree acquisition means 1342 acquires a fund code concentration degree based on γmax, which is the maximum value of the most frequently appearing codes among fund codes paired with each of two or more document information to be processed.

[0089] The degree acquisition means 1342 acquires, for example, αmax, which is the maximum value of ∫∫f(x,y,z)dydz≡α(x), as the paper code concentration degree. The degree acquisition means 1342 acquires, for example, βmax, which is the maximum value of ∫∫f(x,y,z)dxdz≡β(y), as the patent code concentration degree. The degree acquisition means 1342 acquires, for example, γmax, which is the maximum value of ∫∫f(x,y,z)dydz≡γ(z), as the fund code concentration degree.

[0090] The degree acquisition means 1342 preferably acquires the total number of first classification codes and the total number of second classification codes corresponding to one or more pieces of document information for codes of types not associated with scores, acquires a code concentration degree based on the total number, determines a minimum score to be adopted so that the difference between the total number of codes of types associated with scores and the total number of codes of types not associated with scores satisfies an allowable condition, acquires the total number of first classification codes and the total number of second classification codes using only coordinate information corresponding to classification codes associated with scores equal to or greater than the minimum value, and acquires a code concentration degree based on both the total numbers. The code concentration degrees are paper code concentration degree, patent code concentration degree, and fund code concentration degree.

[0091] The acceptable condition is a condition related to the score to be adopted, for example, that the difference between the total number of types of chords not associated with a score and the total number of types of chords associated with a score is within a threshold or less than a threshold.

[0092] The degree acquisition means 1342 acquires, for example, a paper code concentration degree, which is the maximum share rate (αmax / Σ, where Σ≡ the total number of document information to be processed) of a paper code attached to each of two or more document information to be processed. The degree acquisition means 1342 acquires, for example, a patent code concentration degree, which is the maximum share rate (βmax / Σ, where Σ≡ the total number of document information to be processed) of a patent code attached to each of two or more document information to be processed. The degree acquisition means 1342 acquires, for example, a fund code concentration degree, which is the maximum share rate (γmax / Σ, where Σ≡ the total number of document information to be processed) of a fund code attached to each of two or more document information to be processed.

[0093] The degree acquisition means 1342, for example, acquires a share rate of each of two or more paper codes attached to two or more document information to be processed, acquires a maximum share rate of the paper codes from the two or more share rates, calculates a deviation value in the document information to which the maximum share rate is a target, and acquires a paper code concentration degree that is the deviation value. The degree acquisition means 1342, for example, acquires a share rate of each of two or more patent codes attached to two or more document information to be processed, acquires a maximum share rate of the patent codes from the two or more share rates, calculates a deviation value in the document information to which the maximum share rate is a target, and acquires a patent code concentration degree that is the deviation value. The degree acquisition means 1342, for example, acquires a share rate of each of two or more fund codes attached to two or more document information to be processed, acquires a maximum share rate of the fund codes from the two or more share rates, calculates a deviation value in the document information to which the maximum share rate is a target, and acquires a fund code concentration degree that is the deviation value.

[0094] The quasi-Stokes type determination means 1343 determines MAX(paper code concentration, patent code concentration, fund code concentration) and determines the type using the determined information. Note that MAX(paper code concentration, patent code concentration, fund code concentration) is a function that determines the information (paper code concentration, patent code concentration, or fund code concentration) that is the maximum value among paper code concentration, patent code concentration, and fund code concentration.

[0095] The quasi-Stokes type determination means 1343, for example, determines the type to be quasi-Bohr if MAX(paper code concentration, patent code concentration, fund code concentration)=paper code concentration, determines the type to be quasi-Edison if MAX(paper code concentration, patent code concentration, fund code concentration=patent code concentration, and determines the type to be quasi-Pasteur if MAX(paper code concentration, patent code concentration, fund code concentration=fund code concentration).

[0096] The quasi-Stokes type determination means 1343 may determine one of nine types, namely, [quasi-Bohr-single type, quasi-Bohr-following Edison type, quasi-Bohr-following Pasteur type], [quasi-Edison-single type, quasi-Edison-following Bohr type, quasi-Edison-following Pasteur type], and [quasi-Pasteur-single type, quasi-Pasteur-following Bohr type, quasi-Pasteur-following Edison type], by processing the flowchart of Fig. 8 described later. In this case, too, it can be said that the quasi-Stokes type determination means 1343 determines the type, quasi-Bohr type, quasi-Edison type, or quasi-Pasteur type.

[0097] The evaluation unit 135 acquires an evaluation value for a group including two or more individuals. The evaluation value can be said to be the degree to which innovation can be expected to be created in the group. The evaluation value can be said to be the degree to which innovation can be expected to be created when two or more individuals in the group conduct joint research.

[0098] The evaluation unit 135 obtains an evaluation value for innovation creation of one group, for example, by using the types of two or more individuals. The two or more individuals all belong to one group. The evaluation unit 135 usually obtains a lower evaluation value when the types of two or more individuals are the same, compared to when they are different.

[0099] The evaluation unit 135, for example, judges whether the types match for each pair of two individuals that make up one group. Next, the evaluation unit 135 calculates, for example, the rate at which the types match (match rate) or the rate at which the types do not match (mismatch rate). Next, the evaluation unit 135, for example, acquires a smaller evaluation value as the match rate increases, or acquires a larger evaluation value as the mismatch rate increases.

[0100] For example, the evaluation unit 135 regards each individual belonging to a team as a node, and when there is co-authored literature information, configures an edge connecting the nodes corresponding to each person who is a co-author. Next, the evaluation unit 135, for example, if the types of individuals corresponding to two or more nodes connected by an edge are the same, regards the edge as a homo edge, and if the types are different, regards the edge as a hetero edge, and acquires the number of homo edges or the number of hetero edges. Next, the evaluation unit 135, for example, acquires a lower evaluation value as the number of homo edges increases, or acquires a higher evaluation value as the number of hetero edges increases. Note that the evaluation unit 135, for example, calculates the homo edge rate "the number of homo edges / the total number of edges" and calculates the evaluation value by a decreasing function with the homo edge rate as a parameter. The evaluation unit 135, for example, calculates the hetero edge rate "the number of hetero edges / the total number of edges" and calculates the evaluation value by an increasing function with the hetero edge rate as a parameter.

[0101] The evaluation unit 135 determines an intersection line ab between a plane a corresponding to a first type determined by the MAX (paper code concentration, patent code concentration, competitive funding code concentration) of one group and a plane b corresponding to a second type determined by the MAX (paper code concentration, patent code concentration, competitive funding code concentration) of another group, obtains the number or share rate for each piece of coordinate information corresponding to both the one group and the other group, which is coordinate information on the intersection line ab, and obtains a larger evaluation value the larger the number or share rate. Note that a group usually includes two or more individuals, but it may be a single individual.

[0102] The output unit 14 outputs various types of information, such as the type and the evaluation value.

[0103] Here, output is a concept that includes display on a display, projection using a projector, printing on a printer, sound output, transmission to an external device, storage on a recording medium, and handing over the processing results to other processing devices or other programs, etc.

[0104] The type output unit 141 outputs the type determined by the type determination unit 134 .

[0105] The evaluation output unit 142 outputs the evaluation value acquired by the evaluation unit 135 .

[0106] The storage unit 11, the document storage unit 111, and the coordinate storage unit 112 are preferably non-volatile recording media, but may also be realized as volatile recording media.

[0107] There is no restriction on the process by which information is stored in the storage unit 11, etc. For example, information may be stored in the storage unit 11, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 11, etc., or information inputted via an input device may be stored in the storage unit 11, etc.

[0108] The reception unit 12 and the condition reception unit 121 are preferably realized by wireless or wired communication means, but may also be realized by a means for receiving broadcasts, a device driver for an input means such as a touch panel or keyboard, or control software for a menu screen.

[0109] The processing unit 13, the document acquisition unit 131, the code assignment unit 132, the coordinate storage unit 133, the type determination unit 134, the evaluation unit 135, the coordinate acquisition means 1341, the degree acquisition means 1342, and the quasi-Stokes type determination means 1343 can usually be realized by a processor, a memory, or the like. The processing procedure of the processing unit 13, etc. is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may also be realized by hardware (dedicated circuit). The processor may be a CPU, MPU, GPU, or the like, and the type is not important.

[0110] The output unit 14, the type output unit 141, and the evaluation output unit 142 may or may not include an output device such as a display or a speaker. The output unit 14, etc. may be realized by, for example, driver software for an output device, or a combination of driver software for an output device and an output device, etc. The output unit 14, etc. may be realized by wireless or wired communication means.

[0111] Next, an example of the operation of the information processing device 1 will be described with reference to the flowchart of FIG.

[0112] (Step S201) The reception unit 12 judges whether or not a coordinate accumulation instruction has been received. If a coordinate accumulation instruction has been received, the process proceeds to step S202, and if not, the process proceeds to step S206.

[0113] (Step S202) The coordinate accumulator 133 assigns 1 to a counter i.

[0114] (Step S203) Coordinate accumulation unit 133 judges whether or not the i-th document information for which coordinate information is to be acquired exists in document storage unit 111. If the i-th document information exists, the process proceeds to step S204, and if not, the process returns to step S201.

[0115] (Step S204) The coordinate accumulation unit 133 performs a process of acquiring coordinate information corresponding to the i-th document information. An example of such coordinate acquisition process will be described with reference to the flowchart of FIG.

[0116] (Step S205) The coordinate accumulation unit 133 increments the counter i by 1. The process returns to step S203.

[0117] (Step S206) Condition receiving unit 121 judges whether or not the selection conditions have been received. If the selection conditions have been received, the process proceeds to step S207, and if not, the process proceeds to step S210.

[0118] (Step S207) The type determination unit 134 acquires from the document storage unit 111 two or more pieces of document information that match the selection condition.

[0119] (Step S208) The type determination unit 134 performs a process of determining the type of two or more pieces of document information. An example of such a type determination process will be described with reference to the flowchart of FIG.

[0120] (Step S209) The type output unit 141 outputs the type acquired in step S208. The process returns to step S201.

[0121] (Step S210) The reception unit 12 judges whether or not a first evaluation instruction has been received. If a first evaluation product instruction has been received, the process proceeds to step S211, and if not, the process proceeds to step S223.

[0122] (Step S211) The evaluation unit 135 assigns 1 to a counter i.

[0123] (Step S212) The evaluation unit 135 judges whether or not the i-th individual in the group corresponding to the first evaluation instruction exists. If the i-th individual exists, the process proceeds to step S213, and if not, the process proceeds to step S216.

[0124] (Step S213) The type determination unit 134 acquires two or more pieces of document information corresponding to the i-th individual. Each of the two or more pieces of document information is associated with one or more personal identifiers. The personal identifier here is information that identifies the author or inventor of the document information.

[0125] (Step S214) The type determination unit 134 determines the type of the i-th individual by using two or more pieces of document information acquired in step S213. An example of such type determination processing will be described with reference to the flowchart of FIG.

[0126] (Step S215) The evaluation unit 135 increments the counter i by 1. The process returns to step S212.

[0127] (Step S216) The evaluation unit 135 assigns 1 to the counter j.

[0128] (Step S217) The evaluation unit 135 judges whether or not a j-th pair of two people who are considered to be co-authors or co-inventors exists. If the j-th pair exists, the process proceeds to step S218, and if not, the process proceeds to step S221.

[0129] (Step S218) The evaluation unit 135 acquires the types of each of the two individuals who make up the j-th pair. Note that these types have been acquired in step S214.

[0130] (Step S219) The evaluation unit 135 determines whether the types of the two individuals acquired in step S218 match or do not match.

[0131] (Step S220) The evaluation unit 135 increments the counter j by 1. The process returns to step S217.

[0132] (Step S221) The evaluation unit 135 acquires an evaluation value using the determination result (whether or not there is a match) in step S219.

[0133] It is preferable that the evaluation unit 135 obtains a lower evaluation value as the number of coincidences in the determination result in step S219 increases, and it is preferable that the evaluation unit 135 obtains a higher evaluation value as the number of non-matching in the determination result in step S219 increases.

[0134] (Step S222) The evaluation output unit 142 outputs the evaluation value acquired in step S221. Then, the process returns to step S201.

[0135] (Step S223) The accepting unit 12 judges whether or not a second evaluation instruction has been accepted. If a second evaluation product instruction has been accepted, the process proceeds to step S224, and if not, the process returns to step S201.

[0136] (Step S224) The evaluation unit 135 performs a second evaluation process. An example of the second evaluation process will be described with reference to the flowchart of FIG.

[0137] (Step S225) The evaluation output unit 142 outputs the evaluation value acquired in step S224. The process returns to step S201.

[0138] In the flowchart of FIG. 2, the code assigning unit 132 may assign a code to one or more pieces of document information by the method described above.

[0139] Furthermore, in the flowchart of FIG. 2, the process ends when the power is turned off or an interrupt for ending the process occurs.

[0140] Next, an example of the coordinate acquisition process in step S204 will be described with reference to the flowchart in FIG.

[0141] (Step S301) The coordinate accumulation unit 133 acquires one or more paper codes corresponding to the literature information to be processed.

[0142] (Step S302) The coordinate accumulation unit 133 acquires one or more patent codes corresponding to the document information to be processed.

[0143] (Step S303) The coordinate storage unit 133 acquires one or more funding codes associated with the document information to be processed.

[0144] (Step S304) The coordinate accumulation unit 133 acquires from the storage unit 11 the x coordinate value paired with each of the one or more paper codes acquired in step S301.

[0145] (Step S305) The coordinate accumulation unit 133 acquires from the storage unit 11 the y coordinate value paired with each of the one or more patent codes acquired in step S302.

[0146] (Step S306) The coordinate accumulation unit 133 acquires from the storage unit 11 the z coordinate value paired with each of the one or more fund codes acquired in step S303.

[0147] (Step S307) The coordinate accumulation unit 133 acquires top coordinate information which is coordinate information (x, y, z) having x corresponding to the top paper code among the one or more paper codes acquired in step S301, y corresponding to the top patent code among the one or more patent codes acquired in step S302, and z corresponding to the top funding code among the one or more funding codes acquired in step S303, and accumulates the top coordinate information in the coordinate storage unit 112 in association with the document information to be processed. Note that the paper code, patent code, and funding code may be only the top code.

[0148] (Step S308) The coordinate accumulator 133 assigns 1 to the counter i.

[0149] (Step S309) Coordinate accumulation unit 133 determines whether or not an i-th combination of (x, y, z) excluding the head coordinate information exists among the x coordinate values ​​(1 or more) acquired in step S304, the y coordinate values ​​(1 or more) acquired in step S305, and the z coordinate values ​​(1 or more) acquired in step S306. If the i-th combination exists, the process proceeds to step S310, and if not, the process returns to the upper level process.

[0150] (Step S310) The coordinate accumulation section 133 acquires the sub-coordinate information which is the i-th combination, and accumulates the sub-coordinate information in the coordinate storage section 112 in association with the document information to be processed.

[0151] (Step S311) The coordinate accumulation unit 133 increments the counter i by 1. The process returns to step S309.

[0152] Next, an example of the type determination process in step S208 will be described with reference to the flowchart in FIG.

[0153] (Step S401) The coordinate obtaining means 1341 obtains all coordinate information paired with each of two or more pieces of target document information.

[0154] (Step S402) The degree acquisition unit 1342 performs a paper code concentration degree acquisition process using the two or more pieces of coordinate information acquired in step S401. An example of the paper code concentration degree acquisition process will be described with reference to the flowchart of FIG.

[0155] (Step S403) The degree acquisition means 1342 performs a patent code concentration degree acquisition process using two or more pieces of coordinate information acquired in step S401. An example of the patent code concentration degree acquisition process will be described with reference to the flowchart of FIG.

[0156] (Step S404) The degree acquisition unit 1342 performs a fund code concentration degree acquisition process using the two or more pieces of coordinate information acquired in step S401. An example of the fund code concentration degree acquisition process will be described with reference to the flowchart of FIG.

[0157] (Step S405) The quasi-Stokes type determination means 1343 performs a type acquisition process using the paper code concentration degree acquired in step S402, the patent code concentration degree acquired in step S403, and the fund code concentration degree acquired in step S404. After that, it returns to the upper process. An example of the type acquisition process will be described with reference to the flowchart of FIG. 8.

[0158] Next, an example of the paper code concentration degree acquisition process in step S402 will be described with reference to the flowchart in FIG.

[0159] (Step S501) The degree obtaining means 1342 assigns 1 to a counter i.

[0160] (Step S502) The degree acquisition means 1342 judges whether or not the i-th x exists in the two or more pieces of coordinate information acquired in step S401. If the i-th x exists, the process proceeds to step S503, and if it does not exist, the process proceeds to step S506. Note that the i-th x is a value of x that is different from the other x values.

[0161] (Step S503) The degree acquiring means 1342 acquires the number of pieces of coordinate information including the i-th x, which is the number of pieces of coordinate information among the two or more pieces of coordinate information acquired in step S401.

[0162] (Step S504) The degree obtaining means 1342 obtains the share of the i-th x (the number of coordinate information including the i-th x / the number of document information to be processed) using the number of coordinate information obtained in step S503.

[0163] (Step S505) The degree obtaining means 1342 increments the counter i by 1. The process returns to step S502.

[0164] (Step S506) The degree obtaining means 1342 obtains αmax, which is the maximum value among the number of pieces of coordinate information obtained in step S503.

[0165] (Step S507) The degree acquiring unit 1342 acquires the maximum share rate, which is the maximum value among the share rates acquired in step S504.

[0166] (Step S508) The degree acquiring unit 1342 acquires the standard deviation of the maximum share rate acquired in step S507.

[0167] It is preferable that the degree acquiring means 1342 acquires αmax, or the maximum share rate, or the standard deviation of the maximum share rate as the degree of paper code concentration. However, the degree acquiring means 1342 may acquire another value using αmax, or the maximum share rate, or the standard deviation of the maximum share rate as the degree of paper code concentration.

[0168] Moreover, it is preferable that the degree acquiring means 1342 acquires the value of x or the paper code corresponding to the paper code concentration degree, and temporarily stores the value of x or the paper code in association with the paper code concentration degree.

[0169] Next, an example of the patent code concentration degree acquisition process in step S403 will be described with reference to the flowchart of FIG.

[0170] (Step S601) The degree obtaining means 1342 assigns 1 to a counter i.

[0171] (Step S602) The degree acquisition means 1342 judges whether or not the i-th y exists in the two or more pieces of coordinate information acquired in step S401. If the i-th y exists, the process proceeds to step S603, and if it does not exist, the process proceeds to step S606. Note that the i-th y is a y value that is different from the other y values.

[0172] (Step S603) The degree acquiring means 1342 acquires the number of pieces of coordinate information including the i-th y, which is the number of pieces of coordinate information among the two or more pieces of coordinate information acquired in step S401.

[0173] (Step S604) The degree obtaining means 1342 obtains the share of the i-th y (the number of coordinate information including the i-th y / the number of document information to be processed) using the number of coordinate information obtained in step S603.

[0174] (Step S605) The degree obtaining means 1342 increments the counter i by 1. The process returns to step S602.

[0175] (Step S606) The degree obtaining means 1342 obtains βmax, which is the maximum value among the number of pieces of coordinate information obtained in step S603.

[0176] (Step S607) The degree acquiring unit 1342 acquires the maximum share rate, which is the maximum value among the share rates acquired in step S604.

[0177] (Step S608) The degree acquiring unit 1342 acquires the standard deviation of the maximum share rate acquired in step S607.

[0178] It is preferable that the degree acquisition means 1342 acquires βmax, the maximum share rate, or the deviation value of the maximum share rate as the patent code concentration degree. However, the degree acquisition means 1342 may acquire another value using βmax, the maximum share rate, or the deviation value of the maximum share rate as the patent code concentration degree.

[0179] In addition, it is preferable that the degree acquiring means 1342 acquires a value of y or a patent code corresponding to the patent code concentration degree, and temporarily stores the value of y or the patent code in association with the patent code concentration degree.

[0180] Next, an example of the fund code concentration degree acquisition process in step S404 will be described with reference to the flowchart of FIG.

[0181] (Step S701) The degree obtaining means 1342 assigns 1 to a counter i.

[0182] (Step S702) The degree acquisition means 1342 judges whether or not the i-th z exists in the two or more pieces of coordinate information acquired in step S401. If the i-th z exists, the process proceeds to step S703, and if it does not exist, the process proceeds to step S706. Note that the i-th z is a z value different from the other z values.

[0183] (Step S703) The degree acquiring means 1342 acquires the number of pieces of coordinate information including the i-th piece of z, which is the number of pieces of coordinate information among the two or more pieces of coordinate information acquired in step S401.

[0184] (Step S704) The degree obtaining means 1342 obtains the share rate of the i-th z (the number of coordinate information including the i-th z / the number of document information to be processed) using the number of coordinate information obtained in step S703.

[0185] (Step S705) The degree obtaining means 1342 increments the counter i by 1. The process returns to step S702.

[0186] (Step S706) The degree obtaining means 1342 obtains γmax, which is the maximum value among the number of pieces of coordinate information obtained in step S703.

[0187] (Step S707) The degree acquiring unit 1342 acquires the maximum share rate, which is the maximum value among the share rates acquired in step S704.

[0188] (Step S708) The degree acquiring unit 1342 acquires the standard deviation of the maximum share rate acquired in step S707.

[0189] It is preferable that the degree acquiring means 1342 acquires γmax, the maximum share rate, or the deviation value of the maximum share rate as the fund code concentration degree. However, the degree acquiring means 1342 may acquire another value using γmax, the maximum share rate, or the deviation value of the maximum share rate as the fund code concentration degree.

[0190] Moreover, it is preferable that the degree acquiring unit 1342 acquires the value of z or the fund code corresponding to the fund code concentration degree, and temporarily stores the value of z or the fund code in association with the fund code concentration degree.

[0191] Next, an example of the type acquisition process in step S405 will be described with reference to the flowchart in FIG.

[0192] (Step S801) The quasi-Stokes type determination means 1343 judges whether the paper code concentration is the maximum among the paper code concentration, the patent code concentration, and the fund code concentration. If the paper code concentration is the maximum, the process proceeds to step S802. If not, the process proceeds to step S807.

[0193] (Step S802) The quasi-Stokes type determination means 1343 judges whether the patent code concentration satisfies a predetermined condition (adoption condition). If the adoption condition is satisfied, the process proceeds to step S803, and if not, the process proceeds to step S804. The adoption condition is, for example, that the patent code concentration is the second largest among the paper code concentration, the patent code concentration, and the fund code concentration, and that the patent code concentration is equal to or greater than the threshold. The adoption condition is, for example, that the patent code concentration is the second largest, and that the difference between the paper code concentration and the patent code concentration is within the threshold or less than the threshold. The adoption condition is, for example, that the patent code concentration is the second largest.

[0194] (Step S803) The quasi-Stokes type determination means 1343 determines the type to be "quasi-Bohr-quasi-Edison type", and returns to the upper process.

[0195] (Step S804) The quasi-Stokes type determination means 1343 determines whether the fund code concentration degree satisfies a predetermined condition (adoption condition). If the adoption condition is satisfied, the process proceeds to step S805, and if not, the process proceeds to step S806. The adoption condition is, for example, that the fund code concentration degree is the second highest and is equal to or greater than a threshold value. The adoption condition is, for example, that the fund code concentration degree is the second highest value and the difference between the paper code concentration degree and the fund code concentration degree is within a threshold value or less than a threshold value. The adoption condition is, for example, that the fund code concentration degree is the second highest.

[0196] (Step S805) The quasi-Stokes type determination means 1343 determines the type to be "quasi-Bohr-quasi-Pasteur type". Then, the process returns to the upper level process.

[0197] (Step S806) The quasi-Stokes type determination means 1343 determines the type to be "quasi-Bohr single type", and returns to the upper process.

[0198] (Step S807) The quasi-Stokes type determination means 1343 judges whether the patent code concentration is the largest among the paper code concentration, the patent code concentration, and the fund code concentration. If the patent code concentration is the largest, the process proceeds to step S808. If not, the process proceeds to step S813.

[0199] (Step S808) The quasi-Stokes type determination means 1343 judges whether the paper code concentration satisfies a predetermined condition (adoption condition). If the adoption condition is satisfied, proceed to step S809, and if not, proceed to step S810. The adoption condition is, for example, that the paper code concentration is the second highest and is equal to or greater than the threshold. The adoption condition is, for example, that the paper code concentration is the second highest and the difference between the patent code concentration and the paper code concentration is within the threshold or less than the threshold. The adoption condition is, for example, that the paper code concentration is the second highest.

[0200] (Step S809) The quasi-Stokes type determination means 1343 determines the type to be "quasi-Edison-sub-Bohr type." The process returns to the upper level process.

[0201] (Step S810) The quasi-Stokes type determination means 1343 judges whether the fund code concentration degree satisfies a predetermined condition (adoption condition). If the adoption condition is satisfied, the process proceeds to step S811, and if not, the process proceeds to step S812. An adoption condition is, for example, that the fund code concentration degree is the second highest and is equal to or greater than a threshold value. An adoption condition is, for example, that the fund code concentration degree is the second highest value and the difference between the patent code concentration degree and the fund code concentration degree is within a threshold value or less than a threshold value. An adoption condition is, for example, that the fund code concentration degree is the second highest.

[0202] (Step S811) The quasi-Stokes type determination means 1343 determines the type to be "quasi-Edison-sub-Pasteur type." The process returns to the upper level process.

[0203] (Step S812) The quasi-Stokes type determination means 1343 determines the type to be "quasi-Edison single type", and returns to the upper level process.

[0204] (Step S813) The quasi-Stokes pattern determination means 1343 judges whether the paper code concentration degree satisfies a predetermined condition (adoption condition). If the adoption condition is satisfied, proceed to step S809, and if not, proceed to step S814. The adoption condition is, for example, that the paper code concentration degree is the second highest and is equal to or greater than the threshold value. The adoption condition is, for example, that the paper code concentration degree is the second highest value and the difference between the fund code concentration degree and the paper code concentration degree is within the threshold value or less than the threshold value. The adoption condition is, for example, that the paper code concentration degree is the second highest.

[0205] (Step S814) The quasi-Stokes type determination means 1343 determines the type to be "quasi-Pasteur-sub-Bohr type". Then, the process returns to the upper level process.

[0206] (Step S815) The quasi-Stokes type determination means 1343 judges whether the patent code concentration satisfies a predetermined condition (adoption condition). If the adoption condition is satisfied, proceed to step S816, and if not, proceed to step S817. The adoption condition is, for example, that the patent code concentration is the second highest and is equal to or greater than the threshold. The adoption condition is, for example, that the patent code concentration is the second highest value and the difference between the fund code concentration and the patent code concentration is within the threshold or less than the threshold. The adoption condition is, for example, that the patent code concentration is the second highest.

[0207] (Step S816) The quasi-Stokes type determination means 1343 determines the type to be "quasi-Pasteur-Edison type", and returns to the upper level process.

[0208] (Step S817) The quasi-Stokes type determination means 1343 determines the type to be "quasi-Pasteur-single type", and returns to the upper level processing.

[0209] 8, the quasi-Stokes type determination means 1343 determines one of nine types. However, the quasi-Stokes type determination means 1343 may determine, for example, that the type is quasi-Bohr type if MAX(paper code concentration, patent code concentration, fund code concentration)=paper code concentration, that the type is quasi-Edison type if MAX(paper code concentration, patent code concentration, fund code concentration=patent code concentration, and that the type is quasi-Pasteur type if MAX(paper code concentration, patent code concentration, fund code concentration=fund code concentration).

[0210] Next, an example of the second evaluation process in step S224 will be described with reference to the flowchart in FIG.

[0211] (Step S901) The type determination unit 134 acquires two or more pieces of document information of the first user corresponding to the second evaluation instruction.

[0212] (Step S902) The type determination unit 134 determines the type of the first user by using two or more pieces of document information acquired in step S901. An example of such type determination processing has been described with reference to the flowchart of FIG.

[0213] (Step S903) The type determination unit 134 acquires two or more pieces of document information of the second user corresponding to the second evaluation instruction.

[0214] (Step S904) The type determination unit 134 determines the type of the second user by using the two or more pieces of document information acquired in step S903. An example of such type determination processing has been described with reference to the flowchart of FIG.

[0215] (Step S905) The evaluation unit 135 determines a plane a corresponding to the maximum code concentration degree of the first user. The plane a corresponding to the maximum code concentration degree is a plane cut by the code corresponding to the maximum code concentration degree of the first user in the three-dimensional space.

[0216] (Step S906) The evaluation unit 135 determines a plane b corresponding to the maximum code concentration degree of the second user. The evaluation unit 135 determines a plane b corresponding to the maximum code concentration degree of the second user. The plane b corresponding to the maximum code concentration degree is a plane cut by the code corresponding to the maximum code concentration degree of the second user in the three-dimensional space.

[0217] (Step S907) The evaluation unit 135 determines the intersection line ab between the plane a and the plane b.

[0218] (Step S908) The evaluation unit 135 acquires one or more pieces of coordinate information on the intersection line ab.

[0219] (Step S909) The evaluation unit 135 acquires the code concentration degree corresponding to each of the one or more pieces of coordinate information acquired in step S908.

[0220] (Step S910) The evaluation unit 135 obtains an evaluation value by using the code concentration degree of 1 or more obtained in step S909, and then returns to the upper level process.

[0221] The evaluation value is, for example, a value calculated by an increasing function with a parameter being a code concentration degree equal to or greater than 1. The evaluation value is, for example, the sum of code concentration degrees equal to or greater than 1.

[0222] A specific operation example of the information processing device 1 in this embodiment will be described below. Here, the information processing device 1 uses a JST classification code as the paper code, an IPC as the patent code, and a Grant-in-Aid for Scientific Research classification code as the funding code.

[0223] The storage unit 11 of the information processing device 1 stores x-coordinate values ​​corresponding to two or more JST classification codes (example of paper codes) having an order. The storage unit 11 also stores y-coordinate values ​​corresponding to two or more IPCs (example of patent codes) having an order. The storage unit 11 also stores z-coordinate values ​​corresponding to two or more Grant-in-Aid for Scientific Research classification codes (example of funding codes) having an order. This forms a three-dimensional space having three axes: the x-axis of paper classification based on the JST classification codes, the y-axis of patent classification based on the IPCs, and the z-axis of Grant-in-Aid for Scientific Research classification (funding classification) based on the Grant-in-Aid for Scientific Research classification codes (see FIG. 10).

[0224] The document storage unit 111 of the information processing device 1 stores a large number of document information pairs with the personal identifier "Professor Yoshinori Ohsumi", a large number of document information pairs with the personal identifier "Professor Shinya Yamanaka", and a large number of document information pairs with the personal identifier "Professor Tasuku Honjo". Each document information is associated with one or more JST classification codes, one or more IPCs, and one or more Grant-in-Aid for Scientific Research classification codes. The document information here is, for example, paper information. Each paper information is originally associated with one or more JST classification codes. Each paper information is associated with one or more IPCs assigned by the code assignment unit 132. Furthermore, each paper information is associated with one or more Grant-in-Aid for Scientific Research classification codes assigned by the code assignment unit 132.

[0225] Next, it is assumed that the reception unit 12 of the information processing device 1 receives a user's coordinate storage instruction from a terminal X (not shown). Next, the coordinate storage unit 133 acquires one or more pieces of three-dimensional coordinate information (x, y, z) for each piece of literature information in the literature storage unit 111 by the process described using the flowchart in FIG. 3, and stores the one or more pieces of coordinate information in the coordinate storage unit 112 in association with each piece of literature information.

[0226] In the above situation, the following three specific examples are output. Specific Example 1 is an example in which the type of author (which can be an inventor, etc.) of the literature information is determined and output. Specific Example 2 is an example in which the paper information of one research team is used to perform a first evaluation of the research team and an evaluation value is output. Specific Example 3 is an example in which the paper information of each of two teams is used to perform a second evaluation in the case of joint research between the two teams and an evaluation value is output.

[0227] (Example 1) Next, it is assumed that the condition receiving unit 121 of the information processing device 1 receives from the terminal X selection conditions having the personal identifiers "Professor Yoshinori Ohsumi", "Professor Shinya Yamanaka", and "Professor Tasuku Honjo".

[0228] Next, the type determination unit 134 acquires the paper code concentration, patent code concentration, and funding code concentration of the literature information of Professor Yoshinori Ohsumi through the type determination process described using the flowchart in FIG. 4. Here, the code concentration (paper code concentration, patent code concentration, and funding code concentration) adopts the above-mentioned share rate. Then, the type determination unit 134 acquires the maximum share rate of "37%" of the paper code concentration, and associates the JST classification code "EF01020S: Cell Physiology General" with the paper code concentration (see FIG. 11(a)). Then, since "MAX (paper code concentration, patent code concentration, funding code concentration) = paper code concentration", the type determination unit 134 acquires the type "quasi-Bohr type" in combination with the personal identifier "Professor Yoshinori Ohsumi".

[0229] Next, the type determination unit 134 obtains the paper code concentration, patent code concentration, and funding code concentration of the literature information of Professor Shinya Yamanaka through the type determination process described using the flowchart in FIG. 4. It is assumed here that the code concentration is a share rate. The type determination unit 134 obtains the maximum share rate of "50%" of the patent code concentration, and associates the patent code concentration with the IPC "C12N5: undifferentiated cells of humans, animals, and plants" (see FIG. 11(b)). Since "MAX (paper code concentration, patent code concentration, funding code concentration) = patent code concentration", the type determination unit 134 obtains the type "quasi-Edison type" in combination with the personal identifier "Professor Shinya Yamanaka".

[0230] Next, the type determination unit 134 obtains the paper code concentration, patent code concentration, and funding code concentration of the literature information of Professor Honjo Tasuku by the type determination process described using the flowchart in FIG. 4. It is assumed here that the code concentration is a share rate. The type determination unit 134 obtains the maximum share rate of "28%" of the funding code concentration, and associates the funding code concentration with the Grant-in-Aid for Scientific Research classification code "49070: Immunology" (see FIG. 11(c)). Since "MAX (paper code concentration, patent code concentration, funding code concentration) = funding code concentration", the type determination unit 134 obtains the type "quasi-Pasteur type" in combination with the personal identifier "Professor Honjo Tasuku".

[0231] Next, the type output unit 141 transmits to terminal X the type "quasi-Bohr type" associated with the personal identifier "Professor Yoshinori Ohsumi," the type "quasi-Edison type" associated with the personal identifier "Professor Shinya Yamanaka," and the type "quasi-Pasteur type" associated with the personal identifier "Professor Tasuku Honjo."

[0232] Then, terminal X receives the information and outputs the information, which is the type "quasi-Bohr type" associated with the personal identifier "Professor Yoshinori Ohsumi," the type "quasi-Edison type" associated with the personal identifier "Professor Shinya Yamanaka," and the type "quasi-Pasteur type" associated with the personal identifier "Professor Tasuku Honjo."

[0233] As described above, according to the first concrete example, the type of research of an individual can be determined and output by using the three types of codes assigned to the paper information of the individual.

[0234] (Example 2) In specific example 2, the literature acquisition unit 131 acquires literature information (here, paper information) paired with a personal identifier identifying a researcher for each of the researchers belonging to six research teams (G1, G2, G3, G4, G5, G6) from the literature storage unit 111. Next, the type determination unit 134 selects one of nine types for each researcher belonging to each of the six research teams by the type acquisition process described using the flowchart in Fig. 8, and associates the type with the researcher.

[0235] Next, the evaluation unit 135 treats the researchers as nodes for each of the six research teams, and when co-authored paper information exists, constructs edges that connect the nodes corresponding to each co-author. An image of the network structure for each research team is shown in FIG. 12. Note that the node corresponding to each researcher is associated with the type of each researcher. However, since it is not necessary to disclose the names of each researcher, these are replaced with the keywords of the research of that researcher.

[0236] Next, the evaluation unit 135 acquires a lower evaluation value for each of the six research teams when two or more types of each individual are the same among the nine types, as compared to when they are different among the nine types. That is, the evaluation unit 135 acquires the total number of edges and the number of homo edges or hetero edges for each of the six research teams. Note that a homo edge is an edge where the types corresponding to the two nodes connected by the edge are the same among the nine types. A homo edge is black in FIG. 12, and is proportional to the number of co-authored papers between the researchers located at the nodes at both ends. A hetero edge is an edge where the types corresponding to the two nodes connected by the edge are different among the nine types. A hetero edge is gray in FIG. 12, and is proportional to the number of co-authored papers between the researchers located at the nodes at both ends. Note that the diameter of the nodes at both ends of each edge is proportional to the number of papers by the researcher located there. Next, the evaluation unit 135 calculates the homo edge rate "number of homo edges / total number of edges" or the hetero edge rate "number of hetero edges / total number of edges". Next, the evaluation unit 135 calculates an evaluation value for each of the six research teams using a decreasing function with the homo-edge rate as a parameter, or an increasing function with the hetero-edge rate as a parameter.

[0237] The data used for the calculations in Figure 12 was calculated by calculating the standard deviations of paper code concentration, patent code concentration, and research grant classification code concentration for the population of Kyoto University researchers ranked 1st to 1,000th in the number of physics-related papers published over a five-year period. If the concentration of one code is more dominant than any of the other codes at a threshold value of 5 or more, the code is classified as "quasi-BEP single type," if either code is more dominant than the other code at a threshold value of 5 or more, the code is classified as "quasi-BEP secondary BEP type," and all other cases are classified as "unclassified."

[0238] Next, the evaluation output section 142 outputs the evaluation value acquired by the evaluation section 135 for each of the six research teams.

[0239] FIG. 13 shows the number of papers per person in each of the six research teams (G1, G2, G3, G4, G5, and G6) over a five-year period. FIG. 13 shows the results of multiple regression analysis, and the multiple regression R2 Since the correlation coefficient was 0.61 and the significance F value was 0.066, it can be considered that there is a strong and significant negative correlation between the homo edge rate of each research team and the number of papers produced.

[0240] In addition, six chemistry research teams from Osaka University and five biology research teams from Kobe University also obtained similar results, showing a negative correlation between the homo edge rate and the number of papers produced.

[0241] (Example 3) In specific example 3, a case where an evaluation value is obtained when a research team including Professor Shigetoshi Sugawara of Tohoku University and engineers of Shimadzu Corporation conduct joint research on high-speed cameras will be described. Two or more pieces of document information by Professor Shigetoshi Sugawara of Tohoku University and two or more pieces of document information by engineers of Shimadzu Corporation are stored in the document storage unit 111 of the information processing device 1. Each piece of document information is associated with one or more JST classification codes, one or more IPCs, and one or more Grant-in-Aid for Scientific Research classification codes.

[0242] The type determination unit 134 obtained the type of Professor Shigetoshi Sugawara et al., "Pasteur type," by using information on two or more papers by Professor Shigetoshi Sugawara et al. before the start of the collaborative research. The type determination unit 134 also detected that the fund code corresponding to the maximum fund code concentration degree was (21060: electronic devices). Then, the evaluation unit 135 determined the plane a corresponding to the fund code (21060: electronic devices) (see FIG. 14(a)).

[0243] The type determination unit 134 also obtained the Shimadzu Corporation type "Edison type" using two or more pieces of paper information by Shimadzu Corporation engineers before the start of the joint research. The type determination unit 134 also detected that the patent code corresponding to the maximum patent code concentration degree was (H04 N5: image method). The evaluation unit 135 then determined the plane b corresponding to the patent code (H04 N5: image method) (see FIG. 14(b)).

[0244] Next, the evaluation unit 135 determined the intersection line ab between the plane a and the plane b (see FIGS. 14(c) and (f)). Note that the intersection line ab is the dashed line in FIGS. 14(c) and (f). At this intersection ab, the truth-seeking research stance of Professor Shigetoshi Sugawa and his team and the application-oriented research stance of Shimadzu Corporation's engineers collide, creating a so-called "chemical reaction" that may result in the creation of unexpected papers, patents, etc. An example of predicting and evaluating this is described below.

[0245] Next, the evaluation unit 135 obtained the number of pieces of paper information corresponding to each of the one or more points on the intersection line ab, and calculated the sum of the number of pieces of paper information. Next, the evaluation unit 135 obtained an evaluation value by an increasing function with the sum of the number of pieces of paper information as a parameter.

[0246] Next, the evaluation output unit 142 output the evaluation value acquired by the evaluation unit 135. Note that the evaluation value can be said to be the degree to which innovation can be expected to occur in the case where Professor Shigetoshi Sugawa of Tohoku University and his colleagues and engineers of Shimadzu Corporation conduct joint research on high-speed cameras based on the paper before the joint research.

[0247] FIG. 14(d) is a graph showing the sum of the number of papers published by both teams before the collaborative research at the coordinate point where the papers coexist on the intersection line ab between planes a and b, and FIG. 14(e) is a graph showing the number of papers published during the collaborative research (since 2007) and thereafter. Comparing FIG. 14(d) and (e), it can be seen that the distributions are similar. This shows that a "chemical reaction" caused by the collaborative research of the two teams occurs at the coexistence coordinate of the intersection line ab of the concentration planes of both teams, and that the probability of this occurring is roughly proportional to the sum of the number of papers published before the collaborative research at the coexistence coordinate of both teams.

[0248] As described above, according to this embodiment, the type of an individual or group can be obtained by using three types of codes paired with document information. Note that such a type is, for example, a classification according to the above-mentioned "BEP type."

[0249] Furthermore, according to this embodiment, an evaluation value regarding innovation in one group can be obtained.

[0250] Furthermore, according to this embodiment, an evaluation value regarding innovation in the case where two groups are conducting joint research can be obtained.

[0251] In addition, according to this embodiment, the type of researcher (classification based on the BEP type) can be determined, making it possible to extract Pasteur-type researchers as candidates for CEO or CTO of university-based startups and provide efficient support such as entrepreneurial education.

[0252] Furthermore, according to this embodiment, the impact on team activity level of classification based on researchers' BEP types can be estimated, so that this can be applied to team building in interdisciplinary research, industry-academia collaborative research, and the like, to create highly active teams.

[0253] Furthermore, according to this embodiment, since the evaluation value in the case of conducting joint research can be predicted, joint research can be planned, and interdisciplinary research or industry-academia collaborative research can be promoted.

[0254] The processing in this embodiment may be realized by software. This software may be distributed by software download or the like. This software may also be recorded on a recording medium such as a CD-ROM and distributed. This also applies to the other embodiments in this specification. The software that realizes the information processing device 1 in this embodiment is a program such as the following. In other words, this program causes a computer to function as a type determination unit that obtains a paper code concentration degree based on αmax, which is the maximum number of codes that appear most frequently among the paper codes for classifying paper information, the patent codes for classifying patent information, and three types of codes that are funding codes specified when obtaining competitive funding, and the paper codes that are paired with two or more pieces of literature information and paired with one or more attribute values ​​related to the literature information, obtains a patent code concentration degree based on βmax, which is the maximum number of codes that appear most frequently among the patent codes that are paired with the two or more pieces of literature information, and obtains a funding code concentration degree based on γmax, which is the maximum number of codes that appear most frequently among the funding codes that are paired with the two or more pieces of literature information, and uses the paper code concentration degree, the patent code concentration degree, and the funding code concentration degree to determine a type corresponding to a paper set that is the two or more pieces of literature information, and a type output unit that outputs the type.

[0255] (Embodiment 2) In this embodiment, an information processing device that acquires vectors relating to the type of individual or group will be described.

[0256] In addition, in this embodiment, we will describe an information processing device that can use vectors of two subjects, who are individuals or groups, to obtain the heterogeneity or homogeneity of the two subjects, and can obtain a predicted value of literature information based on the vectors, in other words, the degree of possibility of intellectual property creation through collaborative research between the two subjects.

[0257] FIG. 15 is a block diagram of an information processing device 2 in the present embodiment.

[0258] The information processing device 2 includes a storage unit 11, a reception unit 22, a processing unit 23, and an output unit 24. The reception unit 22 includes a condition reception unit 221. The processing unit 23 includes a document acquisition unit 131, a code assignment unit 132, a coordinate storage unit 133, a type determination unit 134, an evaluation unit 135, and a three-dimensional vector calculation unit 231. The three-dimensional vector calculation unit 231 includes a BEP vector acquisition means 2311, a heterogeneity calculation means 2312, a homogeneity calculation means 2313, and a document information prediction means 2341. The three-dimensional vector calculation unit 231 may include only one of the homogeneity calculation means 2313 and the heterogeneity calculation means 2312. The output unit 24 includes a type output unit 141, an evaluation output unit 142, a BEP vector output unit 241, and a relationship output unit 242.

[0259] The receiving unit 22 included in the information processing device 2 receives various instructions or information, such as a first selection condition (to be described later), a second selection condition (to be described later), a coordinate accumulation instruction, a first evaluation instruction, and a second evaluation instruction.

[0260] Means for inputting various instructions or information include, for example, a touch panel, a keyboard, a mouse, a menu screen, and the like.

[0261] The condition receiving unit 221 receives a selection condition. The selection condition is a condition related to one or more attribute values ​​related to document information, and is a condition for identifying a target person who is an individual or a group. The condition receiving unit 221 may also receive a vector output instruction. The vector output instruction is an instruction to output a BEP vector. The vector output instruction has a selection condition.

[0262] The condition receiving unit 221 receives a first selection condition and a second selection condition. The first selection condition is a condition for identifying a first subject, which is a first individual or a first group. The second selection condition is a condition for identifying a second subject, which is a second individual or a second group. The first selection condition and the second selection condition are conditions for one or more attribute values ​​related to the literature information. The condition receiving unit 221 may receive a degree output instruction. The degree output instruction is an instruction to output one or two pieces of information, the heterogeneity and the homogeneity. The degree output instruction includes the first selection condition and the second selection condition. The condition receiving unit 221 may receive a predicted value output instruction. The predicted value output instruction is an instruction to output one or two predicted values, the predicted value of the literature information based on the heterogeneity and the predicted value of the literature information based on the homogeneity. The predicted value output instruction includes the first selection condition and the second selection condition. In response to the reception of a predicted value output instruction, information on one or more of the heterogeneity used to obtain the predicted value and the homogeneity used to obtain the predicted value may be output.

[0263] The processing unit 23 performs various types of processing. The various types of processing are, for example, processing performed by the document acquisition unit 131, the code assignment unit 132, the coordinate accumulation unit 133, the type determination unit 134, the evaluation unit 135, or the three-dimensional vector calculation unit 231.

[0264] The BEP vector acquisition means 2311 constituting the three-dimensional vector calculation unit 231 acquires a BEP vector (B, E, P) using one or more pieces of coordinate information (x, y, z) acquired by the coordinate acquisition means 1341. The constituting BEP vector acquisition means 2311 acquires a BEP vector (B, E, P) for each of one or two or more candidate coordinates. It is preferable that the BEP vector acquisition means 2311 acquires a BEP vector (B, E, P) for two or more candidate coordinates.

[0265] The BEP vector (B,E,P) is information that indicates the characteristics of the type of subject at the candidate coordinates. The candidate coordinates are points in the three-dimensional space described above. The candidate coordinates are coordinates (x,y,z) that indicate a point in the three-dimensional space. The BEP vector (B,E,P) is a vector whose elements are the B component, the E component, and the P component.

[0266] The B component is a value that indicates the degree of tendency of "researchers who emphasize the pursuit of truth." The B component can be said to be a value that indicates the degree of Bohr type. The E component is a value that indicates the degree of tendency of "researchers who emphasize applications." The E component can be said to be a value that indicates the degree of Edison type. The P component is a value that indicates the degree of tendency of "researchers who emphasize both the pursuit of truth and applications." The P component can be said to be a value that indicates the degree of Pasteur type.

[0267] The BEP vector (B, E, P) acquired by the BEP vector acquisition means 2311 is a vector indicating the tendency of the subject corresponding to the document information that is the basis of the one or more pieces of coordinate information acquired by the coordinate acquisition means 1341.

[0268] The BEP vector acquisition means 2311 calculates α(x) / Σ, β(y) / Σ, and γ(z) / Σ for each of two or more individuals or groups including one individual or one group, for example, using one or more coordinate information (x, y, z). Next, the BEP vector acquisition means 2311 calculates normalized values ​​b(x), e(y), and p(z) that are normalized values ​​of α(x) / Σ, β(y) / Σ, and γ(z) / Σ. Next, the BEP vector acquisition means 2311 acquires a BEP vector (B, E, P) that is a three-dimensional vector with b(x) as the B component, e(y) as the E component, and p(z) as the P component.

[0269] The normalized value here may be, for example, a deviation value, a value obtained by minimum / maximum normalization, a value obtained by standardization to convert the value so that the average is 0 and the standard deviation is 1, or the like.

[0270] Furthermore, the equation for calculating α(x) / Σ, β(y) / Σ, and γ(z) / Σ is the following Equation 1.

[0271]

number

[0272] Next, the processing of the BEP vector acquisition means 2311 will be explained. In other words, it is possible to visualize in which category the literature information of one individual or one group is concentrated as a BEP vector. Since the components of the vector are assumed to be independent, α is a function of x only, β is a function of y only, and γ is a function of z only.

[0273] The heterogeneity calculation means 2312 calculates the first BEP vector (B 1 ,E 1 ,P 1 ) and the second BEP vector (B 2 ,E 2 ,P 2 ) to obtain the degree of heterogeneity. 1 ,E 1 ,P 1 ) is the BEP vector of the first subject. The second BEP vector (B 2 ,E 2 ,P 2 ) is the BEP vector of the second subject.

[0274] It is preferable that the heterogeneity calculation means 2312 obtains a heterogeneity for each of two or more candidate coordinates. The heterogeneity may be a vector or a scalar. The heterogeneity is information on the degree of heterogeneity. The heterogeneity may be called the heterogeneity.

[0275] The heterogeneity calculation means 2312 acquires the heterogeneity, for example, by the following method (1) or (2). (1) Cross product method

[0276] The heterogeneity calculation means 2312 calculates the first BEP vector (B 1,E 1 ,P 1 ) and the second BEP vector (B 2 ,E 2 ,P 2 ) to obtain a heterogeneity which is the cross product of the cross product. Note that the heterogeneity calculation unit 2312 may obtain a heterogeneity using the above cross product as a parameter, for example. (2) Method using the angle θ between two BEP vectors

[0277] The heterogeneity calculation means 2312 acquires the heterogeneity, for example, as follows. That is, the heterogeneity calculation means 2312 acquires the angle θ(x, y, z) between the first BEP vector and the second BEP vector. Next, the angle θ(x, y, z) is used to calculate HT(x, y, z), which is the square of SINθ(x, y, z). Note that HT(x, y, z) is an example of heterogeneity.

[0278] The literature information prediction means 1314 acquires a predicted value of literature information based on the heterogeneity acquired by the heterogeneity calculation means 2312. Also, the literature information prediction means 1314 acquires a predicted value of literature information based on the homogeneity acquired by the homogeneity calculation means 2313. The predicted value of literature information is a predicted value of the number of literature documents that will be created when a first subject and a second subject conduct collaborative research. The literature is, for example, a paper, but it may also be a patent or the like. Also, collaborative research may be broadly interpreted as working together.

[0279] Next, the process of the document information prediction means 2314 acquiring the predicted value of document information based on the heterogeneity will be described. The document information prediction means 2314 calculates, for example, √f1(x,y,z) which is the square root of f1(x,y,z) of the first subject and √f2(x,y,z) which is the square root of f2(x,y,z) of the second subject. Next, the document information prediction means 2314 calculates the predicted value of document information based on the heterogeneity by multiplying, for example, HT(x,y,z) by the square root of f1(x,y,z) of the first subject and the square root of f2(x,y,z) of the second subject. Alternatively, the document information prediction means 2314 calculates, for example, half the sum of f1(x,y,z) of the first subject and f2(x,y,z) of the second subject. Next, the document information prediction means 2314 multiplies HT(x, y, z) by half the sum to calculate a predicted value of document information based on the heterogeneity. Note that f1 and f2 are functions that obtain the number of document information corresponding to the coordinate information (x, y, z).

[0280] That is, the heterogeneity calculation means 2312 and the document information prediction means 2314 respectively calculate the heterogeneity and a predicted value of the document information based thereon, for example, by the following Equation 2 or Equation 3.

[0281]

number

[0282]

number

[0283] Next, a case where the document information prediction means 1314 acquires a predicted value of document information based on the heterogeneity using a correction process will be described.

[0284] The heterogeneity calculation means 2312 obtains the angle θ(x, y, z) between the first BEP vector and the second BEP vector, and obtains HT(x, y, z), which is the square of SINθ(x, y, z), using the angle θ(x, y, z). Next, the literature information prediction means 2314 obtains f1(x, y, z) of the first subject, and obtains f2(x, y, z) of the second subject. Next, the literature information prediction means 2314 calculates the product and sum of f1(x, y, z) and f2(x, y, z). Next, if the product is zero and the sum is positive, the literature information prediction means 2314 multiplies HT(x, y, z) by the square root of the sum to obtain the heterogeneity, and if the product is not zero or the sum is not positive, the literature information prediction means 2314 multiplies HT(x, y, z) by the square root of the product to obtain the heterogeneity. It should be noted that the starting points of the first BEP vector and the second BEP vector are at the same coordinates.

[0285] That is, the heterogeneity calculation means 2312 and the document information prediction means 2314 each calculate the heterogeneity and a predicted value of the document information based thereon, for example, by the following formula 4.

[0286]

number

[0287] The homogeneity calculation means 2313 acquires homogeneity, which is the degree of similarity between the first BEP vector (B1, E1, P1) and the second BEP vector (B2, E2, P2). It is preferable that the homogeneity calculation means 2313 acquires homogeneity for each of two or more candidate coordinates. The homogeneity may be a vector or a scalar. The homogeneity is information regarding the degree of homogeneity. The homogeneity may be referred to as homogeneity.

[0288] The homogeneity calculation means 2313 obtains the homogeneity, for example, by the following method (1) or (2). (1) Inner product method

[0289] The homogeneity calculation means 2313 calculates the first BEP vector (B 1 ,E1 ,P 1 ) and the second BEP vector (B 2 ,E 2 ,P 2 ) is obtained as a homogeneity. The homogeneity calculation means 2313 may obtain a homogeneity using the above-mentioned inner product as a parameter, for example. The first BEP vector and the second BEP vector have starting points at the same coordinates. (2) Method using the angle θ between two BEP vectors

[0290] The homogeneity calculation means 2313 acquires the homogeneity, for example, as follows. That is, the homogeneity calculation means 2313 acquires the angle θ(x, y, z) between the first BEP vector and the second BEP vector. Next, the angle θ(x, y, z) is used to calculate HM(x, y, z), which is the square of COSθ(x, y, z). Note that HM(x, y, z) is an example of homogeneity.

[0291] Next, the process of the document information prediction means 2314 acquiring the predicted value of document information based on the homogeneity will be described. For example, the document information prediction means 2314 calculates √f1(x,y,z) which is the square root of f1(x,y,z) of the first subject and √f2(x,y,z) which is the square root of f2(x,y,z) of the second subject. Next, the document information prediction means 2314 multiplies HM(x,y,z) by the square root of f1(x,y,z) of the first subject and the square root of f2(x,y,z) of the second subject to calculate the predicted value of document information based on the homogeneity. Alternatively, the document information prediction means 2314 calculates half the sum of f1(x,y,z) of the first subject and f2(x,y,z) of the second subject. Next, the document information prediction means 2314 multiplies HM(x, y, z) by half the sum to calculate a predicted value of the document information based on the homogeneity.

[0292] That is, the homogeneity calculation means 2313 and the document information prediction means 2314 calculate the homogeneity and a predicted value of the document information based thereon, for example, by the following formula 5 or formula 6, respectively.

[0293]

number

[0294]

number

[0295] Next, a case where the document information prediction means 1314 acquires a predicted value of document information based on the homogeneity using a correction process will be described.

[0296] The homogeneity calculation means 2313 obtains the angle θ between the first BEP vector and the second BEP vector, and obtains HM(x,y,z), which is the square of cosθ, using the angle θ. Next, the literature information prediction means 2314 obtains f1(x,y,z) of the first subject, and obtains f2(x,y,z) of the second subject. Next, the literature information prediction means 2314 calculates the product and sum of f1(x,y,z) and f2(x,y,z). Next, if the product is zero and the sum is positive, the literature information prediction means 2314 multiplies HM(x,y,z) by the square root of the sum to obtain a predicted value of the literature information based on the homogeneity, and if the product is not zero or the sum is not positive, the literature information prediction means 2314 multiplies HM(x,y,z) by the square root of the product to obtain a predicted value of the literature information based on the homogeneity.

[0297] That is, the homogeneity calculation means 2313 and the document information prediction means 2314 calculate the homogeneity and a predicted value of the document information based thereon by the following formula 7.

[0298]

number

[0299] Next, the above-mentioned processing of the heterogeneity calculation means 2312 and homogeneity calculation means 2313 will be explained. In other words, for the literature information of two subjects (individuals or groups), by calculating the square of SINθ(x,y,z) of the BEP vectors of each of the two subjects, the heterogeneity factor HT(x,y,z) can be found, and similarly, by calculating the square of COSθ(x,y,z), the homogeneity factor HM(x,y,z) can be found.

[0300] Furthermore, by multiplying the two geometric means or arithmetic means, it is possible to calculate a predicted value for the literature information, which is the possibility of intellectual property being created through joint research between the two subjects.

[0301] When intellectual property creation is a process in which the history of past paper publications, etc. has a multiplicative influence, in other words, when it is the result of a complex system, the geometric mean is multiplied (for example, see the URL "https: / / shinomiya-note.hatenablog.com / entry / 2020 / 07 / 31 / 180828").

[0302] In addition, if the creation of intellectual property is a process in which the history of past paper publications, etc. has an additive effect, the arithmetic mean is multiplied.

[0303] Furthermore, if we simply multiply both √f(x, y, z), (A) for example, if a first individual, etc. has produced many papers at a coordinate point but a second individual, etc. has not produced any papers, the product of the roots will be zero, which will give the same result as (B) when both are zero.

[0304] This means that the coordinate points at which intellectual property will be created in the future are subject to strict restrictions, that is, among the past coordinate points, the range of coordinate points where both (C) are not zero.

[0305] Therefore, in order to alleviate this, in the case of (A), the second individual is deemed to have also produced one paper, and the square root of the sum is multiplied by √fi(x,y,z)×√1=√(f1(x,y,z)+0).

[0306] The output unit 24 outputs various information, such as the type, the evaluation value, the BEP vector, the degree of heterogeneity, the degree of homogeneity, and the predicted value of the document information.

[0307] Here, output is a concept that includes displaying on a display, projecting using a projector, printing on a printer, transmitting to an external device, storing on a recording medium, and passing on the processing results to other processing devices or other programs, etc.

[0308] The BEP vector output unit 241 outputs the BEP vector acquired by the BEP vector acquisition means 2311. It is preferable that the BEP vector output unit 241 outputs the BEP vector in a three-dimensional space. It is preferable that the BEP vector output unit 241 outputs the BEP vector graphically in a three-dimensional space. The output of the graphical vector is usually an output of an arrow.

[0309] The three-dimensional space has three axes: an axis of paper classification according to the JST classification code (e.g., the x-axis), an axis of patent classification according to the IPC (e.g., the y-axis), and an axis of Grant-in-Aid for Scientific Research (funding classification) according to the Grant-in-Aid for Scientific Research classification code (e.g., the z-axis). However, in this specification, the x-, y-, and z-axes are in a left-handed coordinate system.

[0310] It is preferable that the BEP vector output unit 241 outputs a BEP vector for each of two or more candidate coordinates in the three-dimensional space.

[0311] The relationship output unit 242 outputs the heterogeneity acquired by the heterogeneity calculation means 2312. The relationship output unit 242 outputs the homogeneity acquired by the homogeneity calculation means 2313. The relationship output unit 242 may output both the heterogeneity and the homogeneity, or may output only one of them. In addition, the relationship output unit 242 also outputs a predicted value of the document information based on the heterogeneity, and a predicted value of the document information based on the homogeneity.

[0312] The reception unit 22 and the condition reception unit 221 are preferably realized by wireless or wired communication means, but may also be realized by a means for receiving broadcasts, a device driver for an input means such as a touch panel or keyboard, or control software for a menu screen.

[0313] The processing unit 23, the three-dimensional vector calculation unit 231, the BEP vector acquisition means 2311, the homogeneity calculation means 2313, the heterogeneity calculation means 2312, and the document information prediction means 2414 can usually be realized by a processor, a memory, etc. The processing procedure of the processing unit 23, etc. is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, they may also be realized by hardware (dedicated circuit). The processor may be a CPU, MPU, GPU, etc., and the type is not important.

[0314] The output unit 24, the BEP vector output unit 241, and the relation output unit 242 may or may not include an output device such as a display or a speaker. The output unit 24, etc. may be realized by, for example, driver software for an output device, or a combination of driver software for an output device and an output device, etc. The output unit 24, etc. may be realized by wireless or wired communication means.

[0315] Next, an example of the operation of the information processing device 2 will be described with reference to the flowchart of Fig. 16. The information processing device 2 may perform all of the operations performed by the information processing device 1. Therefore, in the flowchart of Fig. 16, only the operations unique to the information processing device 2 will be described.

[0316] (Step S1601) The condition receiving unit 221 judges whether or not a vector output instruction has been received. If a vector output instruction has been received, the process proceeds to step S1602, and if not, the process proceeds to step S1611.

[0317] (Step S1602) The document obtaining section 131 obtains a selection condition included in the vector output instruction.

[0318] (Step S1603) The document obtaining section 131 obtains from the document storage section 111 two or more pieces of document information that match the selection condition.

[0319] (Step S1604) The coordinate obtaining means 1341 obtains all coordinate information paired with each of two or more pieces of target document information.

[0320] (Step S1605) The BEP vector acquisition means 2311 assigns 1 to the counter i.

[0321] (Step S1606) The BEP vector acquisition means 2311 judges whether or not the i-th candidate coordinate exists. If the i-th candidate coordinate exists, the process proceeds to step S1607, and if not, the process returns to step S1601.

[0322] (Step S1607) The BEP vector acquisition means 2311 acquires a BEP vector corresponding to the i-th candidate coordinate. An example of such a BEP vector acquisition process will be described with reference to the flowchart in Fig. 17. Note that the BEP vector here is a vector that corresponds to the document information of the person or group (target person) indicated by the selection condition.

[0323] (Step S1608) The BEP vector acquisition means 2311 judges whether any of the three elements of the BEP vector is not 0 or whether all of the elements are 0. If any of the elements are not 0, the process proceeds to step S1609, and if all of the elements are 0, the process proceeds to step S1610.

[0324] (Step S1609) The BEP vector output unit 241 outputs the BEP vector. The BEP vector output unit 241 graphically outputs the BEP vector in, for example, a three-dimensional space. The BEP vector output unit 241 accumulates the BEP vector, for example, in association with the i-th candidate coordinate.

[0325] (Step S1610) The BEP vector acquisition means 2311 increments the counter i by 1. The process returns to step S1606.

[0326] (Step S1611) The condition receiving unit 221 judges whether or not a predicted value output instruction has been received. If a predicted value output instruction has been received, the process proceeds to step S1612, and if not, the process returns to step S1601.

[0327] (Step S1612) The document obtaining unit 131 obtains the first selection condition included in the predicted value output instruction.

[0328] (Step S1613) The document obtaining section 131 obtains a second selection condition included in the predicted value output instruction.

[0329] (Step S1614) The document obtaining unit 131 obtains, from the document storage unit 111, two or more pieces of document information (first document information) that match the first selection condition.

[0330] (Step S1615) The document obtaining section 131 obtains, from the document storage section 111, two or more pieces of document information (second document information) that match the second selection condition.

[0331] (Step S1616) The coordinate obtaining means 1341 obtains all pieces of coordinate information (first coordinate information) paired with each of the two or more pieces of target first document information.

[0332] (Step S1617) The coordinate obtaining means 1341 obtains all pieces of coordinate information (second coordinate information) paired with each of the two or more pieces of second document information to be processed.

[0333] (Step S1618) The BEP vector acquisition means 2311 assigns 1 to the counter i.

[0334] (Step S1619) The BEP vector acquisition means 2311 judges whether or not the i-th candidate coordinate exists. If the i-th candidate coordinate exists, the process proceeds to step S1620, and if not, the process returns to step S1601.

[0335] (Step S1620) The BEP vector acquisition means 2311 acquires a BEP vector (first BEP vector) corresponding to the i-th candidate coordinate and corresponding to the first selection condition by using the first coordinate information. An operation example of such a first BEP vector acquisition process is the operation described in the flowchart (BEP vector acquisition process) of FIG.

[0336] (Step S1621) The BEP vector acquisition means 2311 acquires a BEP vector (second BEP vector) corresponding to the i-th candidate coordinate and corresponding to the second selection condition by using the second coordinate information. An operation example of such second BEP vector acquisition processing is the operation described in the flowchart (BEP vector acquisition processing) of FIG.

[0337] (Step S1622) The document information prediction means 2314 judges whether or not the information to be acquired, which is indicated by the predicted value output instruction, includes a predicted value of the document information based on the heterogeneity (referred to as a "first predicted value" as appropriate). If the heterogeneity is included, the process proceeds to step S1623, and if the heterogeneity is not included, the process proceeds to step S1626.

[0338] (Step S1623) The document information prediction means 2314 etc. acquires a first predicted value using the first BEP vector and the second BEP vector. An example of such a first predicted value acquisition process will be described with reference to the flowchart of FIG.

[0339] (Step S1624) The relationship output unit 242 determines whether or not the first predicted value is 0. If the first predicted value is 0, the process proceeds to step S1626, and if the first predicted value is not 0, the process proceeds to step S1625. Note that "!=" in FIG. 16 indicates "does not match."

[0340] (Step S1625) The relationship output unit 242 outputs the first predicted value. The relationship output unit 242 outputs the first predicted value in, for example, a three-dimensional space. The relationship output unit 242 accumulates the first predicted value in association with, for example, the i-th candidate coordinate.

[0341] (Step S1626) The document information prediction means 2314 judges whether or not the information to be acquired, which is indicated by the predicted value output instruction, includes a predicted value of document information based on homogeneity (referred to as a "second predicted value" as appropriate). If the second predicted value is included, the process proceeds to step S1627, and if the second predicted value is not included, the process proceeds to step S1630.

[0342] (Step S1627) The document information prediction means 2314 etc. acquires a second predicted value using the first BEP vector and the second BEP vector. An example of such a second predicted value acquisition process will be described with reference to the flowchart of FIG.

[0343] (Step S1628) The relationship output unit 242 determines whether or not the second predicted value is 0. If the second predicted value is 0, the process proceeds to step S1630, and if the second predicted value is not 0, the process proceeds to step S1629.

[0344] (Step S1629) The relationship output unit 242 outputs the second predicted value. The relationship output unit 242 outputs the second predicted value in a three-dimensional space, for example. The relationship output unit 242 accumulates the second predicted value in association with the i-th candidate coordinate, for example.

[0345] (Step S1630) The EP vector acquiring means 2311 increments the counter i by 1. The process returns to step S1619.

[0346] In the flowchart of FIG. 16, the process ends when the power is turned off or an interrupt occurs to end the process.

[0347] Next, an example of the BEP vector acquisition process in step S1607 will be described with reference to the flowchart in FIG.

[0348] (Step S1701) The BEP vector acquisition means 2311 acquires the total number of pieces of document information (Σ) acquired in step S1603.

[0349] (Step S1702) The BEP vector acquisition means 2311 acquires the number (α(x)) of coordinate information having x that matches the coordinate value (x). Note that the coordinate value (x) is “x” among the i-th candidate coordinate (x, y, z) of step S1605.

[0350] (Step S1703) The BEP vector acquisition means 2311 acquires the number (β(y)) of coordinate information having y that matches the coordinate value (y). The coordinate value (y) is “y” among the i-th candidate coordinate (x, y, z) of step S1605.

[0351] (Step S1704) The BEP vector acquisition means 2311 acquires the number (γ(z)) of coordinate information having z that matches the coordinate value (z). The coordinate value (z) is “z” among the i-th candidate coordinates (x, y, z) of step S1605.

[0352] (Step S1705) The BEP vector acquisition means 2311 calculates the B component (α(x) / Σ) that constitutes the BEP vector.

[0353] (Step S1706) The BEP vector acquisition means 2311 calculates the E component (β(y) / Σ) that constitutes the BEP vector.

[0354] (Step S1707) The BEP vector acquisition means 2311 calculates the P component (γ(z) / Σ) that constitutes the BEP vector.

[0355] (Step S1708) The BEP vector acquisition means 2311 constructs a BEP vector (B component, E component, P component), and returns to the upper level process.

[0356] Next, an example of the first predicted value acquisition process in step S1623 will be described with reference to the flowchart in FIG.

[0357] (Step S1801) The heterogeneity calculation means 2312 acquires the angle θ between two BEP vectors (the first BEP vector and the second BEP vector). The two BEP vectors are the BEP vectors acquired in steps S1620 and S1621.

[0358] (Step S1802) The heterogeneity calculation means 2312 calculates the following equation using the angle θ: HT(x, y, z)=sinθ 2 " is calculated.

[0359] (Step S1803) The literature information prediction means 2314 acquires the number of pieces of literature information "f1(x, y, z)" corresponding to the coordinate information (x, y, z) from the literature information of the first subject.

[0360] (Step S1804) The literature information prediction means 2314 acquires the number of pieces of literature information "f2(x, y, z)" corresponding to the coordinate information (x, y, z) from the literature information of the second subject.

[0361] (Step S1805) The document information prediction means 2314 calculates the sum of "f1(x,y,z)" and "f2(x,y,z)".

[0362] (Step S1806) The document information prediction means 2314 calculates the product of "f1(x, y, z)" and "f2(x, y, z)".

[0363] (Step S1807) The document information prediction means 2314 judges whether or not "product=0 and sum!=0" is satisfied. If "product=0 and sum!=0" is satisfied, the procedure goes to step S1808, and if "product=0 and sum!=0" is not satisfied, the procedure goes to step S1809.

[0364] (Step S1808) The document information prediction means 2314 calculates a first predicted value by the formula "First predicted value = HT (x, y, z) x the (1 / 2)th power of the sum calculated in step S1805." The process returns to the upper level process.

[0365] (Step S1809) The document information prediction means 2314 calculates a first predicted value by the formula "First predicted value = HT (x, y, z) x the (1 / 2)th power of the product calculated in step S1806." The process returns to the upper level process.

[0366] Next, an example of the second predicted value acquisition process in step S1627 will be described with reference to the flowchart in FIG.

[0367] (Step S1901) The homogeneity calculation means 2313 obtains the angle θ between two BEP vectors (first BEP vector and second BEP vector).

[0368] (Step S1902) The homogeneity calculation means 2313 calculates the degree of homogeneity by using the angle θ by the formula: HM(x, y, z)=cosθ 2 " is calculated.

[0369] (Step S1903) The literature information prediction means 2314 acquires the number of pieces of literature information "f1(x, y, z)" corresponding to the coordinate information (x, y, z) from the literature information of the first subject.

[0370] (Step S1904) The literature information prediction means 2314 acquires the number of pieces of literature information "f2(x, y, z)" corresponding to the coordinate information (x, y, z) from the literature information of the second subject.

[0371] (Step S1905) The document information prediction means 2314 calculates the sum of "f1(x,y,z)" and "f2(x,y,z)".

[0372] (Step S1906) The document information prediction means 2314 calculates the product of "f1(x, y, z)" and "f2(x, y, z)".

[0373] (Step S1907) The document information prediction means 2314 judges whether or not "product=0 and sum!=0" is satisfied. If "product=0 and sum!=0" is satisfied, the procedure goes to step S1908, and if "product=0 and sum!=0" is not satisfied, the procedure goes to step S1909.

[0374] (Step S1908) The document information prediction means 2314 calculates a second predicted value by the formula "Second predicted value=HM(x, y, z)×(1 / 2)th power of the sum calculated in step S1905." The process returns to the upper level process.

[0375] (Step S1909) The document information prediction means 2314 calculates a second predicted value by the formula "Second predicted value=HM(x, y, z)×(1 / 2)th power of the product calculated in step S1906." The process returns to the upper level process.

[0376] Five specific operation examples of the information processing device 2 in this embodiment will be described below. Specific Example 1 is an example of a process for outputting a BEP vector. Specific Example 2 is an example of a process for outputting a heterogeneity. Specific Example 3 is an example of a process for outputting a homogeneity. Specific Example 4 is an example of a process for outputting a predicted value (second predicted value) of document information based on the homogeneity. Specific Example 5 is an example of a process for outputting a predicted value (first predicted value) of document information based on the heterogeneity.

[0377] (Example 1) It is assumed that the condition receiving unit 221 of the information processing device 2 receives the vector output instruction "Subject=Researcher SS".

[0378] Next, the document acquisition unit 131 acquires the selection condition "subject=researcher SS" contained in the vector output instruction. Next, the document acquisition unit 131 acquires two or more pieces of document information of the researcher SS that match the selection condition from the document storage unit 111. Next, the coordinate acquisition means 1341 acquires all coordinate information that is paired with each of the two or more pieces of document information acquired.

[0379] Note that, using the coordinate information acquired by the coordinate acquiring means 1341, Fig. 20 graphically illustrates the number of coordinate information (the number of document information) corresponding to each of two or more candidate coordinates in the three-dimensional space shown in Fig. 10. In Fig. 20, the size of the figure (sphere) above each candidate coordinate in the three-dimensional space indicates the number of document information on each candidate coordinate.

[0380] Next, the BEP vector acquisition means 2311 acquires the BEP vector of the researcher SS for each of the two or more candidate coordinates by the operation described using the flowchart in Fig. 17. Next, the BEP vector output unit 241 outputs the BEP vector as an arrow in three-dimensional space for each candidate coordinate. An example of such output is shown in Fig. 21. Also, a diagram in which the area 2101 in Fig. 21 is enlarged is shown in 2201 in Fig. 22. In Fig. 22, it can be seen that a BEP vector having a magnitude and direction is arranged in three-dimensional space for each candidate coordinate.

[0381] In other words, in Specific Example 1, it was shown that it is possible to visualize in which classification the literature information of a single individual or group (here, researcher SS) is concentrated as a three-dimensional BEP vector.

[0382] (Example 2) It is assumed that the condition receiving unit 221 of the information processing device 2 has received the degree output instruction "heterogeneity (first subject=researcher SS AND second subject=(researcher KY, researcher TH))".

[0383] Next, the document acquisition unit 131 acquires the first selection condition "first subject=researcher SS" and the second selection condition "second subject=(researcher KY, researcher TH)" contained in the degree output instruction. Next, the document acquisition unit 131 acquires two or more pieces of document information (first document information) that match the first selection condition from the document storage unit 111. In addition, the document acquisition unit 131 acquires two or more pieces of document information (second document information) that match the second selection condition from the document storage unit 111.

[0384] Next, the coordinate acquiring means 1341 acquires all the coordinate information (first coordinate information) that is paired with each of the two or more pieces of target first document information. Also, the coordinate acquiring means 1341 acquires all the coordinate information (second coordinate information) that is paired with each of the two or more pieces of target second document information.

[0385] Fig. 23(a) graphically illustrates the number of pieces of coordinate information (the number of pieces of document information) corresponding to each of two or more candidate coordinates in the three-dimensional space shown in Fig. 10, using the first coordinate information acquired by the coordinate acquisition means 1341. Fig. 23(b) graphically illustrates the number of pieces of coordinate information (the number of pieces of document information) corresponding to each of two or more candidate coordinates in the three-dimensional space shown in Fig. 10, using the second coordinate information acquired by the coordinate acquisition means 1341.

[0386] Next, the BEP vector acquisition means 2311 acquires the BEP vector of the first subject (first BEP vector) for each of two or more candidate coordinates through the operation described using the flowchart in Fig. 17. Also, the BEP vector acquisition means 2311 acquires the BEP vector of the second subject (second BEP vector) for each of two or more candidate coordinates through the operation described using the flowchart in Fig. 17.

[0387] Next, the BEP vector output unit 241 outputs the first BEP vector of the first subject as an arrow on the three-dimensional space for each candidate coordinate. An example of such output is shown in FIG. 24(a). Also, the BEP vector output unit 241 outputs the second BEP vector of the second subject as an arrow on the three-dimensional space for each candidate coordinate. An example of such output is shown in FIG. 24(b).

[0388] Next, the heterogeneity calculation means 2312 determines that the information to be acquired in accordance with the degree output instruction includes "heterogeneity."

[0389] Next, the heterogeneity calculation means 2312 calculates the heterogeneity, which is the cross product of the first BEP vector and the second BEP vector, for each candidate coordinate, by the above-mentioned process.

[0390] Next, the relationship output unit 242 outputs the heterogeneity (here, a vector) acquired by the heterogeneity calculation means 2312 for each candidate coordinate. An output example of such heterogeneity is shown in Fig. 25. Also, an output example of the heterogeneity for each candidate coordinate, which is an enlarged view of the area 2501 in Fig. 25, is shown in 2601 in Fig. 26.

[0391] The heterogeneity calculation means 2312 calculates the sinθ 2 may be calculated to obtain the scalar heterogeneity. The relation output unit 242 may also output the scalar heterogeneity for each candidate coordinate. Such a diagram is shown in FIG. 27. FIG. 27 shows the value of the heterogeneity, which is a scalar at each candidate coordinate, by the difference in the color of the dots on each candidate coordinate (difference in color density). It can be said that FIG. 27 is a display of heterogeneity factors.

[0392] In the above, in the specific example 2, the degree of heterogeneity between the literature information of two subjects is calculated as a three-dimensional vector in the case of the cross product, sinθ 2 It was shown that the case can be visualized as a three-dimensional scalar.

[0393] Heterogeneity can be said to be the degree of difference. Heterogeneity is a parameter that indicates the degree to which three-dimensional vectors differ, and is an important factor in the creation of intellectual property that is born from the differences between the two.

[0394] (Example 3) It is assumed that the condition receiving unit 221 of the information processing device 2 receives the degree output instruction "homogeneity (first subject=researcher TK AND second subject=researcher OY)".

[0395] Next, the document acquisition unit 131 acquires the first selection condition "first subject=researcher TK" and the second selection condition "second subject=researcher OY" contained in the degree output instruction. Next, the document acquisition unit 131 acquires two or more pieces of document information (first document information) that match the first selection condition from the document storage unit 111. In addition, the document acquisition unit 131 acquires two or more pieces of document information (second document information) that match the second selection condition from the document storage unit 111.

[0396] Next, the coordinate acquiring means 1341 acquires all the coordinate information (first coordinate information) that is paired with each of the two or more pieces of target first document information. Also, the coordinate acquiring means 1341 acquires all the coordinate information (second coordinate information) that is paired with each of the two or more pieces of target second document information.

[0397] Fig. 28(a) graphically illustrates the number of pieces of coordinate information (the number of pieces of document information) corresponding to each of two or more candidate coordinates in the three-dimensional space shown in Fig. 10, using the first coordinate information acquired by the coordinate acquisition means 1341. Fig. 28(b) graphically illustrates the number of pieces of coordinate information (the number of pieces of document information) corresponding to each of two or more candidate coordinates in the three-dimensional space shown in Fig. 10, using the second coordinate information acquired by the coordinate acquisition means 1341.

[0398] Next, the BEP vector acquisition means 2311 acquires the BEP vector of the first subject (first BEP vector) for each of two or more candidate coordinates through the operation described using the flowchart in Fig. 17. Also, the BEP vector acquisition means 2311 acquires the BEP vector of the second subject (second BEP vector) for each of two or more candidate coordinates through the operation described using the flowchart in Fig. 17.

[0399] Next, the BEP vector output unit 241 outputs the first BEP vector of the first subject as an arrow on the three-dimensional space for each candidate coordinate. An example of such output is shown in FIG. 29(a). Also, the BEP vector output unit 241 outputs the second BEP vector of the second subject as an arrow on the three-dimensional space for each candidate coordinate. An example of such output is shown in FIG. 29(b).

[0400] Next, the homogeneity calculation means 2313 judges that the information to be acquired in accordance with the degree output instruction includes "homogeneity."

[0401] Next, the homogeneity calculation means 2313 calculates, by the above-mentioned process, for each candidate coordinate, cosθ 2 The homogeneity is calculated as:

[0402] Next, the relation output unit 242 outputs the homogeneity (here, a scalar) for each candidate coordinate. Such a diagram is shown in Fig. 30. Fig. 30 shows the value of the homogeneity, which is a scalar, at each candidate coordinate by the difference in dot color (difference in color density) on each candidate coordinate. It can be said that Fig. 30 is a display of homo factors.

[0403] In the above, Specific Example 3 shows that the degree of homogeneity between the literature information of two subjects can be visualized as a three-dimensional scalar.

[0404] Here, homogeneity is a parameter that indicates the degree to which three-dimensional vectors are similar, and is an important factor in the creation of intellectual property that arises from the similarity between the two.

[0405] (Example 4) In specific example 4, a process is described in which the literature information prediction means 2314 uses the heterogeneity for each candidate coordinate between researcher SS and the group to which researcher KY and researcher TH belong obtained by the process in specific example 2 to further obtain a predicted value (first predicted value) of literature information based on the heterogeneity for each candidate coordinate.

[0406] The document information prediction means 2314 calculates a first prediction value for each candidate coordinate by using the heterogeneity (HT(x, y, z)) calculated by the heterogeneity calculation means 2312, in accordance with the operation described using the flowchart in FIG. 18 (according to Equation 4).

[0407] Next, the relationship output unit 242 outputs the first predicted value for each candidate coordinate. Here, the relationship output unit 242 outputs a sphere of a size that specifies the acquired first predicted value for each candidate coordinate in a three-dimensional space. Such a diagram is shown in FIG. 31(a). FIG. 31(a) shows that when researcher SS and a group to which researcher KY and researcher TH belong conduct joint research, it is predicted that results indicated by the size of the sphere will be obtained in the technical field of the candidate coordinate where the large sphere is placed.

[0408] FIG. 31(b) shows the number of papers produced for each candidate coordinate as a result of the actual joint research between researcher SS and the group to which researcher KY and researcher TH belong, as indicated by the size of the sphere. Comparing FIG. 31(a) with FIG. 31(b), it can be seen that the two figures are very similar. Although a detailed explanation is omitted, the mean absolute error of the Bohr component was 0.022, the mean absolute error of the Edison component was 0.018, and the mean absolute error of the Pasteur component was 0.017. In other words, according to FIG. 31(a) and (b), it can be seen that the first predicted value obtained by the literature information prediction means 2314 is very close to the actual data. However, it goes without saying that the papers by researcher SS, researcher KY, and researcher TH, which were the basis for calculating the BEP vector, were published before the actual joint research began.

[0409] (Example 5) In specific example 5, a process is described in which the literature information prediction means 2314 uses the homogeneity for each candidate coordinate between researcher TK and researcher OY obtained by the process in specific example 3 to further obtain a predicted value (second predicted value) of literature information based on the homogeneity for each candidate coordinate.

[0410] The document information prediction means 2314 calculates a second predicted value for each candidate coordinate using the homogeneity (HM(x, y, z)) calculated by the homogeneity calculation means 2313, in accordance with the operation described using the flowchart in Figure 19 (according to Equation 7).

[0411] Next, the relationship output unit 242 outputs the second predicted value for each candidate coordinate. Here, the relationship output unit 242 outputs a sphere of a size that specifies the acquired second predicted value for each candidate coordinate in a three-dimensional space. Such a diagram is shown in FIG. 32(a). FIG. 32(a) shows that when researcher TK and researcher OY conduct joint research, it is predicted that the results indicated by the size of the sphere will be obtained in the technical field of the candidate coordinate where the large sphere is placed.

[0412] FIG. 32(b) shows the number of papers produced for each candidate coordinate as a result of the actual joint research between researcher TK and researcher OY, as indicated by the size of the sphere. Comparing FIG. 32(a) with FIG. 32(b), it can be seen that the two diagrams are very similar. Although a detailed explanation is omitted, the mean absolute error of the Bohr component was 0.032, the mean absolute error of the Edison component was 0.041, and the mean absolute error of the Pasteur component was 0.047. In other words, according to FIG. 32(a)(b), it can be seen that the second predicted value obtained by the literature information prediction means 2314 is very close to the actual data. However, it goes without saying that the papers by researcher TK and researcher OY, which were the basis for calculating the BEP vector, were published before the actual joint research began.

[0413] The above specific examples 4 and 5 can be interpreted as follows. In other words, it is easy to understand if we analogize the BEP vector of researcher A as a magnetic field and the BEP vector of researcher B as the flow velocity of an electrically charged plasma fluid.

[0414] The above specific example 4 can be interpreted as follows. That is, when Researcher B's plasma fluid enters Researcher A's magnetic field, Researcher B's plasma fluid is subjected to the Lorentz force. This Lorentz force does not affect the plasma flow velocity when the direction of Researcher A's magnetic field vector and the direction of Researcher B's plasma flow velocity vector coincide, that is, when the angle between the two vectors is θ and SINθ=0. However, when the directions are different, that is, when the angle θ is not 0, the velocity is affected according to SINθ. This Lorentz force is considered to be the power of creating intellectual property based on heterogeneous combinations. Intellectual property based on heterogeneous combinations can be, for example, basic inventions such as interdisciplinary fusion, which have the potential to cause innovation.

[0415] The interpretation of the above specific example 5 is as follows. If the BEP vector of researcher A and the BEP vector of researcher B are roughly in the same direction, it is believed that intellectual property based on a homogeneous combination will be created. Intellectual property based on a homogeneous combination is, for example, an improvement invention, an applied invention, etc., which can be considered to contribute to gradual technological progress.

[0416] The inventions corresponding to concrete examples 4 and 5 visualize the possibility of creating these two intellectual properties on three-dimensional coordinates, the former calculating the square of sine theta (called the hetero factor HT) with the angle between the two vectors being θ, and the latter calculating the square of cosine theta (called the homo factor HM). Note that the sum of the two is clear from the mathematical formula, so the former and latter have a complementary relationship.

[0417] As described above, according to this embodiment, a vector relating to the type of individual or group can be obtained.

[0418] Furthermore, according to this embodiment, the vectors of two subjects, which are individuals or groups, can be used to obtain the degree of heterogeneity or homogeneity of the two subjects.

[0419] Furthermore, according to this embodiment, the vectors of two subjects, which are individuals or groups, can be used to obtain the degree of heterogeneity or homogeneity of the two subjects.

[0420] Furthermore, according to this embodiment, it is possible to obtain a predicted value of literature information based on the heterogeneity or homogeneity of two subjects by using vectors of two subjects, which are individuals or groups. It can be said that such a predicted value is the degree of possibility of intellectual property creation through collaborative research between the two subjects.

[0421] Fig. 33 shows the appearance of a computer that executes the program described in this specification to realize the information processing device 1 and the like of the various embodiments described above. The above-mentioned embodiments can be realized by computer hardware and a computer program executed thereon. Fig. 33 is an overview of this computer system 300, and Fig. 34 is a block diagram of the system 300.

[0422] In FIG. 33, computer system 300 includes a computer 301 including a CD-ROM drive, a keyboard 302, a mouse 303, and a monitor 304.

[0423] 34, computer 301 includes, in addition to CD-ROM drive 3012, MPU 3013, bus 3014 connected to CD-ROM drive 3012 etc., ROM 3015 for storing programs such as a boot-up program, RAM 3016 connected to MPU 3013 for temporarily storing instructions of application programs and providing temporary storage space, and hard disk 3017 for storing application programs, system programs, and data. Although not shown here, computer 301 may further include a network card for providing connection to a LAN.

[0424] A program for causing computer system 300 to execute the functions of information processing device 1 and the like of the above-mentioned embodiment may be stored on CD-ROM 3101, inserted into CD-ROM drive 3012, and further transferred to hard disk 3017. Alternatively, the program may be transmitted to computer 301 via a network (not shown) and stored on hard disk 3017. The program is loaded into RAM 3016 at the time of execution. The program may be loaded directly from CD-ROM 3101 or the network.

[0425] The program does not necessarily include an operating system (OS) or a third party program that causes the computer 301 to execute the functions of the information processing device 1 of the above-mentioned embodiment. The program only needs to include an instruction portion that calls appropriate functions (modules) in a controlled manner to obtain a desired result. How the computer system 300 operates is well known, and a detailed description will be omitted.

[0426] In addition, in the above program, the steps of transmitting information and receiving information do not include processing performed by hardware, such as processing performed by a modem or interface card in the transmitting step (processing that is performed only by hardware).

[0427] The program may be executed by a single computer or a plurality of computers. That is, the program may be executed by a centralized processing or a distributed processing.

[0428] Furthermore, in each of the above embodiments, it goes without saying that two or more communication means present in one device may be physically realized by one medium.

[0429] In each of the above embodiments, each process may be realized by centralized processing in a single device, or may be realized by distributed processing in a plurality of devices.

[0430] The present invention is not limited to the above-described embodiment, and various modifications are possible, and it goes without saying that these modifications are also included within the scope of the present invention. [Industrial Applicability]

[0431] As described above, the information processing device 1 according to the present invention has an effect of being able to acquire the type of an individual or group by using literature information, and is useful as an information processing device for promoting research. [Explanation of symbols]

[0432] 1, 2 Information processing device 11 Storage area 12, 22 Reception 13, 23 Processing section 14, 24 Output section 111 Document Storage Section 112 Coordinate storage unit 121, 221 Conditions Reception Department 131 Literature Acquisition Department 132 Code Assignment Department 133 Coordinate storage unit 134 Type determination section 135 Evaluation Department 141 type output section 142 Evaluation output unit 241 BEP Vector Output Unit 242 Related Output Section 1341 Coordinate acquisition means 1342 Degree acquisition means 1343 Quasi-Stokes type determination method 2311 BEP Vector Acquisition Method 2312 Heterogeneity calculation method 2313 Homogeneity calculation method 2314 Document Reference Prediction Means

Claims

1. a type determination unit which acquires a paper code concentration degree based on αmax, which is the maximum value of the number of most frequently appearing codes among the paper codes paired with two or more pieces of literature information which are paired with one or more attribute values ​​related to the literature information, and acquires a patent code concentration degree based on βmax, which is the maximum value of the number of most frequently appearing codes among the patent codes paired with the two or more pieces of literature information, and acquires a competitive fund code concentration degree based on γmax, which is the maximum value of the number of most frequently appearing codes among the competitive fund codes paired with the two or more pieces of literature information, and determines a type corresponding to a paper set which is the two or more pieces of literature information, using the paper code concentration degree, the patent code concentration degree, and the competitive fund code concentration degree; and a type output unit that outputs the type.

2. Further comprising a coordinate storage unit that acquires coordinate information (x, y, z) for each of the two or more pieces of literature information, where x is a value corresponding to the paper code paired with the literature information, y is a value corresponding to the patent code paired with the literature information, and z is a value corresponding to the competitive fund code paired with the literature information, and stores the coordinate information (x, y, z) in a coordinate storage unit in association with each of the two or more pieces of literature information; The type determination unit is a coordinate acquisition means for acquiring one or more pieces of coordinate information corresponding to each of the two or more pieces of document information from the coordinate storage unit; The number f(x, y, z) of each piece of coordinate information acquired by the coordinate acquisition means is acquired; Let αmax, which is the maximum value of ∫∫f(x,y,z)dydz≡α(x), be the paper code concentration, The maximum value of ∫∫f(x, y, z)dxdz≡β(y) is defined as βmax, which is the patent code concentration. A degree acquisition means for determining γmax, which is the maximum value of ∫∫f(x,y,z)dydz≡γ(z), as the competitive fund code concentration degree; and a quasi-Stokes type determination means for determining the type as a quasi-Bohr type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=paper code concentration, the type as a quasi-Edison type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=patent code concentration, and the type as a quasi-Pasteur type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=competitive funding code concentration.

3. In one or more of the three types of codes, in addition to a first classification code, one or more sub-classification codes are associated with at least one or more pieces of document information among the two or more pieces of document information; a score is associated with one or more of the paper code, the patent code, and the competitive funding code corresponding to each of the two or more pieces of literature information; a coordinate storage unit that acquires, for each of one or more pieces of document information among the two or more pieces of document information, first coordinate information, which is coordinate information using only the first classification code of the three types of codes, and one or more pieces of sub-coordinate information, which is coordinate information using at least one sub-classification code, and stores the first coordinate information and the one or more pieces of sub-coordinate information in the coordinate storage unit in association with each of the one or more pieces of document information; The type determination unit is a coordinate acquisition means for acquiring two or more pieces of coordinate information, which are primary coordinate information and one or more secondary coordinate information corresponding to each of the one or more pieces of document information; For a code type that does not have a score associated therewith, a total number of first classification codes and a total number of second classification codes corresponding to each of the one or more pieces of document information are obtained, and a code concentration degree is obtained based on the total number of the first classification codes and the second classification codes; said degree acquisition means determining a minimum score to be adopted so that the difference between the total number of codes of a type with associated scores and the total number of codes of a type with no associated scores satisfies an allowable condition, acquiring the total number of top classification codes and the total number of sub classification codes using only coordinate information corresponding to codes with associated scores equal to or greater than said minimum value, and acquiring said code concentration degree based on both totals; and a quasi-Stokes type determination means for determining the type as a quasi-Bohr type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=paper code concentration, the type as a quasi-Edison type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=patent code concentration, and the type as a quasi-Pasteur type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=competitive funding code concentration.

4. a condition receiving unit that receives a selection condition, which is a condition related to one or more attribute values ​​related to the document information and is a condition for identifying one individual or one group; a coordinate storage unit that acquires coordinate information (x, y, z) for each of the two or more pieces of literature information, where x is a value corresponding to the paper code paired with the literature information, y is a value corresponding to the patent code paired with the literature information, and z is a value corresponding to the competitive fund code paired with the literature information, and stores the coordinate information (x, y, z) in a coordinate storage unit in association with each of the two or more pieces of literature information; The type determination unit is a coordinate acquisition means for acquiring, from the coordinate storage unit, one or more pieces of coordinate information corresponding to two or more pieces of document information of the one individual or the one group that satisfies the selection condition; Each of two or more individuals or groups including the one individual or the one group The maximum share of paper codes, patent codes, or competitive funding codes (αmax / Σ, βmax / Σ, γmax / Σ, where Σ≡total number of publications by an individual or group), or calculating a standard deviation of the maximum share rate of each of the three types of codes of the one individual or the one group using the maximum share rate of each of the three types of codes of the two or more individuals or each of the two or more groups; a degree acquisition means for acquiring the paper code concentration degree, the patent code concentration degree, and the competitive fund code concentration degree based on the maximum share rate or the deviation value; and a quasi-Stokes type determination means for determining the type as a quasi-Bohr type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=paper code concentration, the type as a quasi-Edison type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=patent code concentration, and the type as a quasi-Pasteur type if MAX(paper code concentration, patent code concentration, competitive funding code concentration)=competitive funding code concentration.

5. The type determination unit is determining the type of each of two or more individuals in a group; an evaluation unit that obtains an evaluation value of the one group by using the type of each of the two or more individuals; an evaluation value output unit that outputs the evaluation value, The evaluation unit is The information processing device according to claim 1 , wherein when the types of the two or more individuals are the same, a lower evaluation value is acquired compared to when the types are different.

6. an evaluation unit that determines an intersection line ab between a plane a corresponding to a first type determined by the MAX (paper code concentration, patent code concentration, competitive funding code concentration) of one group and a plane b corresponding to a second type determined by the MAX (paper code concentration, patent code concentration, competitive funding code concentration) of another group, obtains coordinate information on the intersection line ab, which is a number, a share rate, or a standard deviation for each piece of coordinate information corresponding to both the one group and the other group, and obtains a larger evaluation value as the number, share rate, or standard deviation is larger; The information processing apparatus according to claim 2 , further comprising an evaluation value output unit that outputs the evaluation value.

7. In addition to or instead of the type determination unit, a three-dimensional vector calculation unit is provided, In addition to or instead of the type output unit, a BEP vector output unit is provided; The three-dimensional vector calculation unit is a coordinate acquisition means for acquiring, from the coordinate storage unit, one or more pieces of coordinate information (x, y, z) corresponding to two or more pieces of document information of the one individual or the one group that satisfies the selection condition; a BEP vector acquisition means for calculating α(x) / Σ=∫∫f(x,y,z)dydz / Σ, β(y) / Σ=∫∫f(x,y,z)dxdz / Σ, and γ(z) / Σ=∫∫f(x,y,z)dydx / Σ for each of two or more individuals or groups including the one individual or the one group, using the one or more pieces of coordinate information (x,y,z), calculating normalized values ​​b(x), e(y), and p(z) which are normalized values ​​of the α(x) / Σ, the β(y) / Σ, and the γ(z) / Σ, and acquiring a BEP vector (B,E,P) which is a three-dimensional vector whose start point is (x,y,z), and whose b(x) is a B component, whose e(y) is an E component, and whose p(z) is a P component; The BEP vector output unit 5. The information processing apparatus according to claim 4, wherein said BEP vector acquisition means outputs the acquired BEP vector.

8. The condition receiving unit: accepting a first selection condition, the first selection condition being a condition for identifying a first subject, the first individual or the first group, and a second selection condition, the second selection condition being a condition for identifying a second subject, the second individual or the second group; The coordinate acquisition means obtain, from the coordinate storage unit, a first coordinate set which is one or more pieces of coordinate information corresponding to two or more pieces of document information of the first subject who satisfies the first selection condition, and obtain a second coordinate set which is one or more pieces of coordinate information corresponding to two or more pieces of document information of the second subject who satisfies the second selection condition; The BEP vector acquisition means Using the first coordinate set, a first BEP vector (B1, E1, P1) is obtained, and using the second coordinate set, a second BEP vector (B2, E2, P2) is obtained; The three-dimensional vector calculation unit is a heterogeneity calculation means for acquiring a heterogeneity which is a degree of difference between the first BEP vector (B1, E1, P1) and the second BEP vector (B2, E2, P2) whose starting points are at the same coordinates, or a homogeneity calculation means for acquiring a homogeneity which is a degree of similarity between the first BEP vector (B1, E1, P1) and the second BEP vector (B2, E2, P2) whose starting points are at the same coordinates, The information processing apparatus according to claim 7 , further comprising a relation output unit for outputting the degree of heterogeneity or the degree of homogeneity.

9. The three-dimensional vector calculation unit includes a document information prediction means, The literature information prediction means includes: As the angle θ between the first BEP vector and the second BEP vector, whose starting points are at the same coordinates, HT(x, y, z) which is the square of sine θ is multiplied by √f1(x, y, z) which is the square root of f1(x, y, z) of the first subject and √f2(x, y, z) which is the square root of f2(x, y, z) of the second subject, or by half the sum of f1(x, y, z) of the first subject and f2(x, y, z) of the second subject, to obtain a predicted value of the literature information based on the heterogeneity; or where the angle θ between the first BEP vector and the second BEP vector, whose starting points are at the same coordinates, is the square of cosθ, and HM(x, y, z) is multiplied by √f1(x, y, z) which is the square root of f1(x, y, z) of the first subject and √f2(x, y, z) which is the square root of f2(x, y, z) of the second subject, or by half the sum of f1(x, y, z) of the first subject and f2(x, y, z) of the second first subject, to obtain a predicted value of the document information based on the homogeneity, The relationship output unit is 9. The information processing apparatus according to claim 8, wherein said document information prediction means outputs a predicted value of said acquired document information.

10. The literature information prediction means includes: With the angle θ between the first BEP vector and the second BEP vector, which have the same starting point coordinate, being HT(x, y, z), which is the square of sine θ, f1(x, y, z) of the first subject is obtained, f2(x, y, z) of the second subject is obtained, and their product and sum are calculated. If the product is zero and the sum is positive, multiply by the square root of the sum, otherwise multiply by the square root of the product to obtain a predicted value of the document information based on the heterogeneity; or 10. An information processing device according to claim 9, wherein the angle θ between the first BEP vector and the second BEP vector, whose starting points are at the same coordinates, is HM(x, y, z) which is the square of cos θ, and f1(x, y, z) of the first subject is obtained, f2(x, y, z) of the second subject is obtained, and the product and sum of these are calculated, and if the product is zero and the sum is positive, the square root of the sum is multiplied, and if not, the square root of the product is multiplied to obtain a predicted value of the document information based on the homogeneity.

11. An information processing method realized by a type determination unit and a type output unit, a type determination step in which the type determination unit obtains a paper code concentration degree based on αmax, which is the maximum number of codes that appear most frequently among the paper codes that are paired with two or more pieces of literature information and paired with one or more attribute values ​​related to the literature information, and obtains a patent code concentration degree based on βmax, which is the maximum number of codes that appear most frequently among the patent codes that are paired with the two or more pieces of literature information, and obtains a competitive fund code concentration degree based on γmax, which is the maximum number of codes that appear most frequently among the competitive fund codes that are paired with the two or more pieces of literature information, and determines a type corresponding to a paper set that is the two or more pieces of literature information using the paper code concentration degree, the patent code concentration degree, and the competitive fund code concentration degree; a type output step in which the type output unit outputs the type.

12. Computer, a type determination unit which acquires a paper code concentration degree based on αmax, which is the maximum value of the number of most frequently appearing codes among the paper codes paired with two or more pieces of literature information which are paired with one or more attribute values ​​related to the literature information, and acquires a patent code concentration degree based on βmax, which is the maximum value of the number of most frequently appearing codes among the patent codes paired with the two or more pieces of literature information, and acquires a competitive fund code concentration degree based on γmax, which is the maximum value of the number of most frequently appearing codes among the competitive fund codes paired with the two or more pieces of literature information, and determines a type corresponding to a paper set which is the two or more pieces of literature information, using the paper code concentration degree, the patent code concentration degree, and the competitive fund code concentration degree; A program for causing the type output unit to function as a type output unit that outputs the type.

Images