Information processing device, information processing method, and program

Abstract

To provide an information processing apparatus, an information processing method, and a program capable of analyzing classification in three-dimensions for a set of technical documents.SOLUTION: An information processing apparatus includes: a coordinate information accumulation unit 32 that accesses a document information storage unit 11 that stores two or more pieces of document information paired with a classification code having three types of sequences, acquires a value corresponding to each of classification code, acquires three-dimensional coordinate information with such values as elements, and accumulates the three-dimensional coordinate information in association with the document information; a condition reception unit 21 that receives a selection condition; a coordinate information acquisition unit 33 that acquires the three-dimensional coordinate information in association with one or more pieces of document information paired with one or more attribute values that match the selection condition; an image configuration unit 34 that configures a three-dimensional integer grid space image using the three-dimensional coordinate information; and an image output unit 41 that outputs the three-dimensional integer grid space image. The information processing apparatus facilitates an analysis of classification in three-dimensions for a set of technical documents.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information processing device or the like that creates a three-dimensional map for a collection of documents that are paired with classification codes. [Background technology]

[0002] Previously, there was a technology that used classification codes and the like to create a two-dimensional patent map from a collection of patent information (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] WO2020 / 138312 publication Summary of the Invention [Problem to be solved by the invention]

[0004] However, the prior art does not allow for three-dimensional classification analysis of collections of technical documents. [Means for solving the problem]

[0005] The information processing device of the first invention accesses a document information storage unit in which a first classification code, a second classification code, a third classification code, and two or more document information pairs with one or more attribute values ​​related to documents are stored, acquires three-dimensional coordinate information (x, y, z) in which a value corresponding to the first classification code paired with each of the two or more document information is x, a value corresponding to the second classification code paired with each of the document information is y, and a value corresponding to the third classification code paired with each of the document information is z, and stores the three-dimensional coordinate information (x, y, z) in the three-dimensional coordinate storage unit for each of the two or more document information in association with the document information; a condition receiving unit receiving selection conditions related to one or more attribute values ​​related to the documents; a coordinate information acquiring unit acquiring three-dimensional coordinate information corresponding to each of the one or more document information pairs with one or more attribute values ​​that match the selection conditions from the three-dimensional coordinate storage unit; the information processing device includes an image construction unit that uses three-dimensional coordinate information to obtain a first document number, which is the number of three-dimensional coordinate information for one or more (y, z), and constructs a YZ plane visually represented by the first document number corresponding to one or more points (y, z) on the YZ plane; obtains a second document number, which is the number of three-dimensional coordinate information for one or more (x, z), and constructs an XZ plane visually represented by the second document number corresponding to one or more points (x, z) on the XZ plane; obtains a third document number, which is the number of three-dimensional coordinate information for one or more (x, y), and constructs an XY plane visually represented by the third document number corresponding to one or more points (x, y) on the XY plane; and constructs a three-dimensional integer grid space image in which the YZ plane, XZ plane, and XY plane are adjacent to each other; and an image output unit that outputs the three-dimensional integer grid space image.

[0006] Such a configuration allows for three-dimensional classification analysis of collections of technical documents.

[0007] The information processing device of the second invention also accesses a document information storage unit in which a first classification code, a second classification code, a third classification code, and two or more document information pairs with one or more attribute values ​​related to documents are stored, and acquires three-dimensional coordinate information (x, y, z) in which a value corresponding to the first classification code paired with each of the two or more document information is x, a value corresponding to the second classification code paired with each of the document information is y, and a value corresponding to the third classification code paired with each of the document information is z, and for each of the two or more document information, it stores the three-dimensional coordinate information (x, y, z) in the three-dimensional coordinate storage unit in association with the document information; a condition receiving unit receiving selection conditions related to one or more attribute values ​​related to documents; and a condition receiving unit receiving from the three-dimensional coordinate storage unit three-dimensional coordinate information corresponding to each of the one or more document information paired with one or more attribute values ​​that match the selection conditions. an image construction unit that uses the three-dimensional coordinate information corresponding to each of the one or more pieces of document information that meet a selection condition to integrate the number of pieces of three-dimensional coordinate information f(x, y, z) along the X axis to obtain a first number of documents ∫f(x, y, z)dx=p(y, z) and construct a YZ plane, integrates the number of pieces of three-dimensional coordinate information f(x, y, z) along the Y axis to obtain a second number of documents ∫f(x, y, z)dy=q(x, z) and constructs an XZ plane, and integrates the number of pieces of three-dimensional coordinate information f(x, y, z) along the Z axis to obtain a third number of documents ∫f(x, y, z)dz=r(x, y) and constructs an XY plane, and constructs a three-dimensional integer grid space image in which the YZ plane, the XZ plane, and the XY plane are orthogonal to each other; and an image output unit that outputs the three-dimensional integer grid space image.

[0008] Such a configuration allows for three-dimensional classification analysis of collections of technical documents.

[0009] Furthermore, the information processing device of the third invention is an information processing device in which, compared to the first or second invention, in one or more classifications of three types of classifications, in addition to a primary classification code, one or more secondary classification codes are associated with at least one or more pieces of document information out of two or more pieces of document information, the coordinate information accumulation unit acquires, for each of one or more pieces of document information out of two or more pieces of document information, primary coordinate information which is three-dimensional coordinate information using only the primary classification code of the three types of classifications and one or more secondary coordinate information which is three-dimensional coordinate information using at least one secondary classification code, and stores the primary coordinate information and the one or more secondary coordinate information in the three-dimensional coordinate storage unit in association with each of the one or more pieces of document information, and the coordinate information acquisition unit acquires two or more pieces of three-dimensional coordinate information which are the primary coordinate information and the one or more secondary coordinate information associated with each of the one or more pieces of document information that are paired with one or more attribute values ​​that match the selection conditions.

[0010] Such a configuration allows for a more accurate analysis of the three-dimensional classification of a collection of technical documents using both primary and secondary classification codes.

[0011] Furthermore, the information processing device of the fourth invention is an information processing device in which, with respect to any one of the first to third inventions, a score is associated with one or more types of classification code from among the first classification code, the second classification code, and the third classification code corresponding to each of two or more pieces of document information, and the coordinate information acquisition unit acquires three-dimensional coordinate information using only the classification code corresponding to the score that satisfies the extraction conditions for the types of classification code associated with the score.

[0012] With this configuration, for scored classification codes, by using the scores to use only appropriate classification codes, it is possible to perform appropriate analysis of classification for a collection of technical documents.

[0013] Furthermore, the information processing device of the fifth invention is an information processing device in which, compared to the fourth invention, the coordinate information acquisition unit acquires, for a classification code type not associated with a score, the total number of sub-classification codes corresponding to each document information of a document information set, which is two or more document information stored in the document information storage unit, or document information paired with one or more attribute values ​​that match the selection conditions, and uses a first parameter acquired using the total number to acquire a score threshold such that the difference between the parameter for the classification code associated with a score and a second parameter acquired in the same manner as the first parameter is smaller than a predetermined condition, and acquires three-dimensional coordinate information using only sub-classification codes corresponding to scores equal to or greater than the threshold.

[0014] With this configuration, when scores are associated with one or two types of classification codes, the scores can be used to use only appropriate classification codes for the scored classification codes, thereby enabling appropriate analysis of the classification of a collection of technical documents.

[0015] Furthermore, the information processing device of the sixth invention is an information processing device in which, compared to the third invention, the image construction unit comprises a planar image construction means for constructing a planar image having one or more directed line segments, starting from a point indicated by the primary coordinate information and ending at a point indicated by the secondary coordinate information, using primary coordinate information and secondary coordinate information on a cutting plane of a three-dimensional integer grid space, the primary coordinate information and secondary coordinate information corresponding to one or more pieces of document information, and the image output unit outputs the planar image.

[0016] With this configuration, analysis can be performed on the classification of a collection of technical documents using the primary coordinate information and secondary coordinate information.

[0017] Furthermore, the information processing device of the seventh invention is an information processing device in which, compared to the sixth invention, at least one of three types of classification codes paired with each of two or more pieces of document information is associated with a score, and the planar image construction means uses the score corresponding to the classification code to obtain a weight for each of one or more directed line segments, and constructs a planar image having directed line segments whose weights are visually recognizable.

[0018] With this configuration, a more precise analysis of classification of a collection of technical documents can be performed using the primary coordinate information, secondary coordinate information, and score. [Effects of the Invention]

[0019] The information processing device according to the present invention allows for three-dimensional classification analysis of a collection of technical documents. [Brief explanation of the drawings]

[0020] [Figure 1] Block diagram of information processing device A according to the first embodiment. [Figure 2] A flowchart illustrating an example of the operation of the information processing device A [Figure 3] Flowchart illustrating an example of coordinate information acquisition processing [Figure 4] Flowchart illustrating an example of extraction condition acquisition processing [Figure 5] Flowchart illustrating an example of the stereoscopic image construction process [Figure 6] Flowchart illustrating an example of coplanar image construction processing [Figure 7] FIG. 10 shows the coordinate information management table. [Figure 8] Figure showing an example of the bird's-eye view output [Figure 9] Figure 10 shows an example of the output of the same three-dimensional integer grid space image. [Figure 10] Figure showing an example of the bird's-eye view output [Figure 11] Figure 10 shows an example of the output of the same three-dimensional integer grid space image. [Figure 12] Figure showing an example of the bird's-eye view output [Figure 13] Figure 10 shows an example of the output of the same three-dimensional integer grid space image. [Figure 14] Figure showing an example of the bird's-eye view output [Figure 15] Figure 10 shows an example of the output of the same three-dimensional integer grid space image. [Figure 16]Figure 10 shows an example of output of a same-plane image. [Figure 17] Figure 10 shows an example of output of a same-plane image. [Figure 18] Overview of the computer system [Figure 19] Block diagram of the computer system DETAILED DESCRIPTION OF THE INVENTION

[0021] Hereinafter, embodiments of an information processing device and the like will be described with reference to the drawings. Note that components with the same reference numerals in the embodiments perform similar operations, and therefore repeated description may be omitted.

[0022] (Embodiment 1) In this embodiment, an information processing device is described that acquires three-dimensional coordinate information for a document information set corresponding to three ordered types of classification codes and one or more attribute values, using the order of the three types of classification codes corresponding to each document information, and associates the three-dimensional coordinate information with each document information. Note that, as will be described in detail later, a three-dimensional space that can be configured using three ordered types of classification codes is one in which the classification codes are discretely mapped into three dimensions while maintaining the order, so it is practically convenient to think of it as a three-dimensional integer grid space (a lattice space of integer coordinates). Therefore, in this specification, this will be referred to as a three-dimensional integer grid space.

[0023] In this embodiment, an information processing device is described that determines two or more pieces of document information that match selection conditions using one or more attribute values, and uses three-dimensional coordinate information paired with each of the determined two or more pieces of document information to construct a map using two-dimensional coordinates corresponding to the two or more pieces of document information for each combination of two of three types of classification codes, and then constructs and outputs a three-dimensional integer grid space image having three orthogonal planes. Note that in this embodiment, each of the three types of classification codes corresponding to the document information may include a primary classification code, which is the most important and usually placed at the top, as well as a secondary classification code, which is a supplementary classification other than the primary classification code and is usually placed after the primary classification code. The three-dimensional integer grid space image is also referred to as a stereoscopic image, as appropriate.

[0024] In this embodiment, a score is associated with each of one or more of the three classification codes corresponding to the document information, and only the classification code corresponding to the score that satisfies the extraction condition is used. The score may be considered to be the likelihood, probability, likelihood, etc., that the classification code corresponds to the document information. The score is, for example, a value output by a machine learning module through the machine learning prediction process described below.

[0025] In this embodiment, an information processing device that constructs a planar image in which directed line segments are expressed using primary coordinate information and secondary coordinate information on a cross section of a three-dimensional integer grid space will be described.

[0026] Furthermore, in this embodiment, we will describe an information processing device that obtains the weight of a directed line segment using a score corresponding to a classification code corresponding to primary coordinate information and secondary coordinate information, and constructs a planar image having a directed line segment whose weight is visually recognizable.

[0027] 1 is a block diagram of an information processing device A according to this embodiment. The information processing device A includes a storage unit 1, a reception unit 2, a processing unit 3, and an output unit 4. The storage unit 1 includes a document information storage unit 11 and a three-dimensional coordinate storage unit 12. The reception unit 2 includes a condition reception unit 21. The processing unit 3 includes a classification code assignment unit 31, a coordinate information accumulation unit 32, a coordinate information acquisition unit 33, and an image construction unit 34. The image construction unit 34 includes a three-dimensional image construction means 341 and a two-dimensional image construction means 342. The output unit 4 includes an image output unit 41.

[0028] Various types of information are stored in a storage unit 1 that constitutes the information processing device A. The various types of information include, for example, document information (to be described later) and three-dimensional coordinate information (to be described later).

[0029] The document information storage unit 11 stores one or more document information. Document information is information on documents related to science, technology, etc. Document information is, for example, information on academic papers, information on scientific documents, and patent information. Patent information is, for example, published patent gazettes, registered patent gazettes, republished patent gazettes, registered utility model gazettes, etc., and the type does not matter. Document information is, for example, files, database records, etc., but the data structure does not matter. Document information is usually associated with a document identifier. Document identifiers are, for example, application numbers, patent numbers, patent application publication numbers, academic paper IDs, and scientific document IDs.

[0030] Furthermore, document information is associated with three or more classification codes and one or more attribute values. When document information is associated with three classification codes, the three classification codes are associated with the X-axis, Y-axis, and Z-axis, and are appropriately referred to as the first classification code, second classification code, and third classification code, respectively. Each of the three or more classification codes is an ordered code. An ordered code means, for example, that the code in the A section of the IPC patent classification code is followed by the code in the B section, ..., and then the code in the H section.

[0031] For example, order information is managed in storage unit 1 in association with two or more classification codes of the same type. Ordinarily, order information can be identified and managed by identifying the ascending or descending order of the symbols that make up the classification code. Alternatively, two or more classification codes of the same type are stored in storage unit 1 in order. In other words, information is stored so that the order of each classification code can be ascertained within a set of classification codes of the same type.

[0032] A document may be associated with two or more primary classification codes, two or more secondary classification codes, or two or more tertiary classification codes. Of the two or more classification codes associated with the document, one is called the primary classification code, and the others are called secondary classification codes. The primary classification code is the most important classification for classifying and displaying the document, such as the classification code that is located first among the two or more classification codes paired with the document information, or the classification code with the highest corresponding score.

[0033] Note that information A and information B correspond to each other as long as information B can be obtained from information A. Information A and information B correspond to each other when, for example, information A and information B are linked, information A and information B exist in the same file, information A and information B exist in the same record, information A includes information B, etc.

[0034] The classification code associated with the document information is information for classifying document information. Examples of classification codes include paper classification codes, patent classification codes, and Grant-in-Aid for Scientific Research classification codes. Paper classification codes are classification categories for papers established by the Japan Science and Technology Agency (JST), and are referred to as JST classification codes in this specification. For details about JST classification codes, see the explanation at the URL "http: / / t21help.nikkei.co.jp / reference / docs / JD3_sousa.pdf." Patent classification codes are, for example, IPC, F-term, FI, CPC, etc. Grant-in-Aid for Scientific Research classification codes are review categories for Grant-in-Aid for Scientific Research established by the Japan Society for the Promotion of Science, and are referred to as JSPS classification codes in this specification. For details about Grant-in-Aid for Scientific Research classification codes, see the explanation at the URL "https: / / www.jsps.go.jp / j-grantsinaid / 03_keikaku / data / h30 / h30_beppyo2-4.pdf." The three types of classification codes associated with the document information are not limited to the above types of classification codes, and four or more types of classification codes may be associated with the document information.

[0035] A score may be associated with one or more of the three classification codes associated with the document information. Generally, the larger the score, the more likely it is that the classification code associated with the score is associated with the document information.

[0036] Furthermore, the one or more attribute values ​​associated with the document information are attribute values ​​of the document. The attribute values ​​are, for example, quantitative parameters or qualitative parameters. Quantitative parameters are information that can be expressed numerically, such as the publication year of a paper, the number of authors, the impact factor, the number of citations, and the year of application. Qualitative parameters are information that cannot be expressed numerically, such as the name of the paper's author, the name of the affiliated institution, the academic journal in which the paper was published, the applicant, and the inventor.

[0037] The document information storage unit 11 may be present in an external device (not shown). In this case, the processing unit 3 accesses the external device and composes an image (described later).

[0038] The three-dimensional coordinate storage unit 12 stores one or more pieces of three-dimensional coordinate information (x, y, z) in association with each of one or more pieces of document information in the document information storage unit 11. The three-dimensional coordinate information is information (x, y, z) in which x is a value corresponding to a first classification code paired with the document information, y is a value corresponding to a second classification code paired with the document information, and z is a value corresponding to a third classification code paired with the document information. Note that, as will be described in detail later, it is practically convenient to consider (x, y, z) as points in a three-dimensional integer grid space (a lattice space of integer coordinates). Therefore, x, y, and z can be considered to be integers.

[0039] The one or more pieces of three-dimensional coordinate information in the three-dimensional coordinate storage unit 12 are usually information acquired by the coordinate information accumulation unit 32.

[0040] The receiving unit 2 receives various instructions and information, such as selection conditions, coordinate assignment instructions, stereoscopic image output instructions, and flat image output instructions.

[0041] The selection conditions are conditions for selecting two or more pieces of document information from the document information in the document information storage unit 11. The selection conditions are usually conditions having one or more attribute values ​​of the document information. The selection conditions may be conditions for acquiring all of the document information in the document information storage unit 11. The selection conditions are conditions for the image construction unit 34 to select document information to be processed to construct an image. The selection conditions may also be called search conditions.

[0042] The coordinate assignment instruction is an instruction to obtain one or more pieces of three-dimensional coordinate information for each piece of document information and to associate the one or more pieces of three-dimensional coordinate information with each piece of document information.

[0043] The stereoscopic image output instruction is an instruction to output a three-dimensional integer grid space image, which will be described later. The stereoscopic image output instruction has, for example, a selection condition.

[0044] The 2D image output instruction is an instruction to output a 2D image, which will be described later. The 2D image output instruction includes, for example, a selection condition.

[0045] The condition receiving unit 21 receives the selection conditions. The condition receiving unit 21 may receive a stereoscopic image output instruction including the selection conditions, or a flat image output instruction including the selection conditions.

[0046] Here, acceptance usually refers to reception by a communication means from a terminal device not shown, but it may also be a concept that includes acceptance of information input from an input device such as a keyboard, mouse, or touch panel, or acceptance of information read from a recording medium such as a disk, magnetic disk, or semiconductor memory.

[0047] The processing unit 3 performs various types of processing, such as processing performed by a classification code assigning unit 31, a coordinate information storage unit 32, a coordinate information acquisition unit 33, and an image construction unit .

[0048] The classification code assigning unit 31 uses a learning device to perform an assignment process on each piece of document information in the document information storage unit 11 using a machine learning prediction algorithm, assigning one or more classification codes of one type and associating them with each piece of document information. Note that the assignment process may also be called a prediction process. Assigning a classification code may also be considered as obtaining a classification code.

[0049] It is preferable that the classification code assigning unit 31 uses a learning device to perform an assignment process on each piece of scientific literature information in the literature information storage unit 11 using a machine learning algorithm, assign one or more IPCs to each piece of literature information, and store the one or more IPCs in association with each piece of scientific literature information. Note that the learning device is information created by learning, for example, training data, which is two or more pieces of literature information associated with one or more classification codes of one type, using a machine learning learning algorithm. The learning device is also stored in the storage unit 1.

[0050] The machine learning algorithm used in the assignment process does not matter, and examples of machine learning algorithms that can be used include deep learning, SVM, decision trees, and random forests.

[0051] Performing the assignment process using a machine learning algorithm involves, for example, providing each piece of literature information and a learning device to a machine learning function (e.g., a fastText, TinySVM, or tensorflow library) and running the function to obtain one or more classification codes of one type for each piece of literature information. Note that when the classification code assignment unit 31 obtains two or more classification codes and scores corresponding to each classification code using the machine learning function, it may select and acquire 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, or that the score is equal to or greater than a threshold, or the like.

[0052] It should be noted that the information processing device A does not need to include the classification code assigning unit 31. In other words, one or more types of classification codes may be assigned to the document information by an external device (not shown). It should be noted that the classification code may be assigned to the document information as long as the classification code corresponds to the document information.

[0053] The coordinate information accumulation unit 32 accesses the document information storage unit 11 and acquires one or more pieces of three-dimensional coordinate information (x, y, z), where x is a value corresponding to the first classification code paired with each of the two or more pieces of document information, y is a value corresponding to the second classification code paired with each of the two or more pieces of document information, and z is a value corresponding to the third classification code paired with each of the two or more pieces of document information. Next, the coordinate information acquisition unit 33 accumulates the acquired one or more pieces of three-dimensional coordinate information in the three-dimensional coordinate storage unit 12, for each of the two or more pieces of document information, in association with the document information.

[0054] For example, for each of one or more pieces of document information among two or more pieces of document information, the coordinate information accumulation unit 32 acquires primary coordinate information, which is three-dimensional coordinate information using only the primary classification code of three types of classification, and one or more pieces of secondary coordinate information, which is three-dimensional coordinate information using at least one secondary classification code. Then, the coordinate information accumulation unit 32 associates the primary coordinate information and the one or more pieces of secondary coordinate information with each of the one or more pieces of document information and accumulates them in the three-dimensional coordinate storage unit 12.

[0055] The sub-coordinate information may be three-dimensional coordinate information consisting only of values ​​corresponding to the sub-classification codes in each of the three types of classifications, or may be three-dimensional coordinate information including a value corresponding to at least one sub-classification code. That is, the sub-coordinate information may be three-dimensional coordinate information including a value corresponding to the head classification code in one of the three types of classification codes and a value corresponding to the sub-classification codes in each of the other two types of classification codes. The sub-coordinate information may also be three-dimensional coordinate information including a value corresponding to the head classification code in two of the three types of classification codes and a value corresponding to the sub-classification code in the other one type of classification code.

[0056] There is no restriction on the timing or trigger of the operation of the coordinate information accumulation unit 32. In other words, the coordinate information accumulation unit 32 may acquire one or more pieces of three-dimensional coordinate information for each piece of document information before the reception unit 2 receives a stereoscopic image output instruction, a two-dimensional image output instruction, or a selection condition, and may store the one or more pieces of three-dimensional coordinate information in the three-dimensional coordinate storage unit 12 in association with each piece of document information.

[0057] Furthermore, when the receiving unit 2 receives a stereoscopic image output instruction or a flat image output instruction or selection conditions, the coordinate information accumulation unit 32 may acquire one or more pieces of three-dimensional coordinate information for each piece of document information selected as a target for constructing a stereoscopic image or a flat image. In such a case, it is preferable that the three-dimensional coordinate storage unit 12 is a volatile recording medium such as a memory.

[0058] The coordinate information acquisition unit 33 acquires three-dimensional coordinate information corresponding to one or more pieces of document information paired with one or more attribute values ​​that match the selection conditions from the three-dimensional coordinate storage unit 12. The coordinate information acquisition unit 33 acquires one or more pieces of three-dimensional coordinate information from the three-dimensional coordinate storage unit 12 for each piece of document information.

[0059] The three-dimensional coordinate information acquired by the coordinate information acquisition unit 33 and corresponding to one piece of document information always includes primary coordinate information. Furthermore, the three-dimensional coordinate information acquired by the coordinate information acquisition unit 33 and corresponding to one piece of document information may include one or more secondary coordinate information. In other words, the coordinate information acquisition unit 33 may acquire two or more pieces of three-dimensional coordinate information, which are primary coordinate information and one or more secondary coordinate information corresponding to one or more pieces of document information paired with one or more attribute values ​​that match the selection conditions.

[0060] It is preferable that the coordinate information acquisition unit 33 acquires three-dimensional coordinate information using only classification codes corresponding to scores that satisfy the extraction conditions, among classification codes of types associated with scores.

[0061] The extraction conditions are conditions for determining the score threshold to be adopted. Preferably, the extraction conditions are conditions for achieving a balance in the number of three-dimensional coordinate information pieces on each of the YZ plane, XZ plane, and XY plane. Note that the balance here refers to a balance in terms of the likelihood, probability, likelihood, etc., that classification codes correspond to document information. The YZ plane, XZ plane, and XY plane are each of the three orthogonal planes that make up the three-dimensional integer grid space image.

[0062] For example, for classification codes of a type not associated with a score, the coordinate information acquisition unit 33 acquires the total number of primary classification codes and the total number of secondary classification codes corresponding to one or more pieces of document information in the document information set. Next, the coordinate information acquisition unit 33, for example, uses the total number to acquire a first parameter (e.g., the ratio of the total number of primary classification codes to the total number of secondary classification codes). Next, the coordinate information acquisition unit 33 acquires, for example, a score threshold, which is a parameter for classification codes associated with a score, such that the difference between the first parameter and a second parameter acquired in the same manner as the first parameter is small enough to satisfy a condition. Next, the coordinate information acquisition unit 33 acquires three-dimensional coordinate information using, for example, primary classification codes and secondary classification codes corresponding to scores equal to or greater than the acquired threshold. Note that the above condition is, for example, that the difference between the first parameter and the second parameter is within a threshold, is smaller than the threshold, or is minimal.

[0063] The first parameter is, for example, the ratio between the total number of minor classification codes corresponding to one or more pieces of document information in the document information set and the number of one or more pieces of document information in the document information set (the total number of primary classification codes corresponding to one or more pieces of document information). The first parameter is, for example, the total number of minor classification codes corresponding to one or more pieces of document information in the document information set.

[0064] When a score is associated with all three types of classification codes, the coordinate information acquisition unit 33 determines a score threshold for each of the three types of classification codes, for example, so that the difference between the total number of each type of sub-classification code associated with each document information in the document information set and the threshold is equal to or less than a certain number.

[0065] The document information set here is document information paired with one or more attribute values ​​that match the selection conditions, or two or more document information stored in the document information storage unit 11.

[0066] The stereoscopic image construction means 341 constituting the image construction unit 34 acquires a first document number, which is the number of three-dimensional coordinate information for one or more (y, z), using three-dimensional coordinate information corresponding to each of the one or more pieces of document information acquired by the coordinate information acquisition unit 33, and constructs a YZ plane in which the first document number is visually represented corresponding to one or more (y, z) points on the YZ plane. Also, the stereoscopic image construction means 341 acquires a second document number, which is the number of three-dimensional coordinate information for one or more (x, z), using three-dimensional coordinate information corresponding to each of the one or more pieces of document information acquired by the coordinate information acquisition unit 33, and constructs an XZ plane in which the second document number is visually represented corresponding to one or more (x, z) points on the XZ plane. Furthermore, the three-dimensional image construction means 341 calculates a third number of documents, which is the number of three-dimensional coordinate information for each of one or more (x, y), using three-dimensional coordinate information corresponding to each of one or more pieces of document information acquired by the coordinate information acquisition unit 33, and constructs an XY plane in which the third number of documents is visually displayed corresponding to each of one or more (x, y) points on the XY plane. Next, the three-dimensional image construction means 341 constructs a three-dimensional integer grid space image by adjoining the constructed YZ plane, XZ plane, and XY plane, respectively.

[0067] A three-dimensional integer grid space image is a three-dimensional image in a three-dimensional integer grid space, and is usually shaped like a rectangular parallelepiped or cube. The three-dimensional integer grid space is a three-dimensional virtual space in which, for an ordered class X, first classification codes are in ascending or descending order and correspond to integer positions on the X-axis coordinate in the three-dimensional space in ascending or descending order, for an ordered class Y, second classification codes are in ascending or descending order and correspond to integer positions on the Y-axis coordinate in the three-dimensional space in ascending or descending order, and for an ordered class Z, third classification codes are in ascending or descending order and correspond to integer positions on the Z-axis coordinate in the three-dimensional space in ascending or descending order.

[0068] The planar image construction means 342 acquires one or more directed line segments, starting from the point indicated by the primary coordinate information and ending at the point indicated by the secondary coordinate information, on a cut plane of the three-dimensional integer grid space, using the primary coordinate information and secondary coordinate information corresponding to one or more pieces of document information. Next, the planar image construction means 342 constructs a planar image in which the acquired one or more directed line segments are expressed. The cut plane is not limited. The cut plane is, for example, a plane determined by the x-coordinate value, y-coordinate value, or z-coordinate value included in the planar image output instruction received by the receiving unit 2. In other words, if the planar image output instruction includes an x-coordinate value, the cut plane is a YZ plane cut by the x-coordinate value. If the planar image output instruction includes a y-coordinate value, the cut plane is an XZ plane cut by the y-coordinate value. If the planar image output instruction includes a z-coordinate value, the cut plane is an XY plane cut by the z-coordinate value.

[0069] The planar image constructing means 342 obtains the weight of each of one or more directed line segments using the score corresponding to the classification code, and constructs a planar image having directed line segments whose weights are visually recognizable.

[0070] Note that the weight generally increases as the score paired with the classification code increases. The weight is, for example, a value calculated using an increasing function (e.g., product, sum, average, etc.) with parameters being the scores paired with the classification code corresponding to each value of the sub-coordinate information (x, y, z) corresponding to the directed line segment. The weight is, for example, a value calculated using an increasing function with parameters being the scores paired with the classification code corresponding to each value of the head coordinate information (x, y, z) corresponding to the directed line segment. The weight is, for example, a value calculated using an increasing function with parameters being the scores paired with the classification code corresponding to each value of the sub-coordinate information (x, y, z) corresponding to the directed line segment and the scores paired with the classification code corresponding to each value of the head coordinate information (x, y, z) corresponding to the directed line segment.

[0071] Note that literature information can also be viewed as a network of connections between the primary classification code and secondary classification codes of each classification in a three-dimensional grid space. In other words, the network formed by literature information is a radial network with the primary classification code as the starting point and the secondary classification code as the end point, consisting of directed lines connecting the two, each of which points toward the secondary classification code, and whose end point is the secondary classification code. This network displays the co-occurrence relationships of classification codes in literature information, as well as the superior-subordinate relationships of the classification codes.

[0072] The output unit 4 outputs various types of information, such as a three-dimensional image or a two-dimensional image.

[0073] The image output unit 41 outputs the three-dimensional image constructed by the three-dimensional image construction means 341. The image output unit 41 outputs the two-dimensional image constructed by the two-dimensional image construction means 342. The image output unit 41 may output both the three-dimensional image and the two-dimensional image, or may output either the three-dimensional image or the two-dimensional image.

[0074] Furthermore, here, output usually means display on a display, but it may also be a concept that includes projection using a projector, printing on a printer, transmission to an external device (for example, a user terminal not shown), storage on a recording medium, or handing over the processing results to another processing device or another program, etc.

[0075] The storage unit 1, document information storage unit 11, and three-dimensional coordinate storage unit 12 are preferably non-volatile recording media, but can also be realized as volatile recording media.

[0076] There is no restriction on the process by which information is stored in the storage unit 1 etc. For example, information may be stored in the storage unit 1 etc. via a recording medium, information transmitted via a communication line etc. may be stored in the storage unit 1 etc., or information input via an input device may be stored in the storage unit 1 etc.

[0077] The receiving unit 2 and the condition receiving unit 21 can be realized by a device driver for an input means such as a touch panel or a keyboard, control software for a menu screen, etc. The receiving unit 2 etc. may also be realized by a communication means.

[0078] The processing unit 3, classification code assignment unit 31, coordinate information storage unit 32, coordinate information acquisition unit 33, image construction unit 34, 3D image construction means 341, and 2D image construction means 342 can usually be realized by a processor, memory, etc. The processing procedures of the processing unit 3, etc. are usually realized by software, and the software is recorded on 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.

[0079] The output unit 4 and the image output unit 41 may or may not include output devices such as a display or a speaker. The output unit 4, etc. may be realized by driver software for an output device, or by a combination of driver software for an output device and the output device, etc. The output unit 4, etc. may also be realized by a communication means.

[0080] Next, an example of the operation of the information processing device A will be described with reference to the flowchart of FIG.

[0081] (Step S201) The reception unit 2 determines whether or not a coordinate assignment instruction has been received. If a coordinate assignment instruction has been received, the process proceeds to step S202, and if a coordinate assignment instruction has not been received, the process proceeds to step S207.

[0082] (Step S202) The coordinate information accumulation unit 32 assigns 1 to a counter i.

[0083] (Step S203) The coordinate information accumulation unit 32 determines whether or not the i-th document information to which coordinate information is to be assigned exists in the document information storage unit 11. If the i-th document information exists, the process proceeds to step S204; if not, the process returns to step S201.

[0084] (Step S204) The coordinate information accumulation unit 32 acquires all three types of classification codes paired with the i-th document information.

[0085] (Step S205) The coordinate information accumulation unit 32 performs a process of acquiring coordinate information using the three types of classification codes acquired in step S204. An example of such coordinate information acquisition process will be described with reference to the flowchart of FIG.

[0086] (Step S206) The coordinate information accumulation unit 32 increments the counter i by 1. The process returns to step S203.

[0087] (Step S207) The reception unit 2 determines whether or not a stereoscopic image output instruction has been received. If a stereoscopic image output instruction has been received, the process proceeds to step S208, and if a stereoscopic image output instruction has not been received, the process proceeds to step S220.

[0088] (Step S208) The coordinate information acquisition unit 33 acquires the selection conditions included in the stereoscopic image output instruction.

[0089] (Step S209) The coordinate information acquisition unit 33 accesses the document information storage unit 11 and determines one or more pieces of document information that satisfy the selection conditions acquired in step S208. Here, the three-dimensional image construction means 341 acquires, for example, a document identifier for each of the one or more pieces of document information.

[0090] (Step S210) The coordinate information acquisition unit 33 acquires extraction conditions. An example of such extraction condition acquisition processing will be described with reference to the flowchart of FIG.

[0091] (Step S211) The coordinate information acquisition unit 33 assigns 1 to a counter i.

[0092] (Step S212) The coordinate information acquisition unit 33 determines whether the i-th document information exists among the one or more documents determined in step S209. If the i-th document information exists, the process proceeds to step S213, and if the i-th document information does not exist, the process proceeds to step S218.

[0093] (Step S213) The coordinate information acquisition unit 33 assigns 1 to a counter j.

[0094] (Step S214) The coordinate information acquisition unit 33 determines whether the j-th coordinate information paired with the i-th document information matches the extraction conditions. If the j-th coordinate information matches the extraction conditions, the process proceeds to step S215; if the j-th coordinate information does not match the extraction conditions, the process proceeds to step S217. Note that the coordinate information here is three-dimensional coordinate information.

[0095] (Step S215) The coordinate information acquiring unit 33 stores the j-th coordinate information in association with the i-th document information. Preferably, the coordinate information acquiring unit 33 stores the value constituting the j-th coordinate information in association with a score paired with the classification code that is the basis of the value.

[0096] (Step S216) The coordinate information acquisition unit 33 increments the counter j by 1. The process returns to step S214.

[0097] (Step S217) The coordinate information acquisition unit 33 increments the counter i by 1. The process returns to step S212.

[0098] (Step S218) The 3D image construction means 341 constructs a 3D image using the coordinate information accumulated in step S215. An example of such 3D image construction processing will be described with reference to the flowchart of FIG.

[0099] (Step S219) The image output unit 41 outputs the stereoscopic image constructed in step S218. The process returns to step S201.

[0100] (Step S220) The reception unit 2 determines whether or not a 2D image output instruction has been received. If a 2D image output instruction has been received, the process proceeds to step S221, and if a 2D image output instruction has not been received, the process returns to step S201.

[0101] (Step S221) The flat image construction means 342 determines a cutting plane corresponding to the flat image output instruction received in step S220. Here, the flat image construction means 342 typically acquires information specifying the cutting plane (for example, an x ​​value, a y value, or a z value).

[0102] (Step S222) The two-dimensional image construction means 342 performs processing to construct a two-dimensional image. An example of such two-dimensional image construction processing will be described with reference to the flowchart of FIG.

[0103] (Step S223) The image output unit 41 outputs the two-dimensional image constructed in step S222. The process returns to step S201.

[0104] In the flowchart of FIG. 2, it is preferable that the classification code assigning unit 31 assigns a classification code to each piece of document information in advance by the above-mentioned process.

[0105] In the flowchart of FIG. 2, the process ends when the power is turned off or an interrupt occurs to end the process.

[0106] Next, an example of the coordinate information acquisition process in step S205 will be described with reference to the flowchart in FIG.

[0107] (Step S301) The coordinate information accumulation unit 32 assigns 1 to a counter i.

[0108] (Step S302) Coordinate information accumulation unit 32 determines whether or not the i-th first category code exists among the first category codes of category X acquired in step S204. If the i-th first category code exists, proceed to step S303; if the i-th first category code does not exist, return to the upper level processing.

[0109] (Step S303) The coordinate information accumulation unit 32 acquires from the storage unit 1 the value x corresponding to the i-th first classification code.

[0110] (Step S304) The coordinate information accumulation unit 32 acquires a score corresponding to the value x corresponding to the i-th first classification code, which is paired with the i-th first classification code paired with the document information of interest. Note that here, if there is no score paired with the i-th first classification code paired with the document information of interest, the coordinate information accumulation unit 32 does not acquire a score.

[0111] (Step S305) The coordinate information accumulation unit 32 assigns 1 to the counter j.

[0112] (Step S306) Coordinate information accumulation unit 32 determines whether or not the jth second classification code exists among the second classification codes of category Y acquired in step S204. If the jth second classification code exists, proceed to step S307; if the jth second classification code does not exist, proceed to step S316.

[0113] (Step S307) The coordinate information accumulation unit 32 acquires the value y corresponding to the j-th second classification code from the storage unit 1.

[0114] (Step S308) The coordinate information accumulation unit 32 acquires a score corresponding to the value y corresponding to the jth second classification code, which is paired with the jth second classification code paired with the document information of interest. Note that here, if there is no score paired with the jth second classification code paired with the document information of interest, the coordinate information accumulation unit 32 does not acquire a score.

[0115] (Step S309) The coordinate information accumulation unit 32 assigns 1 to a counter k.

[0116] (Step S310) Coordinate information accumulation unit 32 determines whether or not the kth third classification code exists among the third classification codes of category Z acquired in step S204. If the kth third classification code exists, proceed to step S311; if the kth third classification code does not exist, proceed to step S315.

[0117] (Step S311) The coordinate information accumulation unit 32 acquires from the storage unit 1 the value z corresponding to the k-th third classification code.

[0118] (Step S312) The coordinate information accumulation unit 32 acquires a score corresponding to the value z corresponding to the kth third classification code, which is paired with the kth third classification code paired with the document information of interest. Note that here, if there is no score paired with the kth third classification code paired with the document information of interest, the coordinate information accumulation unit 32 does not acquire a score.

[0119] (Step S313) The coordinate information accumulation unit 32 uses the information acquired in steps S303, S304, S307, S308, S311, and S312 to accumulate the coordinate information (x, y, z) and scores paired with each value of the coordinate information in association with the document information of interest. Note that if a score cannot be acquired, the score is not accumulated.

[0120] (Step S314) The coordinate information accumulation unit 32 increments the counter k by 1. The process returns to step S310.

[0121] (Step S315) The coordinate information accumulation unit 32 increments the counter j by 1. The process returns to step S306.

[0122] (Step S316) The coordinate information accumulation unit 32 increments the counter i by 1. The process returns to step S302.

[0123] In the flowchart of Figure 3, it is preferable to store the coordinate information obtained using the first, second, and third classification codes as the primary coordinate information, distinguishing it from the secondary coordinate information, which is the other coordinate information.

[0124] In the flowchart of FIG. 3, the coordinate information and the like are usually stored in the three-dimensional coordinate storage unit 12, but this is not restrictive.

[0125] Furthermore, in the flowchart of FIG. 3, the sub-coordinate information may be composed of only the sub-classification code from among the three types of classification codes.

[0126] Next, an example of the extraction condition acquisition process in step S210 will be described with reference to the flowchart in FIG.

[0127] (Step S401) The coordinate information acquisition unit 33 assigns 1 to a counter i.

[0128] (Step S402) The coordinate information acquisition unit 33 determines whether or not there is an i-th type classification code that is not associated with a score. If there is, the process proceeds to step S403, and if there is not, the process proceeds to step S406.

[0129] (Step S403) The coordinate information acquisition unit 33 acquires the number of all sub-classification codes that are the i-th type of classification code and that correspond to the target document information.

[0130] (Step S404) The coordinate information acquisition unit 33 acquires a first parameter (for example, the ratio between the number of documents and the number of all sub-classification codes) using the number of all sub-classification codes acquired in step S403.

[0131] (Step S405) The coordinate information acquisition unit 33 increments the counter i by 1. The process returns to step S402.

[0132] (Step S406) The coordinate information acquisition unit 33 assigns 1 to the counter j.

[0133] (Step S407) The coordinate information acquisition unit 33 determines whether or not there is a j-th type classification code associated with a score. If there is, the process proceeds to step S408, and if there is not, the process returns to the upper process.

[0134] (Step S408) The coordinate information acquisition unit 33 determines a score threshold for adopting the jth type of classification code using one or two first parameters acquired in step S404. Then, the coordinate information acquisition unit 33 stores the score threshold in a buffer (not shown) in association with information specifying the jth type of classification code.

[0135] (Step S409) The coordinate information acquisition unit 33 increments the counter j by 1. The process returns to step S407.

[0136] In the flowchart of Figure 4, when scores are associated with all types of classification codes, the coordinate information acquisition unit 33 determines the score threshold for adopting each type of classification code, for example, so that the total number of sub-classification codes in each type of classification code is within the threshold.

[0137] Next, an example of the stereoscopic image construction process in step S218 will be described with reference to the flowchart in FIG.

[0138] (Step S501) The stereoscopic image construction means 341 assigns 1 to a counter i for the number of two-dimensional images.

[0139] (Step S502) The stereoscopic image construction means 341 assigns 1 to a counter j for the number of two-dimensional coordinates in the flat image.

[0140] (Step S503) The stereoscopic image construction means 341 determines whether or not the jth coordinate information exists on the ith plane among the YZ plane, XZ plane, and XY plane among the coordinate information accumulated in step S515. If the jth coordinate information exists on the ith plane, the process proceeds to step S504; if not, the process proceeds to step S507. For example, the coordinate information on the YZ plane is (-,0,0)(-,0,1)(-,0,2)... (-, maximum y coordinate, maximum z coordinate). Here, "-" is the x coordinate value, and indicates that it is not an issue (is empty).

[0141] (Step S504) The three-dimensional image construction means 341 acquires the number of coordinate information pieces among the coordinate information accumulated in step S215, which is the number of coordinate information pieces that include two-dimensional coordinate information ((*, y, z) [* is an arbitrary value] in the case of the YZ plane) among the jth coordinate information pieces on the i-th plane.

[0142] (Step S505) The three-dimensional image construction means 341 acquires information corresponding to the number of pieces of coordinate information acquired in step S504 (for example, a circle of a size proportional to the size of the number of pieces of coordinate information, a numerical value indicating the number of pieces of coordinate information, a bar-shaped rectangle of a height corresponding to the number of pieces of coordinate information, etc.). Next, the three-dimensional image construction means 341 places the acquired information at a position in the i-th plane that is indicated by the j-th two-dimensional coordinate information on the i-th plane.

[0143] (Step S506) The stereoscopic image construction means 341 increments the counter j by 1. The process returns to step S503.

[0144] (Step S507) The stereoscopic image construction means 341 increments the counter i by 1. The process returns to step S502.

[0145] (Step S508) The 3D image construction means 341 judges whether or not the counter i is 4. If the counter i is 4, the process proceeds to step S509, and if it is 3 or less, the process returns to step S503.

[0146] (Step S509) The stereoscopic image construction means 341 constructs a stereoscopic image using the YZ plane, XZ plane, and XY plane constructed in step S505, and returns to the upper level processing.

[0147] Next, an example of the two-dimensional image construction process in step S222 will be described with reference to the flowchart in FIG.

[0148] (Step S601) The two-dimensional image construction means 342 assigns 1 to a counter i.

[0149] (Step S602) The two-dimensional image construction means 342 determines whether or not the i-th document information exists among the document information to be processed. If the i-th document information exists, the process proceeds to step S603, and if not, the process returns to the upper process. The document information to be processed is, for example, the document information determined in step S209, the document information that is the basis for constructing the three-dimensional image.

[0150] (Step S603) The planar image construction means 342 acquires one or more pieces of coordinate information that are coordinate information on the cutting plane determined in step S221 and that are paired with the i-th document information. For example, if the cutting plane is the YZ plane specified by (x=10), the planar image construction means 342 acquires three-dimensional image information that matches (10, y, z) [y and z are arbitrary] and that are coordinate information that are paired with the i-th document information.

[0151] (Step S604) If the two-dimensional image construction means 342 was able to acquire the coordinate information in step S603, the process proceeds to step S605, and if not, the process proceeds to step S613.

[0152] (Step S605) The two-dimensional image construction means 342 determines whether or not secondary coordinate information exists in the coordinate information acquired in step S603. If secondary coordinate information exists, the process proceeds to step S606, and if not, the process proceeds to step S613.

[0153] (Step S606) The two-dimensional image construction means 342 assigns 1 to the counter j.

[0154] (Step S607) The two-dimensional image construction means 342 determines whether the j-th sub-coordinate information exists among the coordinate information acquired in step S603. If the j-th sub-coordinate information exists, the process proceeds to step S608; if not, the process proceeds to step S613.

[0155] (Step S608) The planar image construction means 342 determines whether or not a score is associated with the j-th sub-coordinate information. If a score is associated, the process proceeds to step S609, and if a score is not associated, the process proceeds to step S610.

[0156] (Step S609) The planar image construction means 342 acquires a weight using a score of 1 or more corresponding to the j-th sub-coordinate information. If a weight is not acquired here, the weight is assumed to be a default weight (for example, "1").

[0157] (Step S610) The flat image construction means 342 constructs a directed line segment having a shape (e.g., thickness) corresponding to the weight, with the node corresponding to the primary coordinate information of the i-th document information as the start point and the node corresponding to the j-th secondary coordinate information as the end point. Note that if the weight cannot be acquired, the flat image construction means 342 constructs the directed line segment using a default value for the weight (e.g., "1").

[0158] (Step S611) The planar image construction means 342 places the directed line segment constructed in step S610 at a position on the cross section, with the starting point being the position of the primary coordinate information of the i-th document information and the ending point being the position of the j-th secondary coordinate information.

[0159] (Step S612) The two-dimensional image construction means 342 increments the counter i by 1. The process returns to step S607.

[0160] (Step S613) The two-dimensional image construction means 342 increments the counter i by 1. The process returns to step S602.

[0161] A specific example (experimental example) of the operation of the information processing device A according to this embodiment will be described below.

[0162] Assume now that a set of document information to be classified is stored in the document information storage unit 11 of the information processing device A or an external device (not shown). Assume that this set of document information is a group of 15,683 papers published by Osaka Institute of Technology from 2001 to 2020, and is included in the paper database "JDREAM-III" of the Japan Science and Technology Agency (JST).

[0163] It is also assumed that each piece of document information in the document information storage unit 11 is associated with a large number of attribute values ​​including the year of publication, the number of citations, the author's name, the academic journal in which the document was published, and so on.

[0164] Furthermore, each piece of literature information in the literature information storage unit 11 is assigned a paper classification (which may also be called a JST classification) by the JST. As mentioned above, paper classifications consist of a primary classification code that is always assigned, and secondary classification codes that are assigned as needed in addition to the primary classification code. The number of assigned former codes was 15,683, the same as the number of papers, and the number of assigned latter codes was 9,567, with the ratio of the number of assigned former codes to the number of assigned latter codes (the first parameter here) being "γx = 0.61." Note that the JST classification does not have a corresponding score. This is because it is a classification that has not undergone machine learning prediction processing such as CNN.

[0165] Assume that a user operating a terminal device (not shown) inputs a coordinate information acquisition command. The classification code assignment unit 31 of the information processing device A assigned a primary classification code and a secondary classification code of the International Patent Classification (IPC classification) to this group of papers using the machine learning prediction process described above (here, the CNN method). The former numbered 15,683, and the latter numbered 18,349, with the assigned number ratio (first parameter) being "γy = 1.17." Similarly, the classification code assignment unit 31 assigned a primary classification code and a secondary classification code of the Grant-in-Aid for Scientific Research (JSPS classification) using the CNN method. The former numbered 15,683, and the latter numbered 15,709, with the assigned number ratio being "γz = 1.00." The classification code assignment unit stored each of these classification codes in the document information storage unit 11. The IPC classification and the JSPS classification are associated with scores because the machine learning prediction process was performed using the CNN method.

[0166] Then, the coordinate information accumulation unit 32 of the information processing device A acquires the classification code of category X (JST classification), the classification code of category Y (IPC classification), and the classification code of category Z (JSPS classification) that correspond to each paper (for example, papers with "document identifier = P001", "document identifier = P002", etc.) from the document information storage unit 11. Furthermore, the coordinate information accumulation unit 32 acquires values ​​corresponding to each of the three types of classification codes for each paper from the storage unit 1. It is assumed that the storage unit 1 stores values ​​(values ​​that constitute the coordinate information) corresponding to each of the three types of classification codes, as described above.

[0167] The coordinate information accumulation unit 32 also acquires scores corresponding to the classification codes assigned to each paper for categories Y and Z. Note that no score is associated with category X (JST classification). The coordinate information accumulation unit 32 then uses the three classification codes assigned to each paper to acquire, for each paper, values ​​corresponding to a set of the primary classification codes of each of the three classification codes from the storage unit 1 through the process described above, and acquires primary coordinate information having the three values ​​as elements. The coordinate information accumulation unit 32 also uses the three classification codes assigned to each paper to acquire, for each paper, one or more secondary coordinate information corresponding to a set of classification codes including one or more secondary classification codes, for each paper through the process described above. The coordinate information accumulation unit 32 then accumulates the acquired primary coordinate information and secondary coordinate information for each paper in the three-dimensional coordinate storage unit 12. Note that for any type of classification code, secondary coordinate information is not associated with a paper that is not associated with a secondary classification code. The y and z of each coordinate information are associated with a score paired with the classification code corresponding to y and z. It is assumed that the coordinate information storage unit 32 stores the primary coordinate information and the secondary coordinate information in a distinguishable manner.

[0168] As a result of the above processing by the coordinate information accumulation unit 32, the coordinate information management table shown in Figure 7 is stored in the three-dimensional coordinate storage unit 12. The coordinate information management table stores two or more records each having an "ID," "document identifier," "classification X [JST classification]," "classification Y (score) [IPC classification]," "classification Z (score) [JSPS classification]," and "coordinate information." The "ID" is information that identifies the record. The "document identifier" is information that identifies the document. The "coordinate information" is three-dimensional coordinate information acquired by the coordinate information accumulation unit 32. In the "coordinate information," three-dimensional coordinate information corresponding to "*" is primary coordinate information, and three-dimensional coordinate information that does not correspond to "*" is secondary coordinate information.

[0169] In this situation, five specific examples will be described below. Specific Example 1 is an example of outputting a stereoscopic image without using scores. Specific Example 2 is an example of outputting a stereoscopic image using only the scores of one type of classification code. Specific Example 3 is an example of outputting a stereoscopic image using the scores of two types of classification codes. Specific Example 4 is an example of outputting a flat image without taking weights into consideration. Specific Example 5 is an example of outputting a flat image with weights taken into consideration.

[0170] (Example 1) It is assumed that a user operating a terminal device (not shown) inputs a stereoscopic image output instruction including a selection condition of "publication year (2001-2010 or 2011-2013 or 2014-2016 or 2017-2020)." The terminal device then accepts the stereoscopic image output instruction and transmits it to information processing device A. It is assumed that the stereoscopic image output instruction here is a stereoscopic image output instruction that targets papers that fall into each of the four categories of publication year, "2001-2010," "2011-2013," "2014-2016," and "2017-2020," and indicates that the classification of the papers should be represented in a stereoscopic image by dividing them into each of the four categories.

[0171] Next, the condition receiving unit 21 of the information processing device A receives the above-mentioned stereoscopic image output instruction. Then, the coordinate information acquiring unit 33 acquires coordinate information from the coordinate information management table of FIG. 7 for each of the four categories. Next, the stereoscopic image constructing means 341 uses the acquired coordinate information to construct a stereoscopic image in a three-dimensional integer grid space, in this case, in the form of a three-dimensional bubble chart, representing the number of documents at each coordinate point, through the above-mentioned process. An example of such output is the bird's-eye view of FIG. 8. Here, the three-dimensional integer grid space is a three-dimensional space in a three-dimensional left-handed coordinate system (X-axis: JST classification, Y-axis: IPC classification, Z-axis: JSPS classification).

[0172] Furthermore, the three-dimensional image construction means 341 uses four types of frequency distributions, with the publication years "2001 to 2010" in blue, "2011 to 2013" in green, "2014 to 2016" in yellow, and "2017 to 2020" in red, that is, blue B, green G, yellow Y, and red R in this order, and calculates the number of three-dimensional coordinate information items that match the selection conditions, fB(x, y, z), fG(x, y, z), and fY The three-dimensional image construction means 341 integrates (x, y, z) and fR(x, y, z) in the X-axis direction to obtain the first number of documents, ∫fB(x, y, z)dx=pB(y, z), ∫fG(x, y, z)dx=pG(y, z), ∫fY(x, y, z)dx=pY(y, z), and ∫fR(x, y, z)dx=pR(y, z), and constructs the YZ plane as a concentric heat map of red, yellow, green, and blue. Furthermore, the three-dimensional image construction means 341 integrates the four types of frequency distributions in the Y-axis direction using the coordinate information in the coordinate information management table of FIG. 7 to obtain the second number of documents, qB(x, z), qG(x, z), qY(x, z), and qR(x, z), and constructs the XZ plane as a concentric heat map of red, yellow, green, and blue. 7, the three-dimensional image construction means 341 integrates the four types of frequency distributions in the Z-axis direction to obtain the third document numbers, rB(x,y), rG(x,y), rY(x,y), and rR(x,y), and constructs the XY plane as a concentric heat map of red, yellow, green, and blue. Next, the three-dimensional image construction means 341 constructs the integral projection (three-dimensional image) of FIG. 9 by orthogonally intersecting the constructed YZ plane, XZ plane, and XY plane.

[0173] Next, the image output unit 41 outputs the constructed stereoscopic image. Such a stereoscopic image is shown in FIG. 8 and / or FIG. 9. That is, the bird's-eye view and the integral projection may be overlapped, displayed in parallel, or displayed separately. The image output unit 41 may output either the bird's-eye view or the integral projection. In FIG. 8 and FIG. 9, 801 and 901 are the X-axis (JST classification), 802 and 902 are the Y-axis (IPC classification), and 803 and 903 are the Z-axis (JSPS classification).

[0174] Furthermore, it was difficult to grasp which parts of the three-dimensional grid space new papers were being published in using conventional technology such as the bird's-eye view pie chart parameter display in Figure 8. However, in the three-dimensional image (integral projection) in Figure 9, the frequency distribution is integrated to form a heat map, making it possible to accurately and clearly grasp the parts of the three-dimensional space where new papers are being published.

[0175] (Example 2) In specific example 2, when the coordinate information acquisition unit 33 acquires coordinate information, it does not change the score threshold for the JSPS classification but changes the score threshold for the IPC classification, and so that the second parameter (the above-mentioned assignment ratio) corresponding to the IPC classification is the same as or has a difference within the threshold compared to the first parameter of the JST classification, "γx = 0.61," the coordinate information acquisition unit 33 raises the score threshold to be adopted (here, raises it to 0.07) and sets the second parameter (γy) corresponding to the IPC classification to "γy = 0.61." Then, the coordinate information acquisition unit 33 acquires coordinate information from the coordinate information management table of Fig. 7 , excluding coordinate information including a y value corresponding to a classification code paired with a score smaller than the raised score threshold.

[0176] Next, the stereoscopic image construction means 341 constructs a stereoscopic image using the acquired coordinate information through the above-mentioned processing. Examples of such output are shown in FIGS. 10 and 11. In FIGS. 10 and 11, when the threshold value for the IPC classification score is increased to 0.07, γy decreases to "γy = 0.61," and coordinates below the threshold value among those corresponding to the IPC classification subclassification codes disappear. As a result, γx of the JST classification and γy of the IPC classification become "γx = γy," and the stereoscopic image as a whole is more balanced than in Example 1.

[0177] (Example 3) In specific example 3, when the coordinate information acquisition unit 33 acquires coordinate information, it varies both the score threshold for the JSPS classification and the score threshold for the IPC classification, and raises the score threshold to be adopted (here, to 0.04) so ​​that the second parameters (the above-mentioned assignment ratios) corresponding to the JSPS classification and the IPC classification are the same as or less than the first parameter "γx = 0.61" of the JST classification. The coordinate information acquisition unit 33 then sets the second parameter (γz) corresponding to the JSPS classification to "γz = 0.61" and the second parameter (γy) corresponding to the IPC classification to "γy = 0.61." Then, the coordinate information acquisition unit 33 acquires coordinate information from the coordinate information management table of FIG. 7 , excluding coordinate information including any of the z and y values ​​corresponding to classification codes paired with scores smaller than the raised score threshold.

[0178] Next, the stereoscopic image construction means 341 constructs a stereoscopic image using the acquired coordinate information through the above-described process. Examples of such output are shown in FIGS. 12 and 13. In FIGS. 12 and 13, when the threshold score for the JSPS classification is increased to 0.04 and the threshold score for the IPC classification is increased to 0.07, γz and γy decrease to "γz = 0.61" and "γy = 0.61." The coordinates corresponding to the JSPS classification subclassification codes that are below the threshold disappear, and the coordinates corresponding to the IPC classification subclassification codes that are below the threshold disappear. As a result, the γx of the JST classification, the γy of the IPC classification, and the γz of the JSPS classification become "γx = γy = γz." This results in a very well-balanced stereoscopic image overall, even compared to Example 2.

[0179] (Example 4) In specific example 4, it is assumed that the condition receiving unit 21 of the information processing device A receives a stereoscopic image output instruction that includes the selection condition "a collection of 15,683 papers from Osaka Institute of Technology published in JDREAM-III from 2001 to 2020, and papers from a specific research group in the field of organic chemistry." It is also assumed here that the score threshold for the IPC classification is 0.07, and the score threshold for the JSPS classification is 0.04. It is assumed that information regarding these score thresholds is stored in the storage unit 1.

[0180] Then, the stereoscopic image construction means 341 constructs a bird's-eye view integral projection through the above-described processing. Next, the image output unit 41 outputs the constructed stereoscopic image. The constructed stereoscopic image is shown in Fig. 14 (bird's-eye view) and Fig. 15 (integral projection).

[0181] Next, it is assumed that the receiving unit 2 of the information processing device A receives a flat image output instruction. The flat image output instruction includes the above-mentioned selection condition "a collection of 15,683 papers from Osaka Institute of Technology published in JDREAM-III from 2001 to 2020, and papers from a specific research group in the field of organic chemistry," and also includes information specifying the cross section (IPC=B01J31 / 00).

[0182] Then, the planar image construction means 342 uses the document information that meets the selection conditions, i.e., by processing to obtain three-dimensional coordinate information corresponding to IPC=B01J31 / 00, constructs, on a cutting plane (IPC=B01J31 / 00) that is parallel to the XZ plane and includes an integer position of the Y-axis coordinate, primary coordinate information that is the starting point, secondary coordinate information that is the end point, and directed line segments extending from the primary coordinate to the secondary coordinate.The planar image construction means 342 then places the constructed directed line segments on a plane.Through the above processing, the planar image construction means 342 has constructed a planar image.Next, the image output unit 41 outputs the planar image.An example of such output is shown in Figure 16.

[0183] Figure 16 shows the network statistics calculated by setting the weights of the starting point, directed line segments, and end points to an equal value of 1, and making the starting point and end point of the directed line segments proportional to their degree (the number of directed line segments connected to the starting point or end point), with the width of the directed line segments being a constant width.

[0184] The coordinates with the highest degree are X=CB06122I[8] Precious Metal Catalysts, Z=27030 Catalysis and Resource Chemical Processes, and from this origin, three directed lines radiate to the subclassifications, and conversely, directed lines radiate from X=CB06122I[8], Z=33020 Synthetic Chemistry.

[0185] Thus, the planar image in Figure 11 shows the co-occurrence and superior-subordinate relationships of the classifications, and it can be seen that the most important coordinates in the IPC=B01J31 / 00 plane are X=CB06122I[8] precious metal catalysts, Z=27030 catalysts and resource chemical processes.

[0186] (Example 5) Assume that the reception unit 2 of information processing device A receives a 2D image output instruction. The 2D image output instruction includes the selection criteria "a collection of 15,683 papers from Osaka Institute of Technology published in JDREAM-III from 2001 to 2020, and papers from a specific research group in the field of organic chemistry." The 2D image output instruction also includes information specifying a cross section (IPC=B01J31 / 00). Also, assume that the IPC classification score threshold is 0.07 and the JSPS classification score threshold is 0.04.

[0187] In this specific example, for JST classifications that do not have scores, if there is one primary classification code and zero secondary classification codes, the alternative score corresponding to the primary classification code is set to 1, if there is one primary classification code and one secondary classification code, the alternative scores for each code are set to 1 / 2, 1 / 2, and if there is one primary classification code and two secondary classification codes, the alternative scores for each code are set to 1 / 3, 1 / 3, 1 / 3. The alternative scores are provisional scores automatically assigned by information processing device A.

[0188] In this specific example, the weights of the start point, directed line segment, and end point are not all set to the same value of 1. Instead, the weight of the start point is set to the product of the scores of each element of the primary coordinate information: "x-coordinate score (Sx1) × y-coordinate score (Sy1) × z-coordinate score (Sz1)." In this specific example, the weight of the end point is set to the product of the scores of each element of the secondary coordinate information: "x-coordinate score (Sxi) × y-coordinate score (Syj) × z-coordinate score (Szk)" (where i, j, and k are integers that are not simultaneously 1, and 1≦i≦p, 1≦j≦q, and 1≦k≦r). In this specific example, the weight of the directed line segment is set to the same as the weight of the end point. Note that the weight may be a value calculated using another calculation formula instead of the product of the scores. For example, it is preferable for the weight to be a value calculated using an increasing function using the scores.

[0189] Under these conditions, the planar image construction means 342 uses the document information that meets the selection conditions to construct, through the above-described process, the primary coordinate information as the starting point, the secondary coordinate information as the end point, and a directed line segment extending from the primary coordinate to the secondary coordinate on a cutting plane (IPC=B01J31 / 00) that is parallel to the YZ plane and includes an integer position on the X-axis coordinate. The planar image construction means 342 then places the constructed directed line segment on a plane. Through the above process, the planar image construction means 342 has constructed a planar image. Next, the image output unit 41 outputs the planar image. An example of such an output is shown in FIG. 17.

[0190] Figure 17 shows the network statistics calculated, with the start and end points of the directed segments proportional to betweenness centrality (a centrality index of coordinates using critical paths), and the width of the directed segments corresponding to the scores of the end points.

[0191] In Figure 17, the coordinate with the highest betweenness centrality is X=CB06122I[8] Precious metal catalyst, Z=27030 Catalytic and resource chemical processes. From this point, three directed lines radiate to the subclassifications. However, since the width corresponds to the width of the score, the main directed line is directed toward X=CF02030M[8] Addition and elimination reactions, Z=27030 Catalytic and resource chemical processes, and it can be seen that the other two directed lines are less important.

[0192] In addition, X=CG03030W[8] Polymer Support · Catalytic Reaction, Z=47010 Medicinal Chemistry is the third most important coordinate in terms of betweenness centrality calculated using the score value, and it has a strong influence on the two coordinates in Z=47010 Medicinal Chemistry, and it can be seen that it influences and is influenced by X=CG03030W[8] Polymer Support · Catalytic Reaction, Z=2703027030 Catalyst · Resource Chemical Processes. However, it can also be seen that it has almost no influence on the three coordinates in Z=33020 Synthetic Chemistry, as the width of the directed line is almost zero.

[0193] In this way, the planar image allows the co-occurrence and superior-subordinate relationships of classification codes to be understood quantitatively, providing a deeper understanding that the most important coordinates on the IPC=B01J31 / 00 plane are X=CB06122I precious metal catalysts and Z=27030 catalysts and resource chemical processes.

[0194] As described above, according to this embodiment, it is possible to perform three-dimensional classification analysis of a collection of technical documents.

[0195] Furthermore, according to this embodiment, both the primary classification code and the secondary classification code can be used to perform a more accurate analysis of the three-dimensional classification of a collection of technical documents.

[0196] Furthermore, according to this embodiment, by using only appropriate classification codes with scores, it is possible to perform an appropriate analysis of classification for a collection of technical documents.

[0197] Furthermore, according to this embodiment, it is possible to perform analysis relating to classification of a set of technical documents using the primary coordinate information and secondary coordinate information.

[0198] Furthermore, according to this embodiment, the primary coordinate information, secondary coordinate information, and score can be used to perform a more precise analysis of classification of a collection of technical documents.

[0199] 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 A in this embodiment is the following program. That is, this program causes a computer to access a document information storage unit in which a first classification code, a second classification code, a third classification code, and two or more pieces of document information paired with one or more attribute values ​​related to documents are stored, and acquires three-dimensional coordinate information (x, y, z) in which a value corresponding to the first classification code paired with each of the two or more pieces of document information is x, a value corresponding to the second classification code paired with each of the two or more pieces of document information is y, and a value corresponding to the third classification code paired with each of the two or more pieces of document information is z, and stores the three-dimensional coordinate information (x, y, z) in the three-dimensional coordinate storage unit in association with the document information for each of the two or more pieces of document information; a condition receiving unit receiving selection conditions related to one or more attribute values ​​related to documents; a coordinate information acquiring unit acquiring, from the three-dimensional coordinate storage unit, three-dimensional coordinate information corresponding to each of the one or more pieces of document information paired with one or more attribute values ​​that match the selection conditions; and a program to cause the computer to function as an image construction unit that constructs a three-dimensional integer grid space image in which the YZ plane, the XZ plane, and the XY plane are adjacent to each other, and an image output unit that outputs the three-dimensional integer grid space image.

[0200] 18 shows the appearance of a computer that executes the programs described in this specification to realize the information processing devices and the like of the various embodiments described above. The above-described embodiments can be realized by computer hardware and computer programs executed thereon. FIG. 18 is an overview diagram of this computer system 300, and FIG. 19 is a block diagram of system 300.

[0201] In FIG. 18, a computer system 300 includes a computer 301 including a CD-ROM drive, a keyboard 302, a mouse 303, and a monitor 304.

[0202] 19, 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.

[0203] A program that causes computer system 300 to execute the functions of information processing device A and the like of the above-described embodiment may be stored on CD-ROM 3101, inserted into CD-ROM drive 3012, and then 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 when executed. The program may also be loaded directly from CD-ROM 3101 or the network.

[0204] 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 A of the above-described embodiment. The program only needs to include instructions that call appropriate functions (modules) in a controlled manner and achieve the desired results. How the computer system 300 operates is well known, and a detailed description thereof will be omitted.

[0205] 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 transmission step (processing that can only be performed by hardware).

[0206] The computer that executes the program may be a single computer or a plurality of computers, that is, it may perform centralized processing or distributed processing.

[0207] 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.

[0208] Furthermore, 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.

[0209] 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]

[0210] As described above, the information processing device A according to the present invention has the effect of being able to perform three-dimensional classification analysis of a collection of technical documents, and is useful as a server or the like that creates a technology classification map. [Explanation of symbols]

[0211] 1 Storage area 2. Reception 3 Processing section 4 Output section 11 Literature Storage Unit 12 Three-dimensional coordinate storage unit 21 Conditions Reception Department 31 Classification Code Assignment Section 32 Coordinate information storage unit 33 Coordinate information acquisition unit 34 Image Composition Section 41 Image output unit 341 Stereoscopic Image Composition Means 342 Planar image composition means

Claims

1. a coordinate information storage unit that accesses a document information storage unit in which two or more document information items paired with a first classification code, a second classification code, a third classification code, and one or more attribute values ​​related to documents are stored, acquires three-dimensional coordinate information (x, y, z) in which a value corresponding to the first classification code paired with each of the two or more document information items is x, a value corresponding to the second classification code paired with each of the document information items is y, and a value corresponding to the third classification code paired with each of the document information items is z, and stores the three-dimensional coordinate information (x, y, z) in the three-dimensional coordinate storage unit for each of the two or more document information items in association with the document information; a condition receiving unit that receives selection conditions regarding one or more attribute values ​​related to documents; a coordinate information acquisition unit that acquires, from the three-dimensional coordinate storage unit, three-dimensional coordinate information corresponding to one or more pieces of document information that are paired with one or more attribute values ​​that match the selection conditions; an image constructing unit that uses the three-dimensional coordinate information corresponding to each of the one or more document information that meets the selection condition to obtain a first document number, which is the number of three-dimensional coordinate information for one or more (y, z), and configures a YZ plane visually represented by the first document number corresponding to each of the one or more (y, z) points on the YZ plane; obtains a second document number, which is the number of three-dimensional coordinate information for one or more (x, z), and configures an XZ plane visually represented by the second document number corresponding to each of the one or more (x, z) points on the XZ plane; obtains a third document number, which is the number of three-dimensional coordinate information for one or more (x, y), and configures an XY plane visually represented by the third document number corresponding to each of the one or more (x, y) points on the XY plane; and configures a three-dimensional integer grid space image in which the YZ plane, the XZ plane, and the XY plane are orthogonal to each other; an image output unit that outputs the three-dimensional integer grid space image, The information processing device, wherein the first classification code, the second classification code, and the third classification code are each a different type of classification code.

2. a coordinate information storage unit that accesses a document information storage unit in which two or more document information items paired with a first classification code, a second classification code, a third classification code, and one or more attribute values ​​related to documents are stored, acquires three-dimensional coordinate information (x, y, z) in which a value corresponding to the first classification code paired with each of the two or more document information items is x, a value corresponding to the second classification code paired with each of the document information items is y, and a value corresponding to the third classification code paired with each of the document information items is z, and stores the three-dimensional coordinate information (x, y, z) in the three-dimensional coordinate storage unit for each of the two or more document information items in association with the document information; a condition receiving unit that receives selection conditions regarding one or more attribute values ​​related to documents; a coordinate information acquisition unit that acquires, from the three-dimensional coordinate storage unit, three-dimensional coordinate information corresponding to one or more pieces of document information that are paired with one or more attribute values ​​that match the selection conditions; using the three-dimensional coordinate information corresponding to each of the one or more pieces of document information that meet the selection conditions, an image constructing unit that integrates the number of three-dimensional coordinate information f(x, y, z) along the X axis to obtain a first number of documents ∫f(x, y, z)dx=p(y, z) to construct a YZ plane, that integrates the number of three-dimensional coordinate information f(x, y, z) along the Y axis to obtain a second number of documents ∫f(x, y, z)dy=q(x, z) to construct an XZ plane, that integrates the number of three-dimensional coordinate information f(x, y, z) along the Z axis to obtain a third number of documents ∫f(x, y, z)dz=r(x, y) to construct an XY plane, and that constructs a three-dimensional integer grid space image in which the YZ plane, the XZ plane, and the XY plane are orthogonal to each other; an image output unit that outputs the three-dimensional integer grid space image, The information processing device, wherein the first classification code, the second classification code, and the third classification code are each a different type of classification code.

3. In one or more of the three types of classifications, in addition to a primary classification code, one or more secondary classification codes are associated with at least one or more pieces of document information among the two or more pieces of document information, The coordinate information storage unit For each of one or more pieces of document information among the two or more pieces of document information, acquire primary coordinate information, which is three-dimensional coordinate information using only the primary classification code of the three types of classification, and one or more pieces of secondary coordinate information, which is three-dimensional coordinate information using at least one secondary classification code, and store the primary coordinate information and the one or more pieces of secondary coordinate information in the three-dimensional coordinate storage unit in association with each of the one or more pieces of document information; The coordinate information acquisition unit 3. An information processing device according to claim 1, further comprising: acquiring two or more pieces of three-dimensional coordinate information, which are primary coordinate information and one or more secondary coordinate information, corresponding to one or more pieces of document information that are paired with one or more attribute values ​​that match the selection conditions.

4. a score is associated with one or more classification codes among the first classification code, the second classification code, and the third classification code corresponding to each of the two or more pieces of document information; The coordinate information acquisition unit The information processing device according to claim 1 , wherein, for classification codes of a type associated with a score, three-dimensional coordinate information is acquired using only classification codes corresponding to scores that satisfy an extraction condition.

5. The coordinate information acquisition unit 5. The information processing device according to claim 4, wherein for classification codes of a type not associated with a score, a total number of sub-classification codes corresponding to each document information of a document information set that is two or more document information stored in the document information storage unit or document information paired with one or more attribute values ​​that match the selection condition is obtained, and a first parameter obtained using the total number is used to obtain a threshold value for the score, which is a parameter for classification codes of a type associated with a score and whose difference from a second parameter obtained in a similar manner to the first parameter is smaller than a predetermined condition, and three-dimensional coordinate information is obtained using only sub-classification codes corresponding to scores equal to or greater than the threshold value.

6. The image construction unit a planar image constructing means for constructing a planar image having one or more directed line segments, each having a starting point at a point indicated by the primary coordinate information and an ending point at a point indicated by the secondary coordinate information, on a section of the three-dimensional integer grid space image, using primary coordinate information and secondary coordinate information on the section, the primary coordinate information and secondary coordinate information corresponding to each of the one or more document information; The image output unit The information processing apparatus according to claim 3 , wherein the planar image is output.

7. At least one of the three classification codes paired with each of the two or more pieces of document information is associated with a score; The planar image constructing means The information processing apparatus according to claim 6 , further comprising: obtaining a weight for each of the one or more directed line segments using a score corresponding to the classification code; and constructing the planar image having directed line segments whose weights are visually recognizable.

8. An information processing method realized by a coordinate information storage unit, a condition receiving unit, a coordinate information acquisition unit, an image construction unit, and an image output unit, a coordinate information storage step in which the coordinate information storage unit accesses a document information storage unit in which two or more pieces of document information paired with a first classification code, a second classification code, a third classification code, and one or more attribute values ​​related to documents are stored, acquires three-dimensional coordinate information (x, y, z) in which a value corresponding to the first classification code paired with each of the two or more pieces of document information is x, a value corresponding to the second classification code paired with each of the two or more pieces of document information is y, and a value corresponding to the third classification code paired with each of the two or more pieces of document information is z, and stores the three-dimensional coordinate information (x, y, z) in the three-dimensional coordinate storage unit in association with each of the two or more pieces of document information; a condition receiving step in which the condition receiving unit receives selection conditions regarding one or more attribute values ​​related to documents; a coordinate information acquisition step in which the coordinate information acquisition unit acquires, from the three-dimensional coordinate storage unit, three-dimensional coordinate information corresponding to one or more pieces of document information that are paired with one or more attribute values ​​that match the selection conditions; an image construction step in which the image construction unit acquires a first document number, which is the number of three-dimensional coordinate information for one or more (y, z), using the three-dimensional coordinate information corresponding to each of the one or more document information that meets the selection condition, constructing a YZ plane visually represented by the first document number corresponding to each of the one or more (y, z) points on the YZ plane, acquiring a second document number, which is the number of three-dimensional coordinate information for one or more (x, z), constructing an XZ plane visually represented by the second document number corresponding to each of the one or more (x, z) points on the XZ plane, acquiring a third document number, which is the number of three-dimensional coordinate information for one or more (x, y), constructing an XY plane visually represented by the third document number corresponding to each of the one or more (x, y) points on the XY plane, and constructing a three-dimensional integer grid space image in which the YZ plane, the XZ plane, and the XY plane are orthogonal to each other; an image output step in which the image output unit outputs the three-dimensional integer grid space image; An information processing method, wherein the first classification code, the second classification code, and the third classification code are each a different type of classification code.

9. Computer, a coordinate information storage unit that accesses a document information storage unit in which two or more document information items paired with a first classification code, a second classification code, a third classification code, and one or more attribute values ​​related to documents are stored, acquires three-dimensional coordinate information (x, y, z) in which a value corresponding to the first classification code paired with each of the two or more document information items is x, a value corresponding to the second classification code paired with each of the document information items is y, and a value corresponding to the third classification code paired with each of the document information items is z, and stores the three-dimensional coordinate information (x, y, z) in the three-dimensional coordinate storage unit for each of the two or more document information items in association with the document information; a condition receiving unit that receives selection conditions regarding one or more attribute values ​​related to documents; a coordinate information acquisition unit that acquires, from the three-dimensional coordinate storage unit, three-dimensional coordinate information corresponding to one or more pieces of document information that are paired with one or more attribute values ​​that match the selection conditions; an image construction unit that uses the three-dimensional coordinate information corresponding to each of the one or more document information that meets the selection condition to obtain a first document number, which is the number of three-dimensional coordinate information for one or more (y, z), and constructs a YZ plane visually represented by the first document number corresponding to each of the one or more (y, z) points on the YZ plane; obtains a second document number, which is the number of three-dimensional coordinate information for one or more (x, z), and constructs an XZ plane visually represented by the second document number corresponding to each of the one or more (x, z) points on the XZ plane; obtains a third document number, which is the number of three-dimensional coordinate information for one or more (x, y), and constructs an XY plane visually represented by the third document number corresponding to each of the one or more (x, y) points on the XY plane; and constructs a three-dimensional integer grid space image in which the YZ plane, the XZ plane, and the XY plane are orthogonal to each other; a program for causing the image output unit to function as an image output unit that outputs the three-dimensional integer grid space image, The first classification code, the second classification code, and the third classification code are each a different type of classification code.

Images