JP2023077516A - Patent analysis method, patent analysis device, patent analysis program, information recording medium and patent analysis system - Google Patents

Abstract

To provide a patent analysis method, a patent analysis device, a patent analysis program, an information recording medium, and a patent analysis system that display a graph of patent analysis results in a bird's-eye view while allowing technical content to be easily understood.SOLUTION: A patent analysis device 1 includes: an assignment number data acquisition unit 11 that acquires a combination of a classification code of patent classification assigned to a set of patent applications and the number of assignments assigned to the classification code; a relative coordinate value calculation unit 12 that calculates relative coordinate values of the classification code of each layer based on the assignment number data; an absolute coordinate value calculation unit 13 that calculates, for each of the classification codes in the lowest layer, absolute coordinate values in plane coordinates by multiplying the relative coordinate values for all the upper layers to which the classification code belongs and the relative coordinate values for the lowest layer by a layer based scale factor that becomes larger for the upper layer and then adding them; and a graph display unit 16 that displays a graph in which appropriate symbols are drawn on absolute coordinates in the plane coordinates and labels corresponding to the symbols are superimposed and drawn.SELECTED DRAWING: Figure 1

Description

The present invention relates to a patent analysis method, a patent analysis device, a patent analysis program, an information recording medium, and a patent analysis system capable of presenting technical contents in an easy-to-understand and bird's-eye view.

Software that analyzes and graphs information related to patent classifications, which is one type of patent information, is used when you want to analyze what kind of technology a patent application for a specific company or technical field is about (for example, , see Patent Document 1). With such software, it is possible to tabulate the number of patent applications to which a classification symbol indicating an individual technology has been assigned in a set of patent applications to be analyzed. In addition, this software can display the aggregated results using a bar graph, a line graph, or a matrix chart in which bubbles of arbitrary size are arranged vertically and horizontally at regular intervals.

By appropriately using these graphs to display the aggregated results, the user can understand the increasing or decreasing tendency of the number of applications from the related technical fields and the number of applications for each filing year. It is also possible to grasp the technology possessed by a specific company and the trend of technology from the trends in related technology fields and the number of applications. In addition, it is possible to compare the technology owned by the company with that of other companies to confirm its strengths and weaknesses, and to search for promising markets from technical fields close to the technology fields owned by the company. Through these patent analyzes, patent information can be used to examine development policies and intellectual property strategies in companies.

JP 2006-209174 A

In recent years, the use of such patent analysis methods for the IP landscape as well as for the above-mentioned purposes has been considered. An IP landscape is an activity in which a company's intellectual property department proposes a management strategy or business strategy to the top management by conducting an analysis that incorporates intellectual property information as information related to management and business. As an activity of IP Landscape, such analysis is used to create materials that provide a bird's-eye view of the current business environment and predictions of future prospects, and to provide them to managers and business managers. However, since it is difficult for managers and business managers to read detailed materials related to intellectual property and make strategic decisions, the materials provided to management should be comprehensive, yet concise and technical. The material should be easy to understand.

In addition, due to the recent revision of the Corporate Governance Code, which is a guideline concerning a system for managing and supervising corporate management, companies are now required to disclose their intellectual property management policies and activity results to their stakeholders. be. It is preferable that the activity results disclosed as requested by these guidelines be quantitative materials that are easy to understand for various stakeholders such as shareholders, customers, and employees. Here, conventional graphs such as bar graphs and line graphs that can be displayed as graphs by the above-mentioned software can be used, but it is difficult to present information that is comprehensive and easy to understand the technical content.

A representative example of the means for solving the above problems includes a top layer that classifies technologies most broadly and a lower layer that subdivides the upper layer including the top layer. A patent analysis method for analyzing the number of classification symbols assigned to an arbitrary set of patent applications using a patent classification that has a hierarchical structure of multiple layers and classifies technologies with multiple classification symbols, a first acquisition step of acquiring assigned number data having a plurality of combinations of the classification symbol assigned to a set of patent applications and the assigned number assigned with the classification symbol; A relative coordinate value indicating the relative coordinates of the symbol, and a relative coordinate value indicating the relative coordinates of the classification symbol in the second and lower layers having the same classification symbol in the hierarchy one level higher. a relative coordinate value calculation step of calculating based on the assigned number data; and for each of the classification symbols of the lowest hierarchy, the relative coordinate values of all the upper hierarchies to which the classification symbol belongs and the relative coordinate values of the lowest hierarchy. an absolute coordinate value calculation step of calculating an absolute coordinate value in plane coordinates by multiplying a higher layer by a large scale factor for each layer and then adding the multiplication factor; and a graph display step of displaying the drawn graph.

According to the present invention, technical relevance can be represented as a distance on a plane coordinate, and a graph can be displayed in which symbols indicating classification symbols are drawn with technical relevance. As a result, the technical content of the analysis results can be presented in an easy-to-understand, bird's-eye view.

2 is a diagram showing an example of functional blocks of the patent analysis device of Embodiment 1. FIG. It is an explanatory view explaining a hierarchical structure of a patent classification. 4 is a schematic example of a given number data table in Embodiment 1. FIG. FIG. 10 is an explanatory diagram for explaining the given number data and the calculation process used for the calculation process of the relative coordinate value in the first embodiment; 4 is an explanatory diagram of allocation angles and coordinate angles on XY coordinates in Embodiment 1. FIG. 4 is a graph schematically showing a method of calculating relative coordinate values in Embodiment 1. FIG. 6 is an example of a data table relating to hierarchical scale factors in the first embodiment. 4 is a graph schematically showing a method of calculating absolute coordinate values in Embodiment 1. FIG. 4 is an example of a data table of symbol colors and transparency in the first embodiment. 1 is a schematic diagram showing an example of the configuration of a patent analysis device according to Embodiment 1. FIG. 4 is a flow chart showing the flow of processing of the patent analysis device in Embodiment 1. FIG. 4 is a schematic example of a data table used in calculation processing according to the first embodiment; 4 is a display screen of the patent analysis device according to Embodiment 1. FIG. 4 is an enlarged view of a part of the display screen of the patent analysis device in Embodiment 1. FIG. 4 is an example of a data table regarding correction values for each section in Embodiment 1. FIG. FIG. 10 is a display screen after correction of the patent analysis device in Embodiment 1. FIG. FIG. 11 is a graph schematically showing analysis results for two groups of patent applications in Embodiment 2. FIG. FIG. 11 is a schematic diagram showing an example of functional blocks of a patent analysis system connected to user terminals via a network according to Embodiment 3;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. The present invention should not be limited to these embodiments in any way, and can be implemented in various forms without departing from the spirit of the present invention.
<<Embodiment 1>>
<Embodiment 1: Overview>

The patent analysis method and patent analysis device according to the present embodiment calculate the relative coordinate values for each layer based on the number of patent classification symbols assigned to a set of arbitrary patent applications, and It is characterized in that it calculates the absolute coordinate value by multiplying and then adding a large scale factor for each hierarchy, and displays a graph in which the technical classification symbol is drawn at the absolute coordinate on the plane coordinate.
<Functional configuration>

FIG. 1 is a diagram showing an example of functional blocks of the patent analysis device of this embodiment. As shown in the figure, the patent analysis device 1 of this embodiment includes a given number data acquisition unit 11, a relative coordinate value calculation unit 12, an absolute coordinate value calculation unit 13, a symbol determination unit 14, a label determination unit 15, and a graph display unit. 16 and a correction value acquisition unit 17 . Note that the functional blocks of the patent analysis device 1 described below can be implemented as hardware, software, or both hardware and software, as will be described later. Specifically, if a computer that is an information processing device is used, the CPU, main memory, bus, or secondary recording device (hard disk, nonvolatile memory, storage media such as CDs and DVDs, and reading of these media) drive, etc.), input devices used for information input, printing equipment, display devices, hardware components such as other external peripherals, interfaces for external peripherals, communication interfaces, and controlling those hardware driver programs, other application programs, user interface applications, and the like.

Then, by the arithmetic processing of the CPU according to the program developed on the main memory, the data input from the input device and other interfaces and held on the memory and hard disk are processed and stored, and the above hardware and software An instruction is generated to control the Moreover, the present invention can be implemented not only as an apparatus, but also as a method. Also, part of such an invention can be configured as software. Further, the technical scope of the present invention also includes software program products used to cause a computer to execute such software. Also, computer-readable information recording media (compact discs, flexible discs, hard discs, magneto-optical discs, digital video discs, magnetic tapes, or semiconductor memories) on which such software program products are recorded are of course included. It is included in the technical scope of the present invention. The above is the same throughout the present specification.
<Description of each configuration of Embodiment 1>
<Explanation of number of grant data acquisition unit>

The granted number data acquisition unit 11 receives an external input of granted number data having a plurality of combinations of a patent classification code assigned to a set of arbitrary patent applications and the granted number to which the classification code is assigned. get. Further, the given number data acquisition unit 11 outputs the acquired given number data to the relative coordinate value calculation unit 12 . In many cases, the patent classification is divided into three layers or more, including the first layer that classifies the technology in the largest way, the second layer that subdivides the first layer, and the third layer that further subdivides the second layer. It has a hierarchical structure and classifies technologies with multiple classification symbols. Examples of patent classifications having a hierarchical structure of three or more levels include International Patent Classification (IPC), File Index (FI), Common Patent Classification (CPC), and European Patent Classification (ECLA). These patent classifications have a similar hierarchical structure of five or more levels, and any of them can be used to implement the present invention. In this embodiment, an analysis example using the international patent classification will be described. It should be noted that the present invention can be practiced even with a patent classification having a hierarchical structure less than five hierarchies. As such a patent classification, the F term used in Japan and the USPC used in the United States can also be used to implement the present invention.

FIG. 2 is an explanatory diagram for explaining the hierarchical structure of patent classifications used in international patent classifications and the like. As shown in the figure, the International Patent Classification, etc. consists of Sections, which are the first level, Classes, which are the second level, Subclasses, which are the third level, Main Groups, which are the fourth level, and the fifth level, which is the lowest level. It is composed of five layers of subgroups, which are layers, and has a lower layer that subdivides the upper layer to classify technologies hierarchically.

For example, as shown in the figure, the section of the classification symbol "A" representing "daily necessities" in the International Patent Classification is an example of "A01" representing "agriculture;forestry;livestock;hunting;capturing;fishing." subdivided into multiple classes containing classification symbols for Similarly, the "A01" class is subdivided into multiple subclasses, including the classification symbol "A01B", which stands for "soil work in agriculture or forestry". The "A01B" subclass is subdivided into a number of main groups including the classification symbol "A01B1/00" representing "hand tool". The main group of "A01B1/00" is subdivided into a plurality of "subgroups" including classification symbols of "A01B1/02" representing ".plough;excavator". Thus, in these patent classifications, each layer from section to subgroup functions as a classification symbol. In addition, a lower layer classification symbol is formed by adding a character string of alphanumeric characters or symbols to the end of the upper classification symbol or by rewriting the number "00" of the main group. In these patent classifications, having such a hierarchical structure makes it possible to express technical relationships with classification symbols. For example, a difference in classification symbols in a higher hierarchy means more technical separation than a difference in classification symbols in a lower hierarchy.
<Description of number of grants>

FIG. 3 is a schematic example of a given number data table showing the configuration of given number data. Grant number data exemplified in the present embodiment is obtained by summarizing the number of grants for a set of approximately 30,000 Japanese patent applications filed by a certain automobile manufacturer in the last approximately five years. This assigned number data has a plurality of combinations of patent classification symbols assigned to this set and assigned numbers assigned with the relevant classification symbols (in this example, the top 1000 assigned numbers). In the granted number data, as shown in the same figure, for example, the "granted number" that indicates how many patent applications have been assigned the classification symbol at the lowest level, such as a subgroup in the International Patent Classification, is the assigned number. can be set as Note that the number of granted patents can be generated by aggregating using dedicated patent analysis software or spreadsheet software.

As for the number of assignments, how many types of lower-level classification symbols subdividing the same upper-level classification symbol (in other words, how many types of lower-level classification symbols belonging to the same upper-level classification symbol) are assigned? or the number of times the classification symbol in the lower layer has been assigned (that is, the total number of assignments) may be used. As an example, in the example shown in FIG. 3, for the main group "A61B5/00", the classification symbols of the subgroups "A61B5/11", "A61B5/16" and "A61B5/18" are given. The number of granted species will be 3. In addition, the “number of times given” is 77 times as the sum of the number of times given for these three subgroups. In this way, for the hierarchy above the main group, by using the assigned number data relating to the subgroups, it is possible to easily perform the calculation process described later with the minimum amount of data. Of course, it is also possible to acquire the "number of assignments" for each hierarchy above the main group. Furthermore, as in the given number data table shown in the figure, the given number data acquisition unit 11 can acquire not only the classification symbol and the "number of given cases" but also the text data of the explanation for each classification symbol.
<Description of Relative Coordinate Value Calculation Unit>

The relative coordinate value calculator 12 calculates the relative coordinate values of the classification symbols of each layer based on the assigned number data, and outputs the calculated relative coordinate values to the absolute coordinate value calculator 13 . A relative coordinate value indicates a relative coordinate of a classification symbol in the first hierarchy, which is the highest level. The relative coordinate values in the second and lower layers are relative coordinate values indicating the relative coordinates of the classification symbols having the same classification symbol in the hierarchy one level higher. In this embodiment, the first layer relative coordinate value of the first layer, the second layer relative coordinate value of the second layer, the third layer relative coordinate value of the third layer, and the fourth layer relative coordinate value of the fourth layer A coordinate value and a fifth layer relative coordinate value of the fifth layer are calculated. Each relative coordinate value is a coordinate value indicating a position relative to the origin of relative coordinates in each layer, and is calculated for each layer for calculation of an absolute coordinate value to be described later.

FIG. 4 is an explanatory diagram for explaining the given number data used in the relative coordinate value calculation process and the calculation process. In the figure, relative coordinate value calculation processing is performed based on the "given species number" obtained by totaling the number of main class classification symbol types given to each section. Here, the relative coordinate value calculation unit 12 first sums up the “number of types given” to calculate the total number of types of main classes given in all sections. In the example of FIG. 4, "958" is calculated. Next, the assigned angle .theta.1 assigned to each section is calculated by dividing the "given genus" of each section by the calculated sum and adding "360". As shown in the figure, as an example, the assigned angle θ1a assigned to section “A” in the figure is 2.6°, and the assigned angle θ1b assigned to section “B” is 125.5°. In this way, a larger allocation angle θ1 is allocated to a section with a larger number of additions.

Next, the meaning of the angle calculated for each section and the calculation process for calculating the relative coordinate value will be described with reference to FIG. FIG. 5 is an explanatory diagram of allocation angles and coordinate angles on XY coordinates, which are plane coordinates. As shown in the figure, a larger assigned angle .theta.1 is assigned to a section with a greater number of "given species", thereby securing a wider area for drawing symbols for each section. Note that the angle in the figure is not the angle formed by the line segment inclined counterclockwise from the positive X-axis and the positive X-axis, but the line segment inclined clockwise from the positive Y-axis and the positive X-axis. will be described using the angle formed by the Y-axis of (the same applies to FIG. 6, which will be described later). The reason for taking such an angle is that the graph displayed by the patent analysis apparatus 1 does not show the result of mathematical calculation, but the symbol indicating the classification symbol is used as a material for examining the analysis result. This is because it is intended to be easily arranged. For example, arranging the symbols clockwise from the top of the screen or paper is easier to understand as a study material than arranging the symbols counterclockwise from the right.

Subsequently, the relative coordinate value calculator 12 calculates the coordinate angle θ2 using the assigned angle θ1. Specifically, when there is a previous section (classification symbol) in the order of arrangement, all the allocation angles θ1 of the previous section are added, and then the allocation angle θ1 of the own section (classification symbol) is halved and added. is calculated as the coordinate angle θ2. As an example, in section "A" in the figure, since there is no previous section in the order of arrangement, the coordinate angle θ2a is 1.3°, which is 1/2 of 2.6°. On the other hand, in section "B", since section "A", which is the previous section in the order of arrangement, exists, the allocation angle θ1a of section "A" is 2.6°, and the allocation angle θ1b of its own section is 2.6°. The coordinate angle θ2b of section “B” is 65.4° obtained by adding 62.8° obtained by halving 125.5°. For ease of understanding, only the coordinate angle θ2b is shown in FIG. reference). The coordinate angles θ2a to θ2h of each section calculated in this manner are set so that the symbols can be distributed on the coordinates surrounding the origin while allocating areas with large angles so as to widen the areas where the symbols are drawn. calculated to

Next, the relative coordinate value calculation unit 12 calculates the assignment rank R, which is the descending order of the “number of assigned types” of each section. As shown in FIG. 4, in the set according to the present embodiment, the section "B" having the largest number of 334 "given types" has the highest given rank Rb. Ranks Ra to Rh for each section are calculated by calculating second and lower ranks as shown in the figure. Subsequently, the relative coordinate value calculator 12 calculates the relative coordinate values using a trigonometric function based on the assigned order Ra to Rh of each section and the coordinate angles θ2a to θ2h. FIG. 6 is a graph schematically showing a method of calculating relative coordinate values. Specifically, the X component in the relative coordinate value of each section is calculated as the cosine value of the line segment of the length from the origin proportional to the assigned ranks Ra to Rh inclined at the coordinate angles θ2a to θ2h. be. Similarly, the Y component in the relative coordinate values of each section is calculated as the sine value of the line segment of length from the origin proportional to the assigned ranks Ra to Rh inclined at the coordinate angles θ2a to θ2h.

For example, for section "E" shown in FIG. 6, the relative coordinate values are calculated based on the length Le corresponding to the 7th place in the assigned order Re and the coordinate angle θ2e of 134.0°. The coordinate angle θ2e of 134.0° corresponds to 316° in normal XY coordinates for using trigonometric functions. Therefore, the X component of the relative coordinate value of section "E" is the product of 0.719339, which is the cosine value of 316°, and the length Le. Also, the Y component of the relative coordinate value of section "E" is a value obtained by multiplying -0.694658, which is the sine value of 316°, and the length Le. The length L here is a relative value between sections, and the length Le of the section "E" with the seventh ranking Re is the length Lb of the section "B" with the first ranking Rb, for example. 7 times longer than . As a result, relative coordinate values are set for section "E" at a position seven times farther from the origin than section "B". Similarly, the relative coordinate value calculator 12 calculates relative coordinate values for each section.

In this manner, the relative coordinate value calculation unit 12 calculates the relative coordinate values so that the classification symbols are arranged in the order of arrangement in the patent classification while enclosing the origin, as shown in FIGS. 5 and 6 . Further, the relative coordinate value calculation unit 12 calculates the relative coordinate value using a trigonometric function while increasing the angle as the number of assigned classification symbols increases. Then, the relative coordinate value calculation unit 12 calculates the relative coordinate values so that the classification symbols are brought closer to the origin in descending order of the assigned number. As a result, by calculating the relative coordinate values so that the classification symbols are arranged in the order of the patent classification while enclosing the origin, the coordinates for drawing the symbols indicating the classification symbols can be determined.

By calculating the relative coordinate values in this way, it is possible to calculate the relative coordinate values having technical relevance without special operations or inputting individual setting values. In addition, a large area can be set near the center near the origin for a section in which many symbols need to be drawn because many classification symbols are assigned to lower layers. As a result, it is possible to indicate that a classification symbol with a large number of assignments is relatively core technology. In addition, even if many symbols are displayed, the symbols can be drawn in an easy-to-see manner by suppressing overlapping and crowding of the symbols. On the other hand, for sections where there are not many classification symbols and there is no need to draw many symbols, by setting an area away from the origin, it is possible to identify peripheral technologies that are not core technologies or new technologies. Symbols can be drawn in such a way that they are easy to understand, and overlap with other sections or are less likely to be crowded.

Subsequently, the relative coordinate value calculator 12 also calculates relative coordinate values for the layers below the class. In these hierarchies, the relative coordinate values are basically calculated in the same manner as in the first hierarchy. However, since there is an upper layer for the hierarchy below the class, the relative coordinate value for the section described above is calculated for the classification symbol with the same classification symbol one level higher. is different from the calculation process of For example, the relative coordinate value calculation unit 12 determines that subgroups “A61B5/11”, “A61B5/16” and “A61B5/18” shown in FIG. , and in other words, the classification symbols one level higher are the same, so the relative coordinate values are calculated as these relative values.

In this way, the relative coordinate value calculation unit 12 calculates relative coordinate values for classes only for the number of higher classes in subgroups to which classification symbols are assigned in the assigned number data. As for the relative coordinate values of the subclasses, the relative coordinate value calculation unit 12 calculates only the number of upper subclasses of the subgroups to which the classification symbols are assigned in the assigned number data. As for the relative coordinate values of the main groups, the relative coordinate value calculation unit 12 calculates only the number of upper main groups of the subgroups to which the classification symbols are assigned in the assigned number data. As for the relative coordinate values of the subgroups, the relative coordinate value calculator 12 calculates only the number of subgroups to which the classification symbols are assigned in the assigned number data. As a result, relative coordinate values are calculated for all classification symbols relating to the set of patent applications in this embodiment.

When calculating the relative coordinate values, the reciprocal of the assigned number may be used as the distance (length) from the origin instead of the assigned order Re. In this case, in the highest hierarchy, relative coordinate values are calculated using the reciprocal of the given number so that the greater the given number among the plurality of classification symbols, the closer to the origin. In the hierarchy below the class, relative coordinate values are calculated using the reciprocal of the number of assignments so that the greater the number of assignments among the plurality of classification symbols included in the same higher hierarchy, the closer to the origin. As a result, classification symbols with a large number of assignments may be brought closer to the origin, and classification symbols with a small number of assignments may be moved further away from the origin.
<Description of Absolute Coordinate Value Calculator>

The absolute coordinate value calculation unit 13 calculates the relative coordinate values of all the upper layers to which the classification symbol belongs and the relative coordinate values of the lowest layer for each of the classification symbols of the lowest layer. The absolute coordinate value in the plane coordinates is calculated by multiplying by another magnification and then adding. FIG. 7 is an example of a data table relating to hierarchical scale factors, which are hierarchical scale ratios. As shown in the figure, the hierarchy-specific magnification M is set to increase as the hierarchy increases. Specifically, as the hierarchical magnification M in this embodiment, the hierarchical magnification M1 of the first hierarchy section is 50, the hierarchical magnification M2 of the second hierarchy class is 15, and the third hierarchy is Initial values are set to 10 for the hierarchical magnification M3 of the subclass, 5 for the fourth hierarchical main group M4, and 3 for the fifth hierarchical subgroup hierarchical magnification M5. In this way, by calculating the absolute coordinate value after multiplying the relative coordinate value of the upper layer by a large number, the technical relationship (closeness, distance) can be expressed as the distance on the plane coordinates. .

For example, in the International Patent Classification, if the sections are different, the technology is technically almost unrelated (distant technology), so the relative coordinate values for the sections are multiplied by the largest hierarchical magnification M as shown in FIG. , the different symbols of the section can be drawn relatively far apart in plane coordinates. On the other hand, for example, if the main group is the same, even if the subgroups are different, the technology is related (close technology) as the main group. By multiplying the scale factor M, different symbols in the subgroup can be drawn relatively close in plane coordinates.

FIG. 8 is a graph schematically showing a method of calculating absolute coordinate values. As shown in the figure, the absolute coordinate value calculator 13 multiplies the X component and the Y component of the relative coordinate value by the hierarchical scale factor M for each layer, and then adds all the X components to obtain the X component of the absolute coordinate value. , and all of the Y components are added to calculate the Y component of the absolute coordinate value. Specifically, the X component Xn of the n-th absolute coordinate value is the X component Xn-1 of the relative coordinate value of the section multiplied by the hierarchical magnification M1 for the section, the class relative X component multiplied by the hierarchical magnification M2 for the class X component Xn-2 of the coordinate value, X component Xn-3 of the relative coordinate value of the subclass multiplied by the hierarchical magnification M3 for the subclass, X component Xn of the relative coordinate value of the main group multiplied by the hierarchical magnification M4 for the main group -4 and the X component Xn-5 of the relative coordinate value of the subgroup multiplied by the hierarchical magnification M5 for the subgroup are all added. Similarly, the Y component Yn of the n-th absolute coordinate value is the Y component Yn-1 of the relative coordinate value of the section multiplied by the hierarchical magnification M1 for the section, and the Y component of the relative coordinate value of the class multiplied by the hierarchical magnification M2 for the class. component Yn-2, the Y component Yn-3 of the relative coordinate value of the subclass multiplied by the hierarchical magnification M3 for the subclass, the Y component Yn-4 of the relative coordinate value of the main group multiplied by the hierarchical magnification M4 for the main group, and , Y component Yn−5 of relative coordinate values of subgroups multiplied by hierarchical magnification M5 for subgroups. In this manner, the absolute coordinate value calculation unit 13 calculates absolute coordinate values for all (1000 in this embodiment) classification symbols in the lowest hierarchy, and outputs the absolute coordinate values to the graph display unit 16 .

As shown in FIG. 7, it is more preferable that the hierarchy-specific magnification ratio M is set so that the higher the hierarchy, the higher the hierarchy-specific magnification ratio M is. As a whole, technical relevance can be expressed as a distance on plane coordinates. For example, even if the same hierarchical magnification M is set for the main group and the subgroups, and a larger hierarchical magnification M is set for the higher layers, the upper hierarchy can be multiplied by a large hierarchical magnification. Also, by multiplying only the relative coordinate values of the lower layers by a positive number smaller than 1, the upper layers may be substantially multiplied by a large scale factor M for each layer. include.
<Description of Symbol Determination Unit>

The symbol determination unit 14 determines the size of the symbol and the appearance of the symbol, such as the shape, pattern, color, and transparency of the form of the symbol, for each of the lowest-level classification symbols for drawing symbols, and graphically displays the symbols as symbol data. Output to unit 16 . In the present embodiment, the symbol determining unit 14 determines to draw the symbol as a semi-transparent single-color circular bubble. Also, the symbol determination unit 14 determines to draw the symbol of the classification symbol with the large number of additions in a large size. FIG. 9 is an example of a data table of symbol color and transparency. Symbol colors are represented by RGB values. Moreover, it is preferable that the symbol colors are different colors, shapes, or patterns between sections so that the user can recognize the boundaries of the sections in the process of drawing the symbols, which will be described later. In other words, by making the color, shape, or pattern of the symbol the same for each group of technologies in the patent classification, the group of technologies can be represented by the color, shape, or pattern of the symbol. This can prevent confusion of symbols between sections, for example.

For this purpose, it is more preferable to set the adjacently displayed sections so as to have colors on the complementary side of the color wheel. Also, it is preferable that the user can change the symbol color based on the displayed graph. It should be noted that it is not always necessary to increase the size of symbols depending on the number of given symbols. For example, a symbol of a classification symbol with a large number of assignments may be drawn in a conspicuous color (for example, dark color or red). Also, the symbols of the classification symbols with a small number of additions may be drawn in a less conspicuous color (for example, light color or blue). In this case, it is preferable to display a legend attached to the graph showing the relationship between the number of grants and the color.

Also, it is preferable to display the symbols semi-transparently. As shown in FIG. 9, the symbol transparency is determined by a percentage (%). For example, when increasing the size of bubbles in proportion to the number of symbols attached, if the bubbles are dense and many bubbles overlap, if the symbol is translucent, the size of the overlapping bubbles can be checked. This is because the number of bubbles to be provided can be grasped as the size of the bubbles even in a dense state.

Also, as the shape of the symbol, the symbol may be drawn with dots in any polygonal or star-like shape instead of the circular bubble. In this case, it is preferable to use the same symbol shape for each group of technologies in the patent classification. As a result, it is possible to draw a symbol indicating a classification symbol that makes it easy to grasp the grouping of technologies. For example, by making the shape of the symbol the same for each section and different from that of other sections, it is possible to recognize the boundary between the sections and prevent the confusion of the symbols between the sections.
<Description of the label determination part>

The label determining unit 15 determines the appearance of the label, such as the contents, size and character color of the label to be overlaid on the symbol drawn on the plane coordinates, and outputs the label to the graph display unit 16 . For example, the label determination unit 15 combines all or part of the classification symbols included in the assigned number data shown in FIG. to determine the content to be drawn as a label. For example, as shown in FIG. 8, the label determination unit 15 arranges the classification symbols, their explanations, and the number of assignments in this order, and determines the content of the label as a character string separated by spaces and commas.

By the way, for example, if the number of symbols drawn on the graph displayed by the graph display unit 16 is large, it is conceivable that the letters of the labels attached to each symbol overlap and become unreadable. In addition, there are many cases where descriptions of more than 100 characters are attached to descriptions of classification symbols. On the other hand, a person skilled in the art can often guess what technology is involved from only the first few letters of the description of the classification symbol. Therefore, the label determining unit 15 can also determine that only the initial few characters of the description of the classification symbol are to be displayed as the label. By drawing the description of the classification symbol in such a small number of characters, the label can be displayed in an easy-to-read manner.

Also, for example, by drawing a larger bubble as the number of given symbols increases, the user can understand the relative amount of the given number. However, the absolute value of the given number cannot be understood only by the size of the bubble. Also, if the number of symbols to be provided is not represented by the size of the symbol, the number of symbols to be provided cannot be known. Therefore, the label determination unit 15 can include the number of assignments in the label. In addition, since the detailed number of applications is not so important when examining the analysis results from a bird's-eye view, it is possible to display the labels in an easy-to-read manner by not displaying the number of granted applications. In addition, by displaying the size of the symbol as a legend together, it is possible not to include the assigned number in the label, thereby making it difficult for the labels of adjacent symbols to overlap each other.

Normally, unless all digits of a particular classification symbol are displayed as labels, it is impossible to identify which classification symbol it is. However, when the present invention is implemented by collecting only the classification symbols of a specific technical field, or when the division of sections can be distinguished by the color of the symbols, the upper layer of the classification symbols can be omitted. For example, when changing the color of a bubble by section, one letter of the section can be omitted and part of the classification symbol can be included in the label.

As described above, the label determination unit 15 determines the content of the label to be displayed and the color and size of the characters for displaying the content of the label as described above for the number of symbols to be displayed on the plane coordinates. Output to the graph display unit 16 as data. Since the label is not necessarily required, the label determining unit 15 can also determine that the label is not displayed. For example, when a graph in which labels are hidden and only bubbles are displayed by the patent analysis device 1 is captured as an image, and explanations that are easy to understand even for those who are not familiar with patent classification are added in post-processing to create materials, it is convenient not to display labels. because it is good.
<Description of graph display>

The graph display unit 16 receives the outputs from the absolute coordinate value calculation unit 13, the symbol determination unit 14, and the label determination unit 15, and displays the coordinates of the absolute coordinate values in plane coordinates (absolute coordinates), and displays a graph in which labels corresponding to the symbols are overlaid. In this way, the relative coordinates are calculated for each layer in the patent classification based on the granted number data, and the coordinates in the plane coordinates of the absolute coordinate values obtained by multiplying these by the magnification M for each layer and then adding the symbol corresponding to the classification symbol. By drawing , the technical relationship can be represented as a distance on the plane coordinates. As a result, it is possible to display a graph in which the symbols indicating the classification symbols are drawn on the plane coordinates while having technical relevance.

Note that the graph display unit 16 may display not only the graph in which the symbols and labels are drawn, but also an input field for prompting the user to input. For example, it is possible to display setting items such as the appearance of the symbol and the content of the label described above, or the scale factor M for each layer so as to be inputtable. Also, an input field in which a correction value, which will be described later, can be input may be displayed.
<Description of Correction Value Acquisition Unit>

The correction value acquiring unit 17 acquires a correction value input by the user who viewed the displayed analysis result graph. When the user examines the graph of the analysis results displayed on the graph display unit 16, it is difficult to understand because the symbols and labels overlap too much, and the symbols are too far apart from each other, causing the area on the plane coordinates to be blurred. There are times when you want to re-display the graph of the analysis results after correcting the position or distribution when taking too wide. In such a case, the correction value acquisition unit 17 receives an input from the user, acquires a correction value, and outputs it to the absolute coordinate value calculation unit 13 as correction value data.

The correction value obtaining unit 17 obtains an addition correction value for correcting the absolute coordinate values for some of the classification symbols and an integrated correction value for correcting the absolute coordinate values for some of the classification symbols. The addition correction value is used to correct the absolute coordinate value by adding it to the absolute coordinate value. The integrated correction value is used to correct the absolute coordinate value by integrating the integrated correction value to the absolute coordinate value.

For example, when two sets of symbols in different sections overlap, it is preferable to move one symbol away. Also, when two sets of symbols are too far apart than necessary, it is preferable to move these groups closer together. In such a case, the absolute coordinate value calculation unit 13 adjusts the position of the symbol by adding a positive or negative addition correction value to the absolute coordinate value for each section, correcting the absolute coordinate value, and then outputting the result to the graph display unit 16 . be able to.

It is also preferable, for example, to scatter the symbols for sections where the symbols are too close together, and to cluster the symbols for sections where the symbols are too far apart. In such a case, the position of the symbol can be adjusted by the absolute coordinate value calculator 13 accumulating and correcting the integrated correction value to the absolute coordinate value for each section and outputting the result to the graph display unit 16 . The graph display unit 16, to which these corrected absolute coordinate values are input, again displays a graph in which symbols and labels are drawn on the corrected absolute coordinates. By performing the correction as described above, the graph display unit 16 displays a graph in which symbols are drawn at positions desired by the user by moving a part of the symbols or scaling a group of symbols. be able to.
<Hardware configuration>

FIG. 10 is a schematic diagram showing an example of the configuration of the patent analysis device when the above functional components are implemented as hardware. With this hardware configuration, the assigned number data acquisition unit 11, the relative coordinate value calculation unit 12, the absolute coordinate value calculation unit 13, the symbol determination unit 14, the label determination unit 15, the graph display unit 16, and the correction value acquisition unit shown in FIG. 17 functions are implemented and various other related computations are performed.

As shown in FIG. 10, the patent analysis apparatus 1 includes a CPU 101 as a central processing unit, a main memory 102 as a main storage device, a non-volatile memory 103 as an auxiliary storage device, and an interface for connecting to external input/output devices. 104 and the like. These pieces of hardware are interconnected by a data communication path such as the system bus 105, and perform transmission/reception and processing of information. Connected to the interface 104 are a mouse 106 and a keyboard 107, which are input devices for receiving information input from the user, a monitor 108 for displaying information to the user, a LAN 109 capable of communicating with an external database via the Internet, and the like. ing.

Various programs and various data are read into the main memory 102, and the CPU 101 executes various arithmetic processing by referring to the read programs and data. A non-volatile memory 103 configured by a hard disk or a semiconductor memory stores patent classification data in which classification symbols of patent classifications and descriptions thereof are associated with each other, and the above-described granted number data. As patent classification data, if it is an international patent classification, classification symbols having a hierarchical structure of sections, classes, subclasses, main groups, and subgroups, and descriptions for each hierarchy of the classification symbols are stored as linked data. be. The non-volatile memory 103 also stores the initial values of the hierarchical magnification M, symbol data, label data, and correction value data. Various programs are stored in the nonvolatile memory 103 .

In the main memory 102, each program such as a given number data acquisition program, a coordinate value calculation program, a graph display program, a correction program, etc. is read out with the user's operation of an input device as a trigger. The CPU 101 interprets the grant number data acquisition program read out to the main memory 102 and performs processing according to the program. For example, in the patent analysis apparatus 1 of the present embodiment, grant number data generated in an external database according to the user's operation of the input device is stored in the nonvolatile memory 103 via the interface 104, the system bus 105 and the LAN 109. be.

Alternatively, the CPU 101, as a pre-process for the processing in the patent analysis apparatus 1 of the present embodiment, uses a dedicated patent analysis program or a spreadsheet program to generate data on the number of grants, which is in CSV (Comma-Separated Values) format received from an external database. Generate the grant number data from the data of Usually, databases provided by patent information providers contain information on filing dates, information on the content of inventions such as the title of the invention, information on the status of examination and registration, and information on various granted patent classifications. etc. are provided in a CSV format file organized by application, publication, or application family. However, since this format cannot be used in the patent analysis apparatus 1 of the present embodiment, it is necessary to add up the assigned number for each classification symbol of the patent classification. Therefore, when the given number data is not provided from the external database, the CPU 101 uses various given number data generation programs to generate the given number data from the CSV format file and stores it in the non-volatile memory 103 .

In addition, the CPU 101 functions as a graph display unit 16 by controlling the monitor 108 according to the graph display program to display a graph of analysis results in which symbols and labels are drawn on plane coordinates and providing the graph to the user. When the user determines that the graph showing the analysis results is difficult to see, and inputs symbol data, label data, correction value data, etc. using the mouse 106 and keyboard 107 as input devices, the non-volatile memory 103 follows the correction program. These data are stored in . In this way, the CPU 101 also functions as the symbol determination unit 14, the label determination unit 15, and the correction value acquisition unit 17 by various programs.
<Process flow>

FIG. 11 is an example of a flowchart showing the process flow of the patent analysis device of the first embodiment. Note that the steps shown below may be steps executed by each hardware configuration of the computer described above, or may be processing steps that constitute a program recorded on an information recording medium and used to control the computer. I do not care.

First, in step S101 (the first acquisition step in the present invention), the CPU 101 functions as the assigned number data acquisition unit 11, and performs is acquired and stored in the non-volatile memory 103 . Here, the CPU 101 acquires the number of grant data (see FIG. 3) from an external database connected via the interface 104, the LAN 109, etc., according to the user's operation using the mouse 106 and keyboard 107, Store in memory 103 . Note that the CPU 101 may generate and acquire the given number data from the data in the CSV format using the given number data generation program.

If patent classification data, which is a combination of a classification symbol and its explanation, is stored in the non-volatile memory 103 as the assigned number data, the process of the present invention can be performed even if only the combination of the classification symbol and its assigned number is acquired. can be executed. In this case, the CPU 101 reads out the description corresponding to the classification symbol of the assigned number data from the non-volatile memory 103 and stores it in the main memory 102 in a state of being linked to the classification symbol. In this way, the CPU 101 acquires combinations of the classification symbols, their descriptions, and the number of assignments of the top 1,000 subgroups in descending order of the number of assignments.

Subsequently, in step S102 (relative coordinate value calculation step in the present invention), the CPU 101 functions as the relative coordinate value calculation unit 12, and for each of the classification symbols of the subgroups in the assigned number data, the classification symbol and its classification symbol Relative coordinate values are calculated based on the assigned number for all layers to which . FIG. 12 is a schematic example of a data table used in the calculation process of this embodiment. In this embodiment, the CPU 101 calculates a total of 5000 relative coordinate values by calculating the relative coordinate values of five layers from the section to the subgroup for the classification symbols of 1000 subgroups. For example, for the first classification symbol "A47C7/62" shown in FIG. is calculated. For the class "A47" of this classification symbol, 0.23 as the X component X1-2 and 0.83 as the Y component Y1-2 of the relative coordinate value are calculated. Regarding the subclass "A47C" of this classification symbol, 0.00 as the X component X1-3 and 0.00 as the Y component Y1-3 of the relative coordinate value are calculated. Regarding the main group "A47C7/00" of this classification symbol, 0.00 as the X component X1-4 and -0.10 as the Y component Y1-4 of the relative coordinate value are calculated. Regarding the subgroup "A47C7/62" of this classification symbol, 0.00 as the X component X1-5 and -0.10 as the Y component Y1-5 of the relative coordinate value are calculated. The numerical values shown in the figure are expressed with a limited number of digits for ease of understanding, but the actual calculation results are calculated using trigonometric functions as shown in FIG. It is expressed using numbers. Similarly, CPU 101 calculates relative coordinate values for the remaining 999 classification symbols, and stores all relative coordinate values in nonvolatile memory 103 .

Next, in step S103 (absolute coordinate value calculation step in the present invention), the CPU 101 functions as the absolute coordinate value calculation unit 13, reads the relative coordinate values from the nonvolatile memory 103, and calculates the absolute coordinate values. In this step, for each subgroup classification symbol, the CPU 101 calculates the relative coordinate values for all upper layers (from section to main group) to which the subgroup classification symbol belongs and the lowest layer (subgroup) The absolute coordinate value is calculated by multiplying the relative coordinate value by the hierarchical scale factor M and then adding all of them. In this embodiment, the CPU 101 multiplies the X component and the Y component of the relative coordinate values shown in FIG. do. As a result, for example, 3.95, which is the X component X1 of the absolute coordinate value of the first classification symbol in the figure, and 49.15, which is the Y component Y1 of the absolute coordinate value of this classification symbol, are calculated. Similarly, CPU 101 calculates absolute coordinate values for the remaining 999 classification symbols, and stores all absolute coordinate values in nonvolatile memory 103 .

Subsequently, in step S104, the CPU 101 functions as the symbol determination unit 14 and determines the appearance of the symbols to be displayed on the plane coordinates. In this embodiment, the CPU 101 reads out the assigned number data from the nonvolatile memory 103, and increases the area of the circular bubble in proportion to the assigned number of classification symbols for each of the classification symbols of the subgroups for which the absolute coordinate values are calculated. Determine the diameter of the bubble so that The diameter of the circular bubble may be determined so as to increase the diameter of the circular bubble in proportion to the number of assigned symbols of the classification symbol. In addition, the CPU 101 determines a single color in which the color of the bubble is different for each section, as shown in FIG. In this way, it is preferable to use the same symbol color, shape, or pattern for each group of technologies in the patent classification. Further, the CPU 101 determines the transparency of the bubbles to be 50% to make them translucent, so that even when the bubbles overlap, the size of each bubble can be determined. The CPU 101 determines the appearance of the symbols for all subgroup classification symbols in the assigned number data, and stores them in the nonvolatile memory 103 as symbol data.

Next, in step S105, the CPU 101 functions as the label determination unit 15, and determines the appearance of the label, such as the content, size, and character color of the label to be overlaid on the bubble drawn on the plane coordinates. In this embodiment, the CPU 101 reads out the assigned number data from the non-volatile memory 103 and determines the classification symbol to be the label. The CPU 101 stores in the nonvolatile memory 103 the content and appearance of labels for all subgroup classification symbols as label data.

Subsequently, in step S106 (graph display step in the present invention), the CPU 101 mainly functions as the graph display unit 16 together with the monitor 108, and draws bubbles at the absolute coordinates of the classification symbols of the subgroups in plane coordinates. FIG. 13 is a graph showing analysis results on the display screen of the patent analysis device 1. FIG. FIG. 14 is an enlarged view of part of the graph on the display screen of the patent analysis device 1. FIG. In these figures, since the color of the bubble shown in FIG. 9 displayed on the actual display screen cannot be identified due to the conversion to gray scale, a single alphabetic character representing the section and the dashed line indicating the boundary between the sections are omitted. are added (the same applies to FIG. 16).

In this step S106, the CPU 101 reads out the absolute coordinate values, the symbol data and the label data from the nonvolatile memory 103 to the main memory 102. FIG. Subsequently, the CPU 101 displays on the monitor 108 a graph drawn based on these pieces of information. Specifically, the CPU 101 mutually compares the absolute coordinate values of the 1000 subgroup classification symbols, and calculates the maximum and minimum values of both the X component and the Y component. Next, the CPU 101 draws XY coordinates in which the maximum and minimum values of the X-axis and Y-axis are set so as to include the range. Also, as shown in FIG. 13, the CPU 101 draws bubbles of 1000 subgroups at the positions (absolute coordinates) of the absolute coordinate values of the subgroups in the XY coordinates. Here, these bubbles are drawn with a diameter such that the area of the bubble is proportional to the given number. Also, these bubbles are drawn with the colors and transparency shown in FIG. As a result, as shown in FIG. 13, semi-transparent bubbles are overlapped in a portion where bubbles are densely packed, so that the image is drawn darkly.

Also, the CPU 101 draws all the labels so as to overlap all the bubbles, as shown in FIG. As shown in the figure, it is possible to confirm how many types of technologies are included in the set to be analyzed from the classification symbols that the bubbles mean and the sizes of the bubbles. Note that it is preferable to enlarge a part of the graph so that it can be displayed on the monitor 108 as shown in FIG. This is because the user can read even if the character of the label is drawn small.

As shown in FIG. 13, in the present embodiment, given number data, which is a combination of 1000 classification symbols, their descriptions, and given numbers, is acquired, so that 1000 symbols and labels are given on the plane coordinates. displayed in a positional relationship according to This makes it possible to express the technical relevance as a positional relationship on the plane coordinates, and to draw the bubbles indicating the classification symbols on the plane coordinates while imparting the technical relevance. Therefore, it is possible to provide an easy-to-understand, bird's-eye view of the technical content of a set of objects to be analyzed, and to provide patent analysis results that are easy to use for examining intellectual property strategies and management strategies. In addition, as shown in the figure, by locating a section with a large number of assignments in the center, it is possible to clearly indicate that it is an important technology (core technology).

However, when displaying in this way, if you want to display the bubbles of the section that the user is paying attention to or some of the classification symbols, or for sections that are more scattered than necessary, you can change the area where the bubbles are drawn. It is also conceivable to have a desire to make it narrower. Therefore, in step S107, the CPU 101 functions as the correction value acquisition unit 17 together with the mouse 106, the keyboard 107, and the monitor 108, which are input devices, and determines whether or not the user needs correction. For this purpose, the CPU 101 accepts an input of a correction value from the user using an input device, for example. FIG. 15 is an example of a data table regarding correction values for each section. The CPU 101 displays, on the monitor 108, an input field for inputting the correction value shown in the figure next to the graph showing the analysis result, for example. Here, although numerical values after input are shown in FIG. Initial values without correction can be displayed.

Then, the CPU 101 determines that correction is necessary when the user inputs the correction value using the input device in step S107. On the other hand, when there is no input, the graph showing the analysis results continues to be displayed, assuming that no correction is required. It should be noted that input fields may be provided in which the hierarchical magnification M shown in FIG. 7, the setting data of the symbols shown in FIG. 9, the content of the label, and the like can be entered. Initial values can also be displayed in these input fields.

When the user inputs an integrated correction value or an added correction value for each section shown in FIG. , and stored in the nonvolatile memory 103 . The integrated correction value in FIG. 15 is a correction value that is integrated to the absolute coordinate value for each section in order to concentrate or disperse the bubbles for each section. The X correction value and the Y correction value are addition correction values to be added to the absolute coordinate values for each section in order to move the bubble in an arbitrary direction for each section on the XY coordinates. For example, section "B" contains the largest number of classification symbols, so it is displayed densely as shown in FIG. Therefore, when it is desired to disperse the bubbles in section "B", by setting the integrated correction value to be greater than 1, the absolute values of the absolute coordinate values are increased and the bubbles can be dispersed on the XY coordinates.

In addition, it is conceivable that the user determines that the position of the bubble in section "B" needs to be corrected because the bubble in section "B" is displayed as being spread out on the entire XY coordinates. In such a case, the additional correction value can be input to the X correction value and Y correction value of section "B" as much as desired to move on the XY coordinates. In addition, since the X correction value of section "B" in FIG. On top, the bubble is moved in the positive direction (upward on the screen). Note that the CPU 101 can also display the scales of the X-axis and the Y-axis and their numerical values when displaying the XY coordinates. This makes it possible to inform the user how far the bubble should be moved in the XY directions, in other words, what values should be input for the X correction value and the Y correction value.

Subsequently, in step S109 (first correction step and second correction step in the present invention), the CPU 101 functions as the absolute coordinate value calculation unit 13 and corrects the absolute coordinate values for each section as part of the classification symbols. Specifically, the CPU 101 integrates the integrated correction value with the absolute coordinate value for each section. Also, the X correction value and the Y correction value are respectively added to the X component and the Y component of the absolute coordinate value after the accumulation correction value is accumulated. As a result, the CPU 101 calculates absolute coordinate values whose distribution and position are corrected according to the user's input. Note that when the hierarchical magnification M is input by the user, the absolute coordinate values are calculated based on the already calculated relative coordinate values and the input hierarchical magnification M, and then the processing using the correction values described above is performed. to run.

Next, when symbol data is input by the user, CPU 101 redetermines the appearance of the symbol in step S104. Further, when the user inputs label data, the CPU 101 re-determines the content of the label in step S105.

Subsequently, the CPU 101 executes step S106 again, and displays a graph in which the bubbles and labels are redrawn so that the graph displayed on the monitor 108 has contents corresponding to the user's input. FIG. 16 is a display screen after correction of the patent analysis device. As shown in FIG. 13, it is possible to disperse and display the bubbles in section "B" and section "F" which were dense in FIG. According to this, it is possible to display the graph so that the user can easily understand the contents of the labels and the distribution of the bubbles, for example.

As described above, the patent analysis device 1 in this embodiment calculates the relative coordinate values for each layer of the classification symbols of the patent classification assigned to a set of about 30,000 patent applications based on the assigned number, Absolute coordinate values are calculated by multiplying the relative coordinate values of the classification symbols of all layers with respect to the classification symbols of 1000 subgroups by the scale factor M for each layer and then adding all of them, and the bubbles and labels can be displayed on the plane coordinates. can. As a result, the technical relevance can be represented as a distance on the plane coordinates, and the bubbles of the classification symbols can be drawn while having the technical relevance as the positional relationship on the plane coordinates. This makes it possible to present technical details in an easy-to-understand, bird's-eye view, and to provide patent analysis results that are easy to use for examining intellectual property strategies and management strategies.
<<Embodiment 2>>

In the above-described embodiment, the analysis was performed for one set of patent applications, but the analysis was performed for the set of multiple sets of patent applications, and the symbols for the sets of multiple sets of patent applications were superimposed on the same plane coordinates. A graph may be displayed. In this case, for example, when two sets of sets are to be analyzed, in step S101 described above, the CPU 101 selects the first set (group A) and the second set (group B), which are sets of arbitrary patent applications. ) and get the grant count data. Subsequently, in step S102, the CPU 101 adds the number of assignments of the same classification symbol to the group A and the group B, and uses the added number of assignments to perform the above-described calculation processing of the relative coordinate values. Next, in step S103, the CPU 101 calculates absolute coordinate values in the same manner as in the embodiment described above. Subsequently, in step S104, the CPU 101 determines symbol colors and the like for group A and group B so that the symbols are different. Next, in step S106, the CPU 101 displays the symbols of group A and group B on the same plane coordinates.

FIG. 17 is a graph schematically showing the analysis results for two groups of patent applications. In the same figure, the number of granted patent application sets for two groups, Group A represented by a bubble hatched with thin slanted lines and Group B represented by a bubble hatched with thick slanted lines, is shown. is obtained, and in the relative coordinate value calculation process, the number of assignments of group A and group B is added, and then the above-described relative coordinate value calculation process is performed. For example, the X components Xn, Xn+1, . . . of the absolute coordinate values and the Y components Yn, Yn+1 . By doing so, the absolute coordinate values of group A and group B are calculated as the same value if they have the same classification symbol.

Also, the colors, shapes, or patterns of the bubbles in these groups are determined to be different from each other so that the user can identify which group the bubbles displayed in the graph belong to. In FIG. 17, the direction and thickness of hatching are made different so that it is possible to identify which group the bubble belongs to. As shown in the figure, the different types of hatching result in a classification symbol that draws only Group A bubbles, a classification symbol that draws only Group B bubbles, and overlapping Group A and Group B bubbles. You can check the classification symbols drawn by

Also, by enlarging the bubbles in proportion to the number of grants to each group, the sizes of the bubbles can be compared between the groups. As a result, technical fields in which there is a possibility that related development is being carried out because applications have been filed by any group, and technical fields in which independent development is being carried out because applications have been filed by only one group etc., can be understood. Here, in the present embodiment, the above-described relative coordinate value calculation processing is performed after adding the values of group A and group B in the relative coordinate value calculation processing. Since the bubbles in these groups are drawn at the same absolute coordinates, they are drawn overlapping. can be done.

According to this embodiment, in addition to the same effects as those of the above-described embodiments, it is possible to draw a plurality of groups of symbols on the same plane coordinates so that they can be compared. As a result, for example, it is possible to compare the application statuses of one's own company and those of other companies, or to grasp technological trends by comparing a set of patent applications over a predetermined period (for example, 10 years). This makes it possible to compare the technologies possessed by companies and to grasp trends, which can be useful in examining various strategies such as intellectual property strategies.
<<Embodiment 3>>

Further, in the above-described embodiment, a configuration example in which a computer (information processing device) owned by a user is used to execute the processing of the present invention has been described. As a patent analysis service, it may be configured to display a graph of analysis results on the user's terminal. FIG. 18 is a schematic diagram showing an example of functional blocks of a patent analysis system connected to user terminals via a network. The patent analysis system 200 shown in FIG. It has an acquisition unit 217 . This patent analysis system 200 is connected to and communicates with a user terminal 400 via a network 300 such as the Internet, thereby transmitting and receiving necessary data. The user terminal 400 comprises, for example, a computer having the same hardware configuration as the patent analysis apparatus 1 shown in FIG. It functions as a functional unit with the input unit 402 that receives user input regarding display.

In the patent analysis system 200, the graph display data output unit 216 does not display a graph in which symbols are drawn at the absolute coordinates of the classification symbols of the lowest hierarchy in plane coordinates, but provides graph display data capable of displaying this graph to the user. The output to the terminal 400 is greatly different from the patent analysis device 1 in the first embodiment. Such graph display data includes related data such as absolute coordinate values, symbol data and label data used for graph display in the first embodiment. Further, the graph display data may include an instruction program for displaying a graph in which symbols, labels, etc. are drawn at absolute coordinates on the plane coordinates in the graph display unit 401 of the user terminal 400 . As a result, the graph display unit 401 of the user terminal 400 that receives the input of the graph display data from the patent analysis system 200 displays the same analysis result graph as in each of the above-described embodiments. Note that the functional units other than the graph display data output unit 216 in the patent analysis system 200 are the functional units with the same name in the patent analysis apparatus 1, except that they receive the output from the input unit 402 of the user terminal 400 and execute processing. Since they have equivalent functions, their description is omitted here.

In the first embodiment described above, the user owns the hardware configuration of the patent analysis device 1 shown in FIG. The patent information provider owns the main part of this hardware configuration and performs the main processing. Further, the present invention can be implemented as a patent analysis service by acquiring grant number data in its own database in response to a user's request and executing the processing of the present invention as the patent analysis system 200 .

On the other hand, in the user terminal 400, the mouse 106 and the keyboard 107, which are input/output devices in FIG. Similarly, the monitor 108 in FIG. 10 constitutes the graph display section 401 . This allows the user to access the patent analysis system 200 from the user terminal 400 via the network 300 and exchange data. Then, the user sends and receives various data to and from the patent analysis system 200 using an Internet browser or dedicated software, thereby causing the patent analysis system 200 to execute the processing of the above-described embodiment including the acquisition of the granted number data, and graph The display data output unit 216 is caused to output the graph display data. As a result, a graph drawn with symbols and labels is displayed as the analysis result on the graph display unit 401 of the user terminal 400 . In this way, the patent analysis system 200 connected to the user terminal 400 via the network 300 can have the same effect as the above embodiment.
<<Modification>>

In the above-described embodiment, the absolute coordinate values are calculated with the subgroup as the lowest hierarchy, but the configuration example is not calculated at the lowest hierarchy in the patent classification (for example, the international patent classification) used for the analysis target. can also For example, the absolute coordinate values may be calculated with the classification symbol of the intermediate hierarchy as the lowest hierarchy. For example, a graph in which absolute coordinate values are calculated and symbols are displayed may be displayed with the subclass as the third hierarchy and the main group as the fourth hierarchy in the International Patent Classification as the lowest hierarchy. According to this, it is possible to reduce the number of symbols to be displayed and display the graph of the analysis result in a straightforward and bird's-eye view. On the other hand, in the International Patent Classification, etc., a hierarchy represented by dots that further subdivide subgroups and a lower hierarchy are set by adding alphanumeric characters after several digits of subgroups. and these hierarchies may be the subject of analysis. As a result, it is possible to analyze the classification symbols in which the techniques are further subdivided, and to display the graph of the detailed analysis results.

As a patent classification, the F-term granted only in Japan may be used. In this case, the first layer is the "theme", the second layer is the "point of view", and the third layer is the "two-digit number" attached after the point of view. As a result, detailed patent analysis can be performed by displaying F-term classification symbols, which are classified more finely depending on the technical field than the international patent classification, on the plane coordinates while maintaining technical relevance. It can be carried out. Alternatively, the USPC granted only in the United States may be used.

Further, in step S106 of FIG. 11, the CPU 101 may display the label on the side or above the symbol in the graph showing the analysis result, instead of superimposing the label on the symbol. When the label is displayed near the symbol, a line connecting them may be displayed on the plane coordinates. Also, the label may be movable on the plane coordinates by the operation of the input device by the user.

Also, the graph displayed in step S106 may be changed according to the passage of time or user's operation. In this case, the CPU 101 changes the symbols according to the elapsed time or the user's operation while the graph displayed in step S106 is displayed. For example, the appearance of the symbol (for example, bubble size) can also be changed. In this case, the number of applications for the number of years from the first year to the last year displayed in the graph is obtained as the number of applications granted, and the number of applications for the last year as the data with the largest number of applications is the number of applications granted, and the relative coordinate value is calculated. calculate. Then, the size of the bubble indicating the classification symbol can be changed based on the number of grants (the number of applications) in the corresponding year according to the elapsed time while the graph is displayed. As a result, it is possible to display the graph of the analysis result while changing it so that the transition of the number of applications can be understood, and it is possible to easily grasp the technological trend.

Further, in the above-described embodiment, an example of calculating the number of grants such as "number of grants", "number of species granted" or "number of grants" based on the number of applications and using it to calculate the relative coordinate value has been described. The number of grants may be calculated based on the number of applications other than For example, a patent family number that indicates a group of patent applications claiming priority based on the same patent application, a registration number that indicates the number of patent applications that were patented and registered after filing, or a patent that was maintained after registration. The number of grants may be calculated based on the number of pending patent applications, which is the number of pending patent applications.

The patent analysis method, patent analysis device, patent analysis program, information recording medium, and patent analysis system of the present invention can display a graph of patent analysis results in an easy-to-understand, overview manner.

1 Patent analysis device 11, 211 Assignment number data acquisition unit 12, 212 Relative coordinate value calculation unit 13, 213 Absolute coordinate value calculation unit 14, 214 Symbol determination unit 15, 215 Label determination unit 16, 401 Graph display unit 17, 217 Correction Value acquisition unit 101 CPU
102 main memory 103 nonvolatile memory 104 interface 105 system bus 106 mouse 107 keyboard 108 monitor 109 LAN
200 Patent analysis system 216 Graph display data output unit M, M1 to M5 Hierarchical magnification θ1, θ1a, θ1b Allocation angle θ2, θ2a to θ2h Coordinate angle

1 Patent analysis device 11, 211 Assignment number data acquisition unit 12, 212 Relative coordinate value calculation unit 13, 213 Absolute coordinate value calculation unit 14, 214 Symbol determination unit 15, 215 Label determination unit 16, 401 Graph display unit 17, 217 Correction Value acquisition unit 101 CPU
102 main memory 103 nonvolatile memory 104 interface 105 system bus 106 mouse 107 keyboard 108 monitor 109 LAN
200 Patent analysis system 216 Graph display data output unit M, M1 to M5 Hierarchical magnification θ1, θ1a to θ1h Allocation angle θ2, θ2a to θ2h Coordinate angle

Claims (18)

A patent classification that has a multi-layered hierarchical structure that includes a top layer that classifies technologies most broadly and sub-layers that subdivide the upper layers including the top layer, and that classifies technologies with multiple classification symbols. A patent analysis method for analyzing the number of classification symbols assigned to a set of arbitrary patent applications using
a first acquisition step of acquiring assigned number data having a plurality of combinations of the classification symbol assigned to the set of arbitrary patent applications and the assigned number assigned with the classification symbol;
A relative coordinate value indicating the relative coordinates of the classification symbol in the first hierarchy, which is the highest level, and the relative coordinates of the classification symbol in the second hierarchy and below where the classification symbol in the hierarchy one level higher is the same and a relative coordinate value calculation step of calculating based on the given number data;
For each of the classification symbols of the lowest hierarchy, after multiplying the relative coordinate values of all the upper hierarchies to which the classification symbol belongs and the relative coordinate values of the lowest hierarchy by a large scaling factor for each higher hierarchy an absolute coordinate value calculation step of calculating an absolute coordinate value in plane coordinates by addition;
a graph display step of displaying a graph in which the symbols of the classification symbols are drawn at the coordinates of the absolute coordinate values in the plane coordinates;
A patented analysis method with wherein the patent classification is the International Patent Classification, File Index, Common Patent Classification or European Patent Classification,
The first hierarchy, which is the highest hierarchy, is a section, the second hierarchy that subdivides the first hierarchy is a class, the third hierarchy that subdivides the second hierarchy is a subclass, and the third hierarchy that subdivides the third hierarchy is a class. 2. The patent analysis method according to claim 1, wherein the fourth hierarchy is the main group and the fifth hierarchy subdividing the fourth hierarchy is the subgroup. The patent classification is F-term,
2. The method according to claim 1, wherein the first hierarchy, which is the highest hierarchy, is a theme, the second hierarchy subdividing the first hierarchy is a viewpoint, and the third hierarchy subdividing the second hierarchy is a two-digit number. Patent analysis method. 2. The patent analysis method according to claim 1, wherein, in said relative coordinate value calculating step, said relative coordinate values are calculated so that said classification symbols are arranged in the order of arrangement in said patent classification while enclosing an origin. 5. The patent analysis method according to claim 4, wherein, in said relative coordinate value calculating step, said relative coordinate value is calculated using a trigonometric function while adding a larger angle as said number of assigned classification symbols increases. 2. The patent analysis method according to claim 1, wherein, in said relative coordinate value calculating step, said relative coordinate values are calculated so that said classification symbols are brought closer to the origin in order of increasing number of assigned numbers. 2. The patent analysis method according to claim 1, wherein, in said relative coordinate value calculating step, said relative coordinate values are calculated such that said classification symbols are brought closer to the origin as said given number increases. a second obtaining step of obtaining addition correction values for correcting the absolute coordinate values for some of the classification symbols;
2. The patent analysis method according to claim 1, further comprising a first correction step of correcting by adding said addition correction value to said absolute coordinate value. a third obtaining step of obtaining integrated correction values for correcting the absolute coordinate values for some of the classification symbols;
2. The patent analysis method according to claim 1, further comprising a second correction step of correcting the absolute coordinate value by integrating the integrated correction value. In the graph display step, all or part of the classification symbols, all or part of the description of the classification symbols, and at least one or a plurality of the assigned numbers are combined to display the label on the plane coordinates as a label. 2. The patent analysis method according to claim 1, wherein the drawing is superimposed on the symbol. 2. The patent analysis method according to claim 1, wherein, in said graph display step, said symbols are drawn with dots of any shape or bubbles represented by circles. 2. The patent analysis method according to claim 1, wherein, in said graph display step, said symbols of said classification symbols with a large number of assignments are drawn large. 2. The patent analysis method according to claim 1, wherein, in said graph display step, the color, shape or pattern of said symbol is set to be the same for each group of technologies in said patent classification. In the first obtaining step, obtaining the granted number data for a plurality of sets that are a set of the arbitrary patent applications;
In the relative coordinate value calculation step, adding the given number with the same classification symbol for the plurality of sets, and calculating the relative coordinate value using the added given number;
2. The patent analysis method according to claim 1, wherein, in said graph display step, said symbols of said plurality of sets are displayed on the same plane coordinates with mutually different colors, shapes or patterns. A patent classification that has a multi-layered hierarchical structure that includes a top layer that classifies technologies most broadly and sub-layers that subdivide the upper layers including the top layer, and that classifies technologies with multiple classification symbols. A patent analysis device that analyzes the number of classification symbols assigned to a set of arbitrary patent applications using
a grant number data acquisition unit that acquires grant number data having a plurality of combinations of the classification symbol assigned to the set of arbitrary patent applications and the grant number to which the classification symbol is assigned;
A relative coordinate value indicating the relative coordinates of the classification symbol in the first hierarchy, which is the highest level, and the relative coordinates of the classification symbol in the second hierarchy and below where the classification symbol in the hierarchy one level higher is the same and a relative coordinate value calculation unit that calculates based on the given number data,
For each of the classification symbols of the lowest hierarchy, after multiplying the relative coordinate values of all the upper hierarchies to which the classification symbol belongs and the relative coordinate values of the lowest hierarchy by a large scaling factor for each higher hierarchy an absolute coordinate value calculation unit that calculates an absolute coordinate value in plane coordinates by addition;
a graph display unit for displaying a graph in which the symbols of the classification symbols are drawn at the coordinates of the absolute coordinate values in the plane coordinates;
A patent analyzer with A patent classification that has a multi-layered hierarchical structure that includes a top layer that classifies technologies most broadly and sub-layers that subdivide the upper layers including the top layer, and that classifies technologies with multiple classification symbols. A patent analysis program that causes an information processing device to analyze the number of classification symbols assigned to a set of arbitrary patent applications using
a first acquisition step of acquiring assigned number data having a plurality of combinations of the classification symbol assigned to the set of arbitrary patent applications and the assigned number assigned with the classification symbol;
A relative coordinate value indicating the relative coordinates of the classification symbol in the first hierarchy, which is the highest level, and the relative coordinates of the classification symbol in the second hierarchy and below where the classification symbol in the hierarchy one level higher is the same and a relative coordinate value calculation step of calculating based on the given number data;
For each of the classification symbols of the lowest hierarchy, after multiplying the relative coordinate values of all the upper hierarchies to which the classification symbol belongs and the relative coordinate values of the lowest hierarchy by a large scaling factor for each higher hierarchy an absolute coordinate value calculation step of calculating an absolute coordinate value in plane coordinates by addition;
a graph display step of displaying a graph in which the symbols of the classification symbols are drawn at the coordinates of the absolute coordinate values in the plane coordinates;
A patent analysis program that causes the information processing device to execute the patent analysis program. 17. An information recording medium on which the patent analysis program according to claim 16 is readable and recorded in said information processing apparatus. A patent classification that has a multi-layered hierarchical structure that includes a top layer that classifies technologies most broadly and sub-layers that subdivide the upper layers including the top layer, and that classifies technologies with multiple classification symbols. A patent analysis system that analyzes the number of classification symbols assigned to a set of arbitrary patent applications using
a grant number data acquisition unit that acquires grant number data having a plurality of combinations of the classification symbol assigned to the set of arbitrary patent applications and the grant number to which the classification symbol is assigned;
A relative coordinate value indicating the relative coordinates of the classification symbol in the first hierarchy, which is the highest level, and the relative coordinates of the classification symbol in the second hierarchy and below where the classification symbol in the hierarchy one level higher is the same and a relative coordinate value calculation unit that calculates based on the given number data,
For each of the classification symbols of the lowest hierarchy, after multiplying the relative coordinate values of all the upper hierarchies to which the classification symbol belongs and the relative coordinate values of the lowest hierarchy by a large scaling factor for each higher hierarchy an absolute coordinate value calculation unit that calculates an absolute coordinate value in plane coordinates by addition;
a graph display data output unit for outputting data capable of displaying a graph in which the symbols of the classification symbols are drawn at the coordinates of the absolute coordinate values in the plane coordinates;
A patent analysis system with

Images