Program, patent information processing device, and patent information processing method

Abstract

To solve the problem that important information could not be conventionally acquired from patent information by appropriately using a regenerative AI.SOLUTION: Important information can be acquired from patent information by appropriately using a regenerative AI by a patent information processing device 1 having a prompt source storage section 111 for storing a subject prompt source as a source for acquiring a prompt for acquiring subject information regarding a subject of a patent, an information reception section 121 for receiving patent specific information for specifying a patent for acquiring subject information, a patent information acquisition section 130 for acquiring patent information regarding the content of a patent corresponding to patent specific information, a subject prompt acquisition section 134 for acquiring a subject prompt for acquiring subject information by using a subject prompt source and patent information, a subject acquisition section 135 for passing a subject prompt acquired by the subject prompt acquisition section 134 to a regenerative AI and acquiring subject information from the regenerative AI, and a subject output section 141 for outputting subject information acquired by the subject acquisition section 135.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a patent information processing device that uses generative AI to extract and output important information from patent information. [Background technology]

[0002] Previously, there was a patent map creation system that was said to create patent maps that made it easy to determine the relationships and similarities between multiple inventions by referring to drawings (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2005-38240 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the prior art, it was not possible to appropriately use generative AI to obtain important information from patent information, such as the problem or the solution. [Means for solving the problem]

[0005] The patent information processing device of the first invention is a patent information processing device that includes a prompt source storage unit in which a task prompt source from which a prompt for obtaining task information related to a patent task is stored, an information receiving unit that receives patent specific information that identifies the patent from which task information is to be obtained, a patent information acquisition unit that acquires patent information related to the content of the patent corresponding to the patent specific information, a task prompt acquisition unit that uses the task prompt source and the patent information to obtain a task prompt for obtaining task information, a task acquisition unit that passes the task prompt acquired by the task prompt acquisition unit to a generation AI and obtains task information from the generation AI, and a task output unit that outputs the task information acquired by the task acquisition unit.

[0006] With this configuration, the generation AI can be appropriately used to obtain important issue information from patent information.

[0007] In addition, compared to the first invention, the patent information processing device of the second invention is a patent information processing device in which a prompt source storage unit stores a solution prompt source that is the source of acquiring a prompt for acquiring solution information related to a solution that solves the problem of the patent, and which is equipped with a solution prompt acquisition unit that acquires a solution prompt for acquiring the solution information using the solution prompt source and patent information, a solution acquisition unit that passes the solution prompt acquired by the solution prompt acquisition unit to the generation AI and acquires the solution information from the generation AI, and a solution output unit that outputs the solution information acquired by the solution acquisition unit.

[0008] With this configuration, by appropriately using the generation AI, important solution information can be obtained from patent information.

[0009] In addition, the patent information processing device of the third invention, compared to the first invention, further includes a task judgment unit that judges whether the task information acquired by the task acquisition unit satisfies task conditions, which are conditions that the task information should satisfy, and if the task acquisition unit determines that the task judgment unit does not satisfy the task conditions, the task acquisition unit again passes a task prompt to the generation AI and acquires task information from the generation AI, and the task output unit outputs task information that satisfies the task conditions.

[0010] With this configuration, by appropriately using the generation AI, more appropriate problem information can be obtained from patent information.

[0011] In addition, the patent information processing device of the fourth invention, compared to the first invention, further includes a solution means judgment unit which judges whether the solution means information acquired by the solution means acquisition unit satisfies solution means conditions which are conditions that the solution means information should satisfy, and if the solution means acquisition unit judges that the solution means conditions are not satisfied, the solution means acquisition unit again passes a solution means prompt to the generation AI and acquires the solution means information from the generation AI, and the solution means output unit outputs the solution means information which satisfies the solution means conditions.

[0012] With this configuration, by appropriately using the generation AI, more appropriate solution information can be obtained from patent information.

[0013] In addition, the patent information processing device of the fifth invention is a patent information processing device in which, compared to the first invention, the issue prompt source is information from which a prompt for acquiring issue information among two or more issue candidate information that are candidates for issue information is acquired, and the issue prompt acquisition unit uses the issue prompt source and patent information to acquire a prompt for acquiring issue information among the two or more issue candidate information.

[0014] With this configuration, by appropriately using the generation AI, appropriate problem information for the patent information can be obtained from two or more candidate problem information.

[0015] In addition, the patent information processing device of the sixth invention is a patent information processing device in which, compared to the fifth invention, the task prompt acquisition unit acquires a task prompt for acquiring task information indicating "other" when the task information does not correspond to any of two or more candidate task information that are candidates for task information, and the task acquisition unit may acquire "other" task information, which is task information that does not correspond to any of the two or more candidate task information.

[0016] With this configuration, the generation AI can be appropriately used to obtain appropriate issue information for patent information.

[0017] In addition, the patent information processing device of the seventh invention is a patent information processing device in which, compared to the sixth invention, the information receiving unit receives two or more patent-specific information, and the task prompt acquisition unit acquires, for each of the two or more patent-specific information, a task prompt for acquiring task information using the task prompt source and patent information, and the task acquisition unit passes the task prompt acquired by the task prompt acquisition unit to the generation AI for each of the two or more patent-specific information, acquires the task information from the generation AI, acquires the number of ``other'' task information, and further includes an other judgment unit that determines whether the number satisfies the retry condition, and when the other judgment unit determines that the retry condition is satisfied, the task acquisition unit again passes the task prompt to the generation AI for the patent-specific information corresponding to the ``other'' task information, and acquires the task information from the generation AI.

[0018] With this configuration, the generation AI can be appropriately used to reduce the number of patent information whose issue information is "other."

[0019] In addition, the patent information processing device of the eighth invention is a patent information processing device in which, compared to the first invention, the information receiving unit receives two or more patent-specific information, the task prompt acquisition unit acquires, for each of the two or more patent-specific information, a task prompt for acquiring task information using the task prompt source and patent information, the task acquisition unit passes the task prompt acquired by the task prompt acquisition unit to the generation AI for each of the two or more patent-specific information, acquires task information from the generation AI, and further includes a task aggregation unit that acquires one or more task information using two or more different task information acquired by the task acquisition unit, and the task output unit outputs the one or more task information acquired by the task aggregation unit rather than the task information acquired by the task acquisition unit.

[0020] With this configuration, by appropriately using the generation AI, an appropriate number of types of problem information can be obtained for two or more patent information without the need to prepare candidate problem information.

[0021] In addition, the patent information processing device of the ninth invention is a patent information processing device in which, compared to the second invention, the solution prompt source is information that serves as a source for obtaining a prompt for obtaining any one of two or more candidate solution means information that are candidates for solution means information, and the solution prompt acquisition unit uses the solution prompt source and patent information to obtain a prompt for obtaining any one of the two or more candidate solution means information.

[0022] With this configuration, by appropriately using the generation AI, appropriate solution information for the patent information can be obtained from two or more candidate solution information.

[0023] Furthermore, the patent information processing device of the tenth invention is a patent information processing device in which, compared to the ninth invention, the solution prompt acquisition unit acquires a solution prompt for acquiring solution information indicating "other" when the solution information does not correspond to any of two or more candidate solution information that are candidates for the solution information, and the solution acquisition unit may acquire "other" solution information, which is solution information that does not correspond to any of the two or more candidate solution information.

[0024] With this configuration, the generation AI can be appropriately used to obtain appropriate solution information for patent information.

[0025] In addition, the patent information processing device of the eleventh invention is a patent information processing device in which, compared to the tenth invention, the information receiving unit receives two or more patent-specific information, and the solution prompt acquisition unit acquires, for each of the two or more patent-specific information, a solution prompt for acquiring solution information using the solution prompt source and patent information, and the solution acquisition unit passes the solution prompt acquired by the solution prompt acquisition unit to the generation AI for each of the two or more patent-specific information, acquires the solution information from the generation AI, acquires the number of ``other'' solution information, and further includes an other judgment unit that determines whether the number satisfies the retry condition, and when the other judgment unit determines that the retry condition is satisfied, the problem acquisition unit again passes the solution prompt to the generation AI for the patent specific information corresponding to the ``other'' solution information, and acquires the solution information from the generation AI.

[0026] With this configuration, by appropriately using the generation AI, the number of patent information whose solution information is "other" can be reduced.

[0027] In addition, the patent information processing device of the twelfth invention is a patent information processing device in which, compared to the second invention, the information receiving unit receives two or more patent-specific information, the solution prompt acquisition unit acquires, for each of the two or more patent-specific information, a solution prompt for acquiring solution information using the solution prompt source and patent information, the solution acquisition unit further includes a solution aggregation unit which passes the solution prompt acquired by the solution prompt acquisition unit to the generation AI for each of the two or more patent-specific information, acquires solution information from the generation AI, and acquires one or more solution information using two or more different solution information acquired by the solution acquisition unit, and the solution output unit outputs the one or more solution information acquired by the solution aggregation unit rather than the solution information acquired by the solution acquisition unit.

[0028] With this configuration, by appropriately using the generation AI, it is possible to obtain an appropriate number of types of solution information for two or more patent information without the need to prepare candidate solution information.

[0029] In addition, the patent information processing device of the thirteenth invention, compared to the first invention, further comprises a task pre-processing unit that acquires task information using patent information related to the content of the patent corresponding to the patent specific information, and a judgment pre-processing unit that determines whether the task information acquired by the task pre-processing unit satisfies the AI ​​processing conditions for acquiring task information from the generation AI without using the task information, and when the judgment pre-processing unit determines that the AI ​​processing conditions are satisfied, the task acquisition unit passes the task prompt acquired by the task prompt acquisition unit to the generation AI and acquires the task information from the generation AI.

[0030] With this configuration, the generation AI can be used appropriately to obtain issue information from patent information only when necessary.

[0031] In addition, the patent information processing device of the fourteenth invention, compared to the second invention, further comprises a solution means pre-processing unit that acquires solution means information using patent information related to the content of the patent corresponding to the patent specifying information, and a judgment pre-processing unit that judges whether the solution means information acquired by the solution means pre-processing unit satisfies the AI ​​processing conditions for acquiring the solution means information from the generation AI without using the solution means information, and when the judgment pre-processing unit judges that the solution means information satisfies the AI ​​processing conditions, the solution means acquisition unit passes the solution means prompt acquired by the solution means prompt acquisition unit to the generation AI and acquires the solution means information from the generation AI.

[0032] With this configuration, solution information can be obtained from patent information only when necessary by appropriately using the generation AI.

[0033] In addition, the patent information processing device of the fifteenth invention is a patent information processing device according to the second invention, further comprising: an information receiving unit receives two or more patent-specific information; a problem prompt acquisition unit acquires, for each of the two or more patent-specific information, a problem prompt for acquiring problem information using a problem prompt source and patent information; a solution means prompt acquisition unit acquires, for each of the two or more patent-specific information, a solution means prompt for acquiring solution information using the solution means prompt source and patent information; the problem acquisition unit passes the problem prompt acquired by the problem prompt acquisition unit to the generation AI for each of the two or more patent-specific information and acquires problem information from the generation AI; the solution means acquisition unit passes the solution means prompt acquired by the solution means prompt acquisition unit to the generation AI for each of the two or more patent-specific information and acquires solution information from the generation AI; a map construction unit constructs a patent map having a problem axis having two or more problem information acquired by the problem acquisition unit and a solution means axis having two or more solution means information acquired by the solution means acquisition unit; and a map output unit outputs the patent map constructed by the map construction unit.

[0034] With this configuration, by appropriately using generation AI, a patent map centered on problems and solutions can be created. Effect of the Invention

[0035] According to the patent information processing device of the present invention, important information can be obtained from patent information by appropriately using generation AI. [Brief description of the drawings]

[0036] [Figure 1] Conceptual diagram of information system A in embodiment 1. [Diagram 2] Block diagram of Information System A [Diagram 3] Block diagram of the patented information processing device 1 [Figure 4] A flowchart explaining an example of the operation of the patent information processing device 1. [Diagram 5] A flowchart illustrating an example of the pre-processing [Figure 6]A flowchart explaining an example of the AI ​​processing for this project [Figure 7] A flowchart explaining an example of the AI ​​processing of the solution [Figure 8] A flowchart explaining an example of post-processing for the same assignment [Figure 9] A flowchart for explaining an example of post-processing of the solving means [Figure 10] A flowchart explaining an example of the patent map creation process [Figure 11] A diagram showing an example of the prompt source for this assignment. [Figure 12] FIG. 13 shows an example of a prompt source for the solution. [Figure 13] A diagram showing an example of the prompt source [Figure 14] A diagram showing an example of the patent list [Figure 15] A diagram showing an example of the patent map [Figure 16] A diagram showing an example of the prompt source [Figure 17] A diagram showing an example of the patent list [Figure 18] Block diagram of patent information processing device 3 in embodiment 2 [Figure 19] A flowchart explaining an example of the operation of the patent information processing device 3 [Figure 20] A flowchart explaining an example of the task aggregation process [Figure 21] A flowchart illustrating an example of the solution means aggregation process. [Figure 22] Overview of the computer system according to the above embodiment. [Figure 23] Block diagram of the computer system DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0037] Hereinafter, an embodiment of a patent information processing device and the like will be described with reference to the drawings. Note that components with the same reference numerals in the embodiments perform similar operations, so repeated description may be omitted.

[0038] (Embodiment 1) In this embodiment, a patent information processing device is described which provides a task acquisition prompt using patent information to a generation AI, acquires task information from the generation AI, and outputs the task information.

[0039] In addition, in this embodiment, a patent information processing device will be described that gives a solution acquisition prompt using patent information to a generation AI, acquires solution information from the generation AI, and outputs it.

[0040] In addition, in this embodiment, we will describe a patent information processing device that, if the acquired assignment information does not satisfy the assignment conditions, again gives an assignment acquisition prompt to the generation AI, again acquires assignment information from the generation AI, and outputs it.

[0041] In addition, in this embodiment, a patent information processing device is described which, if the solution information does not satisfy the solution condition, again sends a solution acquisition prompt to the generation AI, again acquires the solution information from the generation AI, and outputs it.

[0042] In addition, in this embodiment, a case will be described in which the assignment acquisition prompt is a prompt that instructs the generation AI to select one piece of assignment information from two or more pieces of assignment information.

[0043] In addition, in this embodiment, a case will be described in which the assignment acquisition prompt is a prompt that also allows "other" assignment information.

[0044] In addition, in this embodiment, we will describe a patent information processing device that, if the number of ``Other'' satisfies the retry condition, again gives a prompt to the generation AI regarding the patent information corresponding to ``Other,'' and again obtains and outputs the issue information from the generation AI.

[0045] In addition, in this embodiment, a case will be described in which the solution acquisition prompt is a prompt for selecting one solution information from two or more pieces of solution information.

[0046] In addition, in this embodiment, a case will be described in which the solution acquisition prompt is a prompt that also allows "other" solution information.

[0047] In addition, in this embodiment, we will describe a patent information processing device that, when the number of "Other" satisfies the retry condition, again gives a prompt to the generation AI regarding the patent information corresponding to "Other", again obtains solution information from the generation AI, and outputs it.

[0048] In addition, in this embodiment, we will describe a patent information processing device that performs preprocessing to obtain assignment information without using a generation AI, and if the assignment information cannot be obtained, gives an assignment acquisition prompt to the generation AI, obtains the assignment information from the generation AI, and outputs it.

[0049] In addition, in this embodiment, we will describe a patent information processing device that performs preprocessing to obtain solution information without using the generation AI, and if the solution information cannot be obtained, gives a solution acquisition prompt to the generation AI, obtains the solution information from the generation AI, and outputs it.

[0050] Furthermore, in this embodiment, a patent information processing device that creates a patent map for two or more patents, with problem information and solution information as axes, will be described.

[0051] In this specification, information X being associated with information Y means that information Y can be obtained from information X, or information X can be obtained from information Y, and the method of association is not important. Information X and information Y may be linked, may exist in the same buffer, information X may be included in information Y, or information Y may be included in information X, etc.

[0052] Furthermore, in this specification, selecting or determining information Z means obtaining information Z, obtaining a pointer to information Z, obtaining the ID of information Z, setting a flag for information Z, etc., and is sufficient as long as information Z can be accessed.

[0053] 1 is a conceptual diagram of an information system A in this embodiment. The information system A includes a patent information processing device 1 and one or more terminal devices 2. The information system A may also include one or more generating AI devices X.

[0054] The patent information processing device 1 is a device that acquires and outputs patent information issues, etc. The patent information processing device 1 is usually a server, but may be a standalone device. When the patent information processing device 1 is a server, the patent information processing device 1 is, for example, a so-called cloud server or ASP server, but it does not matter.

[0055] The terminal device 2 is a terminal used by a user. The terminal device 2 is, for example, a so-called personal computer, a tablet terminal, or a smartphone.

[0056] The generation AI device X is a server in which software having the function of the generation AI is stored. The function of the generation AI here is to generate text that is an answer to a given prompt. The generation AI is, for example, Chat-GPT, but it does not matter.

[0057] The patent information processing device 1 and one or more terminal devices 2 can usually communicate with each other via a network such as the Internet. Also, the patent information processing device 1 and the generation AI device X can usually communicate with each other via a network such as the Internet.

[0058] Fig. 2 is a block diagram of the information system A in this embodiment. Fig. 3 is a block diagram of the patent information processing device 1.

[0059] The patent information processing device 1 includes a storage unit 11, a reception unit 12, a processing unit 13, and an output unit 14. The storage unit 11 includes a prompt source storage unit 111, a problem candidate management unit 112, and a solution candidate management unit 113. The reception unit 12 includes an information reception unit 121. The processing unit 13 includes a patent information acquisition unit 130, a problem preprocessing unit 131, a solution preprocessing unit 132, a judgment preprocessing unit 133, a problem prompt acquisition unit 134, a problem acquisition unit 135, a problem judgment unit 136, a solution prompt acquisition unit 137, a solution acquisition unit 138, a solution judgment unit 139, a other judgment unit 13A, and a map construction unit 13B. The output unit 14 includes a problem output unit 141, a solution output unit 142, and a map output unit 143.

[0060] The terminal device 2 includes a terminal storage unit 21, a terminal reception unit 22, a terminal processing unit 23, a terminal transmission unit 24, a terminal reception unit 25, and a terminal output unit .

[0061] Various information is stored in the storage unit 11 constituting the patent information processing device 1. The various information is, for example, various prompt sources described later, problem candidate information described later, solution means candidate information described later, and various conditions described later.

[0062] The prompt source storage unit 111 stores one or more prompt sources. A prompt source is information that is the source of a prompt to be given to a generation AI. A prompt source is usually a sentence. A prompt source usually has variables. A prompt source has, for example, variables into which patent information is inserted. A prompt source is, for example, a problem prompt source, a solution prompt source, and an aggregate prompt source. A prompt source may be the prompt itself.

[0063] Patent information is information that specifies the contents of a patent. Patent information is usually text showing the contents of a patent, but it can also be a drawing. Patent information is a patent specification, a patent claim, an abstract, a part of a patent specification, a part of a patent claim, or a part of an abstract. A part of a patent specification is, for example, [Problem to be solved by the invention] or [Means for solving the problem] or [Effect of the invention] or [Form for carrying out the invention] in the specification. A part of a patent claim is, for example, claim 1 or one or more independent claims. A part of an abstract is, for example, [Problem] or [Solution]. Needless to say, the language of patent information is not important. The language of patent information is, for example, Japanese, English, Chinese, or other language.

[0064] The source of the issue prompt is information that is the source for obtaining a prompt to obtain issue information. Issue information is information about the issue of a patent. A patent issue is a problem that a patent invention is intended to solve. A patent issue may also be the effect of a patent, which is the flip side of the issue of the patent. In other words, issue information may be information about the effect, which is the flip side of the issue. It goes without saying that the language of the issue information is not important. The language of the issue information may be, for example, Japanese, English, Chinese, or another language. Furthermore, the language of the issue information may be different from the language of the patent information.

[0065] The assignment prompt source may be information that is a source for acquiring a prompt for acquiring any one of two or more pieces of assignment candidate information that are candidates for assignment information. Such two or more pieces of assignment candidate information are information of the assignment candidate management unit 112.

[0066] The solution prompt source is information that is the source for obtaining a prompt to obtain solution information. Solution information is information related to a solution that solves a problem of a patent. A solution is, for example, a component that constitutes an invention. It goes without saying that the language of the solution information is not important. The language of the solution information is, for example, Japanese, English, Chinese, or another language. Furthermore, the language of the solution information may be different from the language of the patent information.

[0067] The solution prompt source may be information that is a source for acquiring a prompt for acquiring any one of two or more pieces of solution candidate information that are candidates for solution information. Such two or more pieces of solution candidate information are information of the solution candidate management unit 113.

[0068] The problem prompt source and the solution means prompt source may be included in one prompt source.

[0069] The aggregation prompt source is information that is the source of a prompt for aggregating a plurality of pieces of problem information or a plurality of pieces of solution means information into one or more pieces of problem information or one or more pieces of solution means information.

[0070] The aggregation prompt source is, for example, a problem information aggregation prompt source and a solution means aggregation prompt source. The problem information aggregation prompt source is a prompt source for aggregating two or more pieces of problem information into one or more pieces of problem information. The solution means aggregation prompt source is a prompt source for aggregating two or more pieces of solution means information into one or more pieces of solution means information.

[0071] Two or more pieces of assignment candidate information are stored in the assignment candidate management unit 112. The two or more pieces of assignment candidate information in the assignment candidate management unit 112 may be included in an assignment prompt source.

[0072] Two or more pieces of solution means candidate information are stored in the solution means candidate management unit 113. The two or more pieces of solution means candidate information in the solution means candidate management unit 113 may be included in a solution means prompt source.

[0073] The reception unit 12 receives various information and instructions. The reception unit 12 receives various information and instructions from, for example, the terminal device 2. The various information and instructions are patent specific information, a start instruction, two or more pieces of problem candidate information, or two or more pieces of solution candidate information, which will be described later. A start instruction is an instruction to start an operation.

[0074] Here, acceptance typically refers to the reception of information transmitted via a wired or wireless communication line, but may also be a concept that includes the acceptance of information input from an input device such as a keyboard, mouse, or touch panel, and the acceptance of information read from a recording medium such as an optical disk, magnetic disk, or semiconductor memory.

[0075] The information receiving unit 121 receives one or more pieces of patent-specific information. It is preferable that the information receiving unit 121 receives two or more pieces of patent-specific information.

[0076] The patent specific information is information that identifies a patent, such as an application number, publication number, patent number, or patent information.

[0077] The processing unit 13 performs various types of processing. The various types of processing are, for example, processing performed by the patent information acquisition unit 130, the problem preprocessing unit 131, the solution means preprocessing unit 132, the judgment preprocessing unit 133, the problem prompt acquisition unit 134, the problem acquisition unit 135, the problem judgment unit 136, the solution means prompt acquisition unit 137, the solution means acquisition unit 138, the solution means judgment unit 139, the other judgment unit 13A, and the map construction unit 13B.

[0078] The patent information acquisition unit 130 acquires patent information corresponding to one or more pieces of patent-specific information received by the information receiving unit 121.

[0079] The patent information acquisition unit 130 acquires, for example, patent information identified by a patent ID (e.g., application number, publication number) that is the patent identifying information accepted by the information accepting unit 121 from a patent database (not shown) or the storage unit 11. The patent information acquisition unit 130 acquires, for example, patent information that is the patent identifying information accepted by the information accepting unit 121. The patent database is usually an external device (not shown).

[0080] The assignment pre-processing unit 131 acquires assignment information for each of one or more pieces of patent-specific information accepted by the information accepting unit 121, using the patent information acquired by the patent information acquiring unit 130.

[0081] The algorithm by which the problem preprocessing unit 131 acquires problem information from patent information does not matter. The problem preprocessing unit 131 does not usually use a generation AI. For example, the problem preprocessing unit 131 acquires a sentence of [Problem to be solved by the invention] in the patent specification, and acquires problem information corresponding to the part of "~" from the sentence using a clue phrase (for example, "I could not do ~"). For example, the problem preprocessing unit 131 acquires a sentence of [Effect of the invention] in the patent specification, and acquires problem information corresponding to the part of "~" (here, information indicating the effect) from the sentence using a clue phrase (for example, "I can do ~"). For example, the problem preprocessing unit 131 acquires problem information corresponding to the part of "~" using a clue phrase "... can do ~" in the specification. In addition, the problem preprocessing unit 131 acquires problem information from patent information using a known method described in, for example, JP Patent Publication No. 2021-43677, Patent No. 7323138, etc. The patent information used by the task pre-processing unit 131 may be the same as the patent information provided to the generation AI, or it may be different.

[0082] The solution means pre-processing unit 132 acquires solution means information for each of one or more pieces of patent-specific information accepted by the information accepting unit 121, using the patent information acquired by the patent information acquiring unit 130.

[0083] The algorithm by which the solution preprocessing unit 132 acquires the solution information from the patent information does not matter. The solution preprocessing unit 132 does not usually use a generation AI. For example, the solution preprocessing unit 132 acquires two or more components from claim 1, searches the specification for sentences with each component as the subject, and acquires the component with the largest number of such sentences as the solution information. The solution preprocessing unit 132 acquires the solution information corresponding to the "···" part by using, for example, the clue phrase "··· can be made by···" in the specification. In addition, the problem preprocessing unit 131 acquires the solution information from the patent information using a known method described in, for example, JP Patent Publication No. 2021-43677, Patent No. 7323138, etc. Note that the patent information used by the solution preprocessing unit 132 may be the same as or different from the patent information given to the generation AI.

[0084] The judgment pre-processing unit 133 judges whether the assignment information acquired by the assignment pre-processing unit 131 satisfies the AI ​​processing condition. The AI ​​processing condition here is a condition for deciding to acquire assignment information from the generating AI without using the assignment information acquired by the assignment pre-processing unit 131. The AI ​​processing condition is, for example, that the assignment pre-processing unit 131 has not been able to acquire assignment information, that the assignment information acquired by the assignment pre-processing unit 131 is information indicating "unclassified", "other", "unknown", etc.

[0085] The judgment preprocessing unit 133 judges whether the solution means information acquired by the solution means preprocessing unit 132 satisfies the AI ​​processing condition. The AI ​​processing condition here is a condition for deciding to acquire the solution means information from the generation AI without using the solution means information acquired by the solution means preprocessing unit 132. The AI ​​processing condition is, for example, that the solution means preprocessing unit 132 has not acquired the solution means information, that the solution means information acquired by the solution means preprocessing unit 132 is information indicating "unclassified", "other", "unknown", etc.

[0086] The assignment prompt acquisition unit 134 acquires an assignment prompt using the assignment prompt source and the patent information. The assignment prompt acquisition unit 134 acquires an assignment prompt, for example, by substituting the patent information into a variable included in the assignment prompt source, where the patent information is to be substituted. An assignment prompt is a prompt for acquiring assignment information. A prompt is information provided to the generation AI to obtain an answer from the generation AI. A prompt may also be called an inquiry, a query, or the like.

[0087] The patent information used by the problem prompt acquisition unit 134 is preferably the "problem to be solved by the invention" or "effect of the invention" in the abstract and specification.

[0088] It is preferable that the assignment prompt acquisition unit 134 acquires a prompt for acquiring one of two or more pieces of assignment candidate information by using the assignment prompt source and the patent information. The two or more pieces of assignment candidate information are, for example, information provided by the user. The two or more pieces of assignment candidate information are, for example, stored in the assignment candidate management unit 112.

[0089] When the task information does not correspond to any of the two or more pieces of task candidate information, the task prompt acquisition unit 134 may acquire a task prompt for acquiring task information indicating "other."

[0090] It is preferable that the issue prompt acquisition unit 134 acquires an issue prompt for each of two or more pieces of patent-specific information by using an issue prompt source and patent information corresponding to each piece of patent-specific information.

[0091] The task acquisition unit 135 passes the task prompt acquired by the task prompt acquisition unit 134 to the generation AI and acquires task information from the generation AI. There is no restriction on the method by which the task acquisition unit 135 passes the task prompt to the generation AI. Passing information to the generation AI usually means sending information to the generation AI device X. The task acquisition unit 135 passes the task prompt to the generation AI, for example, by substituting the task prompt as an argument of the API of the generation AI and executing the API.

[0092] When the assignment determination unit 136 described later determines that the assignment conditions are not satisfied, it is preferable that the assignment acquisition unit 135 again passes the assignment prompt to the generation AI and acquires assignment information from the generation AI.

[0093] The assignment acquisition unit 135 may acquire assignment information indicating "other", which is assignment information that does not correspond to any of two or more pieces of assignment candidate information. The assignment information indicating "other" may be called "other" assignment information or simply "other".

[0094] It is preferable that the assignment acquisition unit 135 passes the assignment prompt acquired by the assignment prompt acquisition unit 134 to the generation AI for each of the two or more pieces of patent-specific information accepted by the information acceptance unit 121, and acquires assignment information from the generation AI.

[0095] When the other judgment unit 13A determines that the retry conditions are met, it is preferable that the task acquisition unit 135 again passes a task prompt to the generation AI for the patent-specific information corresponding to the ``other'' task information and acquires task information from the generation AI.

[0096] When the pre-judgment processing unit 133 judges that the AI ​​processing conditions are satisfied, it is preferable that the task acquisition unit 135 passes the task prompt acquired by the task prompt acquisition unit 134 to the generation AI and acquires task information from the generation AI.

[0097] The assignment determination unit 136 determines whether the assignment information acquired by the assignment acquisition unit 135 satisfies the assignment conditions. The assignment conditions are conditions that the assignment information should satisfy. It is preferable that the assignment conditions are linguistic conditions and not content conditions. The assignment conditions are, for example, that the number of characters in the assignment information is within a threshold or less than a threshold, that the number of words in the assignment information is within a threshold or less than a threshold, and that the assignment information is in a specific language (for example, Japanese).

[0098] The solution prompt acquisition unit 137 acquires a solution prompt using a solution prompt source and patent information. The solution prompt acquisition unit 137 acquires a solution prompt by substituting patent information into a variable included in the solution prompt source, for example, into which the patent information is substituted. The solution prompt is a prompt for acquiring solution information.

[0099] The patent information used by the solution prompt acquisition unit 137 is preferably claim 1 in the claims, the claims, the abstract, and the [Means for solving the problem] in the specification. The patent information used by the solution prompt acquisition unit 137 is preferably different from the patent information used by the problem prompt acquisition unit 134, but may be the same.

[0100] It is preferable that the solution means prompt acquisition unit 137 acquires a prompt for acquiring any one of two or more pieces of solution means candidate information by using the solution means prompt source and the patent information. The two or more pieces of solution means candidate information are, for example, information provided by the user. The two or more pieces of solution means candidate information are, for example, stored in the solution means candidate management unit 113.

[0101] When the solution information does not correspond to any of the two or more candidate solution information, the solution prompt acquisition unit 137 may acquire a solution prompt for acquiring solution information indicating "other."

[0102] It is preferable that the solution prompt acquisition unit 137 acquires a solution prompt for acquiring solution information for each of two or more patent-specific information using a solution prompt source and patent information corresponding to each patent-specific information.

[0103] The solution means acquisition unit 138 passes the solution means prompt acquired by the solution means prompt acquisition unit 137 to the generation AI and acquires solution means information from the generation AI. The solution means acquisition unit 138 transmits information to the solution means prompt X. For example, the solution means acquisition unit 138 passes the solution means prompt to the generation AI by substituting the problem prompt into an argument of the API of the generation AI and executing the API.

[0104] When the solution determination unit 139 determines that the solution condition is not satisfied, it is preferable that the solution acquisition unit 138 passes a solution prompt to the generation AI again and acquires solution information from the generation AI.

[0105] The solution conditions are conditions that the solution information must satisfy. The solution conditions are preferably linguistic conditions and not content conditions. For example, the solution conditions are that the number of characters in the solution information is within or below a threshold, the number of words in the solution information is within or below a threshold, and the solution information is in a specific language (for example, Japanese). It is preferable that the solution conditions are different from the problem conditions, but they may be the same.

[0106] It is preferable that the solution acquisition unit 138 may acquire solution information indicating "others" which is solution information that does not correspond to any of two or more candidate solution information. The solution information indicating "others" may be called "others" solution information or simply "others".

[0107] It is preferable that the solution acquisition unit 138 passes the solution prompt acquired by the solution prompt acquisition unit 137 to the generation AI for each of the two or more pieces of patent-specific information, and acquires the solution information from the generation AI.

[0108] When the pre-judgment processing unit 133 judges that the AI ​​processing conditions are satisfied, it is preferable that the solution acquisition unit 138 passes the solution prompt acquired by the solution prompt acquisition unit 137 to the generation AI and acquires solution information from the generation AI.

[0109] When the other judgment unit 13A determines that the retry condition is met, it is preferable that the solution acquisition unit 138 again passes a solution prompt to the generation AI for the patent specific information corresponding to the ``other'' solution information and acquires the solution information from the generation AI.

[0110] The solution determining unit 139 determines whether the solution information acquired by the solution acquiring unit 138 satisfies the solution condition.

[0111] The other determination unit 13A obtains the number of pieces of "other" assignment information, and determines whether or not the number satisfies the retry condition.

[0112] The retry condition is a condition for obtaining information from the generation AI again. It is preferable that the retry condition is a condition related to the number of "others." The retry condition here is, for example, that the ratio of "other" assignment information is equal to or greater than a threshold, and that the number of "other" assignment information is equal to or greater than a threshold.

[0113] The other determination unit 13A obtains the number of "other" solution means information and determines whether the number satisfies a retry condition. The retry condition here is, for example, that the ratio of "other" solution means information is equal to or greater than a threshold, and that the number of "other" solution means information is equal to or greater than a threshold.

[0114] The map construction unit 13B constructs a patent map having a problem axis having two or more pieces of problem information acquired by the problem acquisition unit 135 and a solution means axis having two or more pieces of solution means information acquired by the solution means acquisition unit 138. Note that the patent map is, for example, a two-dimensional map with one of the problem axis and the solution means axis as the horizontal axis and the other as the vertical axis, but may be a three-dimensional or higher dimensional map.

[0115] It is preferable that each cell of the patent map acquired by the map construction unit 13B is associated with patent specific information. For example, the number of patent specific information mapped to the cell is placed in each cell of the patent map acquired by the map construction unit 13B. Note that the patent specific information corresponding to the cell or the patent specific information mapped to the cell is the patent specific information that identifies the patent information from which the problem information and the solution information corresponding to the cell are acquired.

[0116] The output unit 14 outputs various information, such as one or more pieces of problem information, one or more pieces of solution information, and a patent map.

[0117] Here, output usually means transmission to terminal device 2, but it may also be a concept including display on a display, projection using a projector, printing on a printer, sound output, storage on a recording medium, and handing over the processing results to other processing devices or other programs.

[0118] The assignment output unit 141 outputs one or more pieces of assignment information acquired by the assignment acquisition unit 135. The assignment output unit 141 outputs the assignment information in association with, for example, two or more pieces of patent-specific information or two or more pieces of patent information. It is preferable that the assignment output unit 141 outputs only the assignment information that satisfies the assignment conditions.

[0119] The solution output unit 142 outputs one or more pieces of solution information acquired by the solution acquisition unit 138. The solution output unit 142 outputs the solution information in association with, for example, two or more pieces of patent-specific information or two or more pieces of patent information. It is preferable that the solution output unit 142 outputs only the solution information that satisfies the solution condition.

[0120] The map output unit 143 outputs the patent map constructed by the map construction unit 13 B. There is no restriction on the output form of the patent map.

[0121] Various types of information are stored in the terminal storage unit 21 constituting the terminal device 2. The various types of information are, for example, one or more pieces of patent-specific information.

[0122] The terminal reception unit 22 receives various information, instructions, etc. The various information, instructions, etc. are, for example, a start instruction, or one or more pieces of patent-specific information.

[0123] The means for inputting various information and instructions may be any means, such as a touch panel, a keyboard, a mouse, or a menu screen.

[0124] The terminal processing unit 23 performs various types of processing. For example, the various types of processing are processing for converting received information, instructions, etc. into information or instructions with a structure to be transmitted. For example, the various types of processing are processing for converting received information into information with a structure to be output.

[0125] The terminal transmission unit 24 transmits various information, instructions, etc. to the patent information processing device 1. The various information, instructions, etc. are, for example, a start instruction, or one or more pieces of patent-specific information.

[0126] The terminal receiving unit 25 receives various information from the patent information processing device 1. The various information is, for example, one or more pieces of problem information, one or more pieces of solution means information, or a patent map.

[0127] The terminal output unit 26 outputs various information, such as one or more pieces of problem information, one or more pieces of solution information, or a patent map.

[0128] Here, output may be a concept that includes display on a display, projection using a projector, transmission to an external device, printing on a printer, sound output, storage on a recording medium, and delivery of processing results to other processing devices or other programs, etc.

[0129] The storage unit 11, the prompt source storage unit 111, the problem candidate management unit 112, the solution candidate management unit 113, and the terminal storage unit 21 are preferably non-volatile recording media, but may also be realized as volatile recording media.

[0130] There is no restriction on the process by which information is stored in the storage unit 11, etc. For example, information may be stored in the storage unit 11, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 11, etc., or information inputted via an input device may be stored in the storage unit 11, etc.

[0131] It is preferable that the reception unit 12 and the information reception unit 121 are realized by wireless or wired communication means, but they may also be realized by a means for receiving broadcasts, a device driver for an input means such as a touch panel or keyboard, or control software for a menu screen.

[0132] The processing unit 13, the patent information acquisition unit 130, the problem preprocessing unit 131, the solution means preprocessing unit 132, the judgment preprocessing unit 133, the problem prompt acquisition unit 134, the problem acquisition unit 135, the problem judgment unit 136, the solution means prompt acquisition unit 137, the solution means acquisition unit 138, the solution means judgment unit 139, the other judgment unit 13A, the map construction unit 13B, and the terminal processing unit 23 can usually be realized by a processor, a memory, etc. The processing procedure of the processing unit 13, etc. is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit). The processor is a CPU, an MPU, a GPU, etc., and the type does not matter.

[0133] The output unit 14, the problem output unit 141, the solution output unit 142, and the map output unit 143 are usually realized by wireless or wired communication means, but may be realized by broadcasting means. Also, the output unit 14 etc. may be realized by driver software for an output device such as a display or speaker, or by driver software for an output device and an output device etc.

[0134] The terminal reception unit 22 can be realized by a device driver for an input means such as a touch panel or a keyboard, or control software for a menu screen.

[0135] The terminal transmitting unit 24 is usually realized by a wireless or wired communication means, but may also be realized by a broadcasting means.

[0136] The terminal receiving unit 25 is usually realized by a wireless or wired communication means, but may be realized by a means for receiving a broadcast.

[0137] The terminal output unit 26 may be considered to include, or may not include, an output device such as a display, a speaker, etc. The terminal output unit 26 may be realized by driver software for an output device, or a combination of driver software for an output device and an output device, etc.

[0138] Next, an example of the operation of the patent information processing device 1 will be described with reference to the flowchart of FIG.

[0139] (Step S401) The reception unit 12 judges whether or not a start instruction has been received. If a start instruction has been received, the process proceeds to step S402, and if not, the process returns to step S401.

[0140] (Step S402) The patent information acquisition unit 130 acquires one or more pieces of patent-specific information corresponding to the start instruction received in step S401. The patent information acquisition unit 130 acquires, for example, one or more pieces of patent-specific information from a file included in the start instruction.

[0141] (Step S403) The processing unit 13 assigns 1 to a counter i.

[0142] (Step S404) The processing unit 13 judges whether or not the i-th patent-specific information exists in the patent-specific information acquired in step S402. If the i-th patent-specific information exists, the process proceeds to step S405, and if not, the process proceeds to step S410.

[0143] (Step S405) The processing unit 13 judges whether or not to perform preprocessing. If preprocessing is to be performed, the process proceeds to step S406. If preprocessing is not to be performed, the process proceeds to step S408. Note that preprocessing refers to the process of acquiring task information by the task preprocessing unit 131 and the process of acquiring solution means information by the solution means preprocessing unit 132. Also, whether or not to perform preprocessing is judged, for example, based on information that is determined in advance or included in the start instruction.

[0144] (Step S406) The assignment pre-processing unit 131 etc. perform pre-processing. An example of the pre-processing will be described with reference to the flowchart in FIG.

[0145] (Step S407) The processing unit 13 increments the counter i by 1. The process returns to step S404.

[0146] (Step S408) The task prompt acquisition unit 134 and the like perform task AI processing. An example of the task AI processing will be described with reference to the flowchart in Fig. 6. The task AI processing is processing for acquiring task information using a generation AI.

[0147] (Step S409) The solution prompt acquisition unit 137 etc. performs solution AI processing. Go to step S407. An example of the solution AI processing will be described with reference to the flowchart in Fig. 7. The solution AI processing is a processing for acquiring solution information using a generation AI.

[0148] (Step S410) The other determination unit 13A etc. performs post-assignment processing. An example of the post-assignment processing will be described with reference to the flowchart in Fig. 8. The post-assignment processing is a process for changing the assignment information corresponding to "other" to other assignment information as necessary.

[0149] (Step S411) The other determination unit 13A etc. performs post-solution means processing. An example of the post-solution means processing will be described with reference to the flowchart of Fig. 9. The post-solution means processing is processing for changing the solution means information corresponding to "other" to other solution means information as necessary.

[0150] (Step S412) The processing unit 13 judges whether or not to create a patent map. If a patent map is to be created, the process proceeds to step S413. If not, the process proceeds to step S415. Note that whether or not to create a patent map is determined, for example, by information that is predetermined or included in the start instruction.

[0151] (Step S413) The map construction unit 13B performs a patent map creation process. An example of the patent map creation process will be described with reference to the flowchart of FIG.

[0152] (Step S414) The map output unit 143 outputs the patent map acquired in step S413.

[0153] (Step S415) The output unit 14 outputs the patent list. Return to step S401. The patent list has one or more records for each patent. The record has, for example, patent specification information, problem information, and solution information. The record has, for example, patent information.

[0154] In the flowchart of FIG. 4, the process ends when the power is turned off or an interrupt is issued to end the process.

[0155] Next, an example of the pre-processing in step S406 will be described with reference to the flowchart in FIG.

[0156] (Step S501) The assignment pre-processing unit 131 acquires patent information for which assignment information is to be acquired.

[0157] (Step S502) The task preprocessing unit 131 acquires task cue phrases from the storage unit 11.

[0158] (Step S503) The task preprocessing unit 131 acquires solution means clue phrases from the storage unit 11.

[0159] (Step S504) The assignment preprocessing unit 131 acquires assignment information from the patent information acquired in step S501, using the assignment clue phrase acquired in step S502. This process is a pattern matching process.

[0160] (Step S505) The assignment pre-processing unit 131 judges whether or not the assignment information was acquired in step S504. If the assignment information was acquired, the process proceeds to step S507, and if not, the process proceeds to step S506.

[0161] (Step S506) The assignment pre-processing unit 131 assigns "not acquired" to the assignment information.

[0162] (Step S507) The solution means preprocessing unit 132 acquires solution means information from the patent information acquired in step S501, using the solution means clue phrase acquired in step S503. This process is a pattern matching process.

[0163] (Step S508) The solution pre-processing unit 132 judges whether the solution information was acquired in step S507. If the solution information was acquired, the process returns to the upper process, and if not, the process proceeds to step S509.

[0164] (Step S509) The solution means pre-processing unit 132 assigns "not yet obtained" to the solution means information, and returns to the upper process.

[0165] Next, an example of the task AI processing in step S508 will be described with reference to the flowchart in FIG.

[0166] (Step S601) The issue prompt acquisition unit 134 acquires patent information that corresponds to the target patent-specific information and is used when acquiring issue information.

[0167] (Step S602) The task prompt acquiring unit 134 acquires a task prompt source from the prompt source storage unit 111.

[0168] (Step S603) The assignment prompt acquisition unit 134 creates an assignment prompt by placing the patent information acquired in step S601 in the variable portion of the patent information of the assignment prompt source acquired in step S602.

[0169] (Step S604) The assignment acquisition unit 135 passes the assignment prompt created in step S603 to the generation AI. The assignment acquisition unit 135, for example, assigns the assignment prompt to the API of the generation AI and executes the API.

[0170] (Step S605) The assignment acquisition unit 135 judges whether assignment information has been acquired from the generating AI. If assignment information has been acquired, the process proceeds to step S606, and if not, the process returns to step S605.

[0171] (Step S606) The assignment determination unit 136 determines whether or not the assignment information acquired in step S605 satisfies the assignment conditions. If the assignment conditions are satisfied, the process proceeds to step S608, and if not, the process proceeds to step S607.

[0172] (Step S607) The issue determination unit 136 determines whether or not the termination condition is satisfied. If the termination condition is satisfied, the process proceeds to step S608, and if not, the process returns to step S604. The termination condition is a condition for terminating the acquisition of information using the generation AI for one patent. The termination condition is, for example, that the number of times the generation AI is used again is equal to or greater than a threshold value.

[0173] (Step S608) The issue acquisition unit 135 associates the acquired issue information with the patent-specific information, and returns to the upper-level processing.

[0174] Next, an example of the solving means AI process in step S409 will be described with reference to the flowchart in FIG.

[0175] (Step S701) The solution prompt acquisition unit 137 acquires patent information that corresponds to the target patent specifying information and is used when acquiring the solution information.

[0176] (Step S702) The solution prompt acquiring unit 137 acquires a solution prompt source from the prompt source storage unit 111.

[0177] (Step S703) The solution prompt acquisition section 137 places the patent information acquired in step S701 in the variable portion of the patent information of the solution prompt source acquired in step S702, to create a solution prompt.

[0178] (Step S704) The solution acquisition unit 138 passes the solution prompt created in step S703 to the generation AI. The solution acquisition unit 138, for example, assigns the solution prompt to the API of the generation AI and executes the API.

[0179] (Step S705) The solution acquisition unit 138 judges whether or not solution information has been acquired from the generation AI. If solution information has been acquired, the process proceeds to step S706, and if not, the process returns to step S705.

[0180] (Step S706) The solution determination unit 139 determines whether or not the solution information acquired in step S705 satisfies the solution condition. If it satisfies the solution condition, the process proceeds to step S708, and if it does not, the process proceeds to step S707.

[0181] (Step S707) The solution means determination unit 139 determines whether or not the termination condition is satisfied. If the termination condition is satisfied, the process proceeds to step S708, and if not, the process returns to step S704. The termination condition is, for example, that the number of times the generation AI is used again is equal to or greater than a threshold value.

[0182] (Step S708) The solution obtaining unit 138 associates the obtained solution information with the patent specifying information, and returns to the upper processing.

[0183] Next, an example of the post-task processing in step S410 will be described with reference to the flowchart in FIG.

[0184] (Step S801) The other judgment unit 13A obtains the number of “other” problem information among the problem information corresponding to one or more pieces of patent-specific information.

[0185] (Step S802) The other determination unit 13A determines whether or not the number acquired in step S801 satisfies the retry condition. If the number satisfies the retry condition, the process proceeds to step S803, and if not, the process returns to the upper process.

[0186] (Step S803) The task obtaining unit 135 assigns 1 to a counter i.

[0187] (Step S804) The task acquisition unit 135 judges whether the i-th patent-specific information exists. If the i-th patent-specific information exists, the process proceeds to step S805. If the i-th patent-specific information does not exist, the process returns to step S801. If the i-th patent-specific information does not exist, the process may return to the upper process.

[0188] (Step S805) The problem acquisition unit 135 judges whether the problem information paired with the i-th patent-specific information is "other" problem information. If it is "other" problem information, proceed to step S806, and if it is not "other" problem information, proceed to step S807.

[0189] (Step S806) The task acquisition unit 135 etc. perform task AI processing. An example of the task AI processing has been described with reference to the flowchart of FIG.

[0190] (Step S807) The counter i is incremented by 1. The process returns to step S804.

[0191] Next, an example of the post-processing of the solution means in step S411 will be described with reference to the flowchart in FIG.

[0192] (Step S901) The other determination unit 13A acquires the number of pieces of "other" solution means information among the solution means information corresponding to one or more pieces of patent specifying information.

[0193] (Step S902) The other determination unit 13A determines whether or not the number acquired in step S901 satisfies the retry condition. If the number satisfies the retry condition, the process proceeds to step S903, and if not, the process returns to the upper process.

[0194] (Step S903) The solution obtaining unit 138 assigns 1 to a counter i.

[0195] (Step S904) The solution acquisition unit 138 judges whether the i-th patent-specific information exists. If the i-th patent-specific information exists, the process proceeds to step S905. If the i-th patent-specific information does not exist, the process returns to step S901. If the i-th patent-specific information does not exist, the process may return to the upper process.

[0196] (Step S905) The solution acquisition unit 138 judges whether the solution information paired with the i-th patent-specific information is "other" solution information. If it is "other" solution information, proceed to step S906, and if it is not "other" solution information, proceed to step S907.

[0197] (Step S906) The solution acquisition unit 138 etc. perform solution AI processing. An example of the solution AI processing has been described with reference to the flowchart of FIG.

[0198] (Step S907) The counter i is incremented by 1. The process returns to step S904.

[0199] Next, an example of the patent map creation process in step S413 will be described with reference to the flowchart of FIG.

[0200] (Step S1001) The map construction unit 13B acquires two or more pieces of issue information corresponding to the two or more pieces of target patent-specific information. If the same issue information exists, the map construction unit 13B performs unique processing on the acquired two or more pieces of issue information. The map construction unit 13B may acquire two or more pieces of issue information from the issue candidate management unit 112.

[0201] (Step S1002) The map construction unit 13B acquires two or more pieces of solution means information corresponding to the two or more pieces of target patent-specific information. If the same solution means information exists, the map construction unit 13B performs unique processing on the acquired two or more pieces of solution means information. The map construction unit 13B may acquire two or more pieces of solution means information from the solution means candidate management unit 113.

[0202] (Step S1003) The map construction unit 13B creates a patent map frame in which two or more pieces of problem information obtained in step S1001 are items on one axis, and two or more pieces of solution information obtained in step S1002 are items on the other axis.

[0203] (Step S1004) The map configuration unit 13B assigns 1 to a counter i.

[0204] (Step S1005) The map construction unit 13B judges whether or not the i-th cell exists in the patent map. If the i-th cell exists, the process proceeds to step S1006, and if not, the process returns to the upper process.

[0205] (Step S1006) The map construction unit 13B acquires the problem information and the solution means information corresponding to the i-th cell.

[0206] (Step S1007) The map construction unit 13B acquires 0 or more pieces of patent-specific information that are paired with the problem information and solution information acquired in step S1006, and associates them with the i-th cell.

[0207] (Step S1008) The map construction unit 13B acquires the number of patent-specific information acquired in step S1007, and places the number in the i-th cell. Note that it is preferable that the map construction unit 13B uses the number as an anchor and links the patent-specific information corresponding to the i-th cell to the anchor.

[0208] (Step S1009) The map configuration unit 13B increments the counter i by 1. The process returns to step S1005.

[0209] Three specific operation examples of the information system A in this embodiment are described below. Specific example 1 is a case where preprocessing, which is the acquisition of task information in advance, is not performed, and task candidate information and solution means candidate information are not provided. Specific example 2 is a case where preprocessing is not performed, and task candidate information and solution means candidate information are provided. Specific example 3 is a case where the generation AI is used after preprocessing.

[0210] (Example 1) The prompt source storage unit 111 of the patent information processing device 1 stores the subject prompt source shown in Fig. 11. In the subject prompt source, "$patent information" is a variable, and the acquired patent information is substituted therein.

[0211] Furthermore, the prompt source storage unit 111 stores the solution means prompt source shown in Fig. 12. In the solution means prompt source, "$patent information" is a variable, and the acquired patent information is substituted therein.

[0212] The prompt source storage unit 111 may also store a prompt source (FIG. 13) in which the problem prompt source and the solution means prompt source are combined into one sentence.

[0213] In addition, the problem condition stored in the storage unit 11 is "10 characters or less AND language = Japanese". In addition, the solution means condition stored in the storage unit 11 is "9 characters or less AND language = Japanese".

[0214] In this situation, it is assumed that the user inputs a start command to the terminal device 2, which includes a file containing the application numbers of many patents.

[0215] Next, the terminal device 2 accepts a start instruction including the file and transmits the start instruction to the patent information processing device 1.

[0216] Next, the reception unit 12 of the patent information processing device 1 receives a start instruction. Next, the patent information acquisition unit 130 acquires the application numbers (patent specific information) of many patents in the file included in the start instruction. Next, the processing unit 13 determines not to perform preprocessing.

[0217] Next, for each of the many application numbers, the assignment prompt acquisition unit 134 acquires patent information (here, an abstract) to be used when acquiring assignment information. The assignment prompt acquisition unit 134 also acquires the assignment prompt source of FIG. 11 from the prompt source storage unit 111. Next, the assignment prompt acquisition unit 134 places the acquired abstract in the "$patent information" section of the assignment prompt source for each application number, and creates an assignment prompt. Next, the assignment acquisition unit 135 assigns the created assignment prompt to the API of the generation AI for each application number, and executes the API. As a result, the assignment acquisition unit 135 acquires assignment information from the generation AI for each application number.

[0218] In addition, the task determination unit 136 determines whether the acquired task information for each application number satisfies the task condition "10 characters or less AND language = Japanese", and if the task condition is not satisfied, it performs a process of acquiring task information from the generation AI until the termination condition (here, for example, the number of retries is up to 3 times) is satisfied.

[0219] Furthermore, for each of the many application numbers, the solution prompt acquisition unit 137 acquires patent information (here, claim 1) used when acquiring a solution. Furthermore, the solution prompt acquisition unit 137 acquires the solution prompt source of FIG. 12 from the prompt source storage unit 111. Next, for each application number, the solution prompt acquisition unit 137 places the acquired claim 1 in the "$patent information" portion of the solution prompt source, and creates a solution prompt. Next, for each application number, the solution prompt acquisition unit 137 assigns the created solution prompt to the API of the generation AI, and executes the API. As a result, the solution prompt acquisition unit 137 acquires solution information from the generation AI for each application number.

[0220] In addition, the solution prompt acquisition unit 137 judges whether the acquired solution information satisfies the solution condition "9 characters or less AND language = Japanese" for each application number, and if the solution condition is not satisfied, it performs a process of acquiring solution information from the generation AI until the termination condition (here, for example, the number of retries is up to 4) is satisfied.

[0221] Through the above process, problem information and solution means information are associated with each application number. An example of this is shown in FIG. 14. In FIG. 14, problem information is "problem words" and solution means information is "solution words." Also, 1401 and 1402 in FIG. 14 do not satisfy the problem conditions. Also, 1403 and 1404 in FIG. 14 do not satisfy the solution means conditions. Then, the problem acquisition unit 135 may classify problem information that does not satisfy the problem conditions as "other." Also, the solution means acquisition unit 138 may classify solution means information that does not satisfy the solution means conditions as "other."

[0222] Next, the map construction unit 13B uses the problem information and solution information associated with each application number to construct a patent map with the problem information on the vertical axis and the solution information on the horizontal axis. That is, the map construction unit 13B creates a patent map frame with the acquired problem information "separation and recovery efficiency, porosity improvement, ..." as the vertical axis item and the acquired solution information "two-stage structure, open-cell porosity, ..." as the horizontal axis item.

[0223] Next, the map construction unit 13B obtains the problem information and solution information corresponding to each cell in the frame of the patent map. The map construction unit 13B also obtains 0 or more patent-specific information that pairs with the problem information and solution information for each cell, and associates them with each cell. The map construction unit 13B also obtains the number of patent-specific information for each cell, and places the number in each cell. Furthermore, the map construction unit 13B sets the number as an anchor for each cell, and links the application number and / or patent information corresponding to each cell to the anchor.

[0224] Next, the map output unit 143 transmits the constructed patent map to the terminal device 2. In addition, the output unit 14 transmits to the terminal device 2 a patent list including a large number of records having patent application numbers, problem terms, and solution terms.

[0225] Next, the terminal device 2 receives and outputs the patent map. An example of such a patent map is shown in FIG. 15. The terminal device 2 also receives and outputs a patent list. An example of such a patent list is a table in which 1401 to 1404 in FIG. 14 are marked as "Others."

[0226] In addition, the processing unit 13 of the patent information processing device 1 may use the prompt source of Figure 13 to obtain problem information and solution means information from the generation AI at once for each patent specific information.

[0227] (Example 2) 16 is stored in the prompt source storage unit 111 of the patent information processing device 1. This prompt source is used for acquiring both problem information and solution means information.

[0228] In addition, the problem candidate management unit 112 stores "durability improvement, conversion efficiency improvement, durability, efficiency, stability improvement, photoelectric conversion characteristics, cost reduction, composition, conductivity improvement, texture improvement." In addition, the solution means candidate management unit 113 stores "means A, means B, . . ., means G." In addition, the variable "$category" of the prompt source (FIG. 16) is substituted with 10 problem candidate information of the problem candidate management unit 112, and a problem prompt is created. In addition, the variable "$category" of the prompt source (FIG. 16) is substituted with 7 problem candidate information of the solution means candidate management unit 113, and a solution means prompt is created. In addition, the prompt source (FIG. 16) is a case where a problem prompt for acquiring problem information indicating "other" is acquired when the problem does not correspond to any of the problem information of two or more problem candidate information. In addition, the prompt source (FIG. 16) is a case where a solution means prompt for acquiring solution information indicating "other" is acquired when the problem does not correspond to any of the problem information of two or more problem candidate information.

[0229] Furthermore, the retry condition stored in the storage unit 11 is that the ratio of "other" is 20% or more.

[0230] In this situation, it is assumed that the user inputs a start command to the terminal device 2, which includes a file containing the application numbers of many patents.

[0231] Next, the terminal device 2 accepts a start instruction including the file and transmits the start instruction to the patent information processing device 1.

[0232] Next, the reception unit 12 of the patent information processing device 1 receives a start instruction. Next, the patent information acquisition unit 130 acquires the application numbers (patent specific information) of many patents in the file included in the start instruction. Next, the processing unit 13 determines not to perform preprocessing.

[0233] Next, for each of the many application numbers, the issue prompt acquisition unit 134 acquires patent information (here, the sentence [Problem that the invention is trying to solve] in the specification) used when acquiring issue information. Also, the issue prompt acquisition unit 134 acquires the prompt source of FIG. 16 from the prompt source storage unit 111. Next, the issue prompt acquisition unit 134 acquires the issue candidate information "Improved durability, Improved conversion efficiency, . . . , Improved texture" from the issue candidate management unit 112, and places it in the variable "$classification". Next, the issue prompt acquisition unit 134 places the acquired patent information in the "$patent information" of the prompt source for each application number, and creates an issue prompt. Next, the issue acquisition unit 135 assigns the created issue prompt to the API of the generation AI for each application number, and executes the API. As a result, the issue acquisition unit 135 acquires issue information from the generation AI for each application number.

[0234] Also, for each of the many application numbers, the solution means prompt acquisition unit 137 acquires patent information (here, the sentence [Means for solving the problem] in the specification) used when acquiring the solution means. Also, the solution means prompt acquisition unit 137 acquires the prompt source of FIG. 16 from the prompt source storage unit 111. Next, the solution means prompt acquisition unit 137 acquires the solution means candidate information "Means A, Means B, . . ., Means G" from the solution means candidate management unit 113 and places it in the variable "$classification". Next, the solution means prompt acquisition unit 137 places the acquired patent information in the "$patent information" of the solution means prompt source for each application number, and creates a solution means prompt. Next, the solution means prompt acquisition unit 137 assigns the created solution means prompt to the API of the generation AI for each application number, and executes the API. As a result, the solution means prompt acquisition unit 137 acquires solution means information from the generation AI for each application number.

[0235] Next, the other judgment unit 13A obtains the number of "Other" in the assignment information corresponding to each application number. The other judgment unit 13A also obtains the number of application numbers (for example, "256"). Next, the other judgment unit 13A uses the obtained number to calculate the percentage of "Other" (number of "Other" / number of application numbers x 100) and determines whether the retry condition of "20% or less" is met. Here, the other judgment unit 13A determines that the percentage of "Other" is 25%, which satisfies the retry condition.

[0236] Next, when the assignment information corresponding to each application number is "other", the assignment acquisition unit 135 acquires assignment information from the generation AI by using the assignment prompt corresponding to the application number again. Then, the task acquisition unit 135 performs a process of acquiring assignment information from the generation AI for application numbers whose assignment information is "other" until the retry condition "20% or less" is satisfied.

[0237] Next, the other judgment unit 13A obtains the number of "Other" in the solution means information corresponding to each application number. The other judgment unit 13A also obtains the number of application numbers (for example, "256"). Next, the other judgment unit 13A uses the obtained number to calculate the percentage of "Other" (number of "Other" / number of application numbers x 100) and determines whether the retry condition of "20% or less" is met. Here, the other judgment unit 13A determines that the percentage of "Other" is 16%, and therefore does not satisfy the retry condition.

[0238] Next, the map construction unit 13B constructs a patent map by the same process as that in the specific example 1 (the process of the flowchart in FIG. 10).

[0239] Next, the map output unit 143 transmits the constructed patent map to the terminal device 2. In addition, the output unit 14 transmits to the terminal device 2 a patent list including a large number of records having patent application numbers, problem terms, and solution terms.

[0240] Next, the terminal device 2 receives and outputs the patent map. The terminal device 2 also receives and outputs the patent list. An example of such a patent list is shown in FIG.

[0241] (Example 3) The prompt source storage unit 111 of the patent information processing device 1 stores the problem prompt source and the solution means prompt source shown in Fig. 11. The problem condition stored in the storage unit 11 is "10 characters or less AND language = Japanese". The solution means condition stored in the storage unit 11 is "9 characters or less AND language = Japanese".

[0242] In this situation, it is assumed that the user inputs a start command to the terminal device 2, which includes a file containing the application numbers of many patents.

[0243] Next, the terminal device 2 accepts a start instruction including the file and transmits the start instruction to the patent information processing device 1.

[0244] Next, the reception unit 12 of the patent information processing device 1 receives the start instruction. Next, the patent information acquisition unit 130 acquires the application numbers (patent specific information) of many patents in the file included in the start instruction. Next, the processing unit 13 determines to perform preprocessing.

[0245] Next, the problem pre-processing unit 131 acquires problem information for each patent, for example, from the specification, by the process described using the flowchart in Figure 5. It is assumed that the problem pre-processing unit 131 acquires "other" if the problem information cannot be acquired from the specification. The solution means pre-processing unit 132 acquires solution means information for each patent, for example, from the specification, by the process described using the flowchart in Figure 5. It is assumed that the solution means pre-processing unit 132 acquires "other" if the solution means information cannot be acquired from the specification. It is assumed that the patent list acquired as a result of the above process is as shown in Figure 17.

[0246] Next, the problem acquisition unit 135 etc. performs the problem AI processing of Fig. 6 for one or more patents (e.g., 1701 in Fig. 17) whose problem information was "other" as a result of preprocessing, and acquires the problem information using the generation AI. Then, the problem acquisition unit 135 changes the problem information corresponding to each patent from "other" to the problem information acquired from the generation AI.

[0247] 7 for one or more patents (e.g., 1702 in FIG. 17) whose solution information was "other" as a result of preprocessing, and acquires the solution information using the generation AI. Then, the solution acquisition unit 138 changes the solution information corresponding to each patent from "other" to the solution information acquired from the generation AI.

[0248] As a result, for example, assume that "Other" 1701 in FIG. 17 is changed to "Improved electrical conductivity" and 1702 is changed to "Measure D."

[0249] Next, the output unit 14 transmits to the terminal device 2 a patent list including a large number of records each having a patent application number, a problem term, and a solution term.

[0250] Next, the terminal device 2 receives and outputs the patent list.

[0251] As described above, according to this embodiment, by appropriately using the generation AI, important information, namely issue information, can be obtained from patent information.

[0252] Furthermore, according to this embodiment, by appropriately using the generation AI, important solution information can be obtained from patent information.

[0253] Furthermore, according to this embodiment, by appropriately using the generation AI, appropriate problem information for the patent information can be obtained from two or more pieces of problem candidate information.

[0254] Furthermore, according to this embodiment, by appropriately using the generation AI, appropriate solution information for the patent information can be obtained from two or more candidate solution information.

[0255] Furthermore, according to this embodiment, by appropriately using the generation AI, the number of patent information whose subject information is "other" can be reduced.

[0256] The processing in this embodiment may be realized by software. This software may be distributed by software download or the like. This software may be recorded on a recording medium such as a CD-ROM and distributed. This also applies to other embodiments in this specification. The software for realizing the patent information processing device 1 in this embodiment is the following program. That is, this program is a program for making a computer that can access a prompt source storage unit in which a prompt source from which a prompt for acquiring problem information related to a patent is acquired be stored, an information receiving unit that receives patent specific information that identifies the patent from which problem information is acquired, a patent information acquisition unit that acquires patent information related to the content of the patent corresponding to the patent specific information, a problem prompt acquisition unit that acquires a problem prompt for acquiring problem information using the problem prompt source and the patent information, a problem acquisition unit that passes the problem prompt acquired by the problem prompt acquisition unit to a generation AI and acquires problem information from the generation AI, and a problem output unit that outputs the problem information acquired by the problem acquisition unit.

[0257] (Embodiment 2) The present embodiment differs from the first embodiment in that when the number of pieces of problem information or the number of pieces of solution means information is large, the problem information or the solution means information is aggregated.

[0258] Except for the reference numerals, the conceptual diagram of an information system B in this embodiment is the same as that in Fig. 1. The information system B includes a patent information processing device 3 and one or more terminal devices 2.

[0259] The patent information processing device 3 is a device that acquires and outputs patent information issues, etc. The patent information processing device 3 is usually a server, but may be a standalone device. When the patent information processing device 3 is a server, the patent information processing device 3 is, for example, a so-called cloud server or an ASP server, but it does not matter.

[0260] 18 is a block diagram of the patent information processing device 3 in this embodiment. The patent information processing device 3 includes a storage unit 31, a reception unit 12, a processing unit 33, and an output unit 14. The storage unit 31 includes a prompt source storage unit 111 and a related word dictionary 311. The processing unit 33 includes a patent information acquisition unit 130, a problem preprocessing unit 131, a solution means preprocessing unit 132, a judgment preprocessing unit 133, a problem prompt acquisition unit 134, a problem acquisition unit 135, a problem judgment unit 136, a solution means prompt acquisition unit 137, a solution means acquisition unit 138, a solution means judgment unit 139, a problem aggregation unit 331, a solution means aggregation unit 332, and a map construction unit 13B.

[0261] Various types of information are stored in the storage unit 31 constituting the patent information processing device 3. The various types of information are, for example, various prompt sources and various conditions. The various prompt sources are, for example, a problem summarizing prompt source, which will be described later, and a solution means summarizing prompt source, which will be described later. The various conditions are, for example, a problem summarizing condition and a solution means summarizing condition.

[0262] The assignment aggregation condition is a condition for aggregating assignment information. The assignment aggregation condition is a condition for reducing the number of assignment information. The number of assignment information is the number that is counted when the character strings of the assignment information are different. The assignment aggregation condition is, for example, that the number of assignment information is equal to or greater than a threshold value.

[0263] The solution aggregation condition is a condition for aggregating solution information. The solution aggregation condition is a condition for reducing the number of solution information. The number of solution information is the number that is counted when the character strings of the solution information are different. The solution aggregation condition is, for example, that the number of solution information is equal to or greater than a threshold value.

[0264] The prompt source storage unit 111 stores, for example, a problem summarizing prompt source and a solution means summarizing prompt source.

[0265] The task aggregation prompt source is information that is the source of creating a prompt used to aggregate task information. For example, the task aggregation prompt source is "Please tell me the hypernym common to "$First task information" and "$Second task information".

[0266] The solution aggregation prompt source is information that is the source of creating a prompt used to aggregate solution information. For example, the problem aggregation prompt source is "Please tell me the hypernym common to "$first solution information" and "$second solution information".

[0267] The related term dictionary 311 stores one or more related term groups. A related term group has two or more related terms. For example, a related term group may be two or more synonyms, two or more homonyms, or one or more hypernyms and one or more hyponyms.

[0268] The processing unit 33 performs various types of processing. The various types of processing are, for example, processing performed by the patent information acquisition unit 130, the problem preprocessing unit 131, the solution means preprocessing unit 132, the judgment preprocessing unit 133, the problem prompt acquisition unit 134, the problem acquisition unit 135, the problem judgment unit 136, the solution means prompt acquisition unit 137, the solution means acquisition unit 138, the solution means judgment unit 139, the problem aggregation unit 331, the solution means aggregation unit 332, or the map construction unit 13B.

[0269] The assignment aggregating unit 331 acquires one or more pieces of assignment information by using two or more different pieces of assignment information acquired by the assignment acquiring unit 135. This process is assignment information aggregating process.

[0270] It is preferable that the assignment aggregating unit 331 performs an assignment information aggregation process when two or more pieces of assignment information acquired by the assignment acquisition unit 135 match the assignment aggregation conditions.

[0271] The task aggregating unit 331 may, for example, perform task information aggregation processing by passing a task aggregation prompt to the generation AI. The task aggregating unit 331, for example, substitutes two or more pieces of task information to be aggregated into the original variable portion of the task aggregating prompt to create a task aggregating prompt. Next, the task aggregating unit 331, for example, passes the task aggregating prompt to the generation AI and obtains one piece of task information from the generation AI.

[0272] The task aggregating unit 331 may perform a process of aggregating task information by using, for example, the related word dictionary 311. For example, the task aggregating unit 331 obtains a superordinate word common to two or more pieces of task information from the related word dictionary 311, and aggregates the two or more pieces of task information into one superordinate word.

[0273] The solution aggregation unit 332 acquires one or more pieces of solution information by using two or more different pieces of solution information acquired by the solution acquisition unit 138. This process is solution information aggregation processing.

[0274] It is preferable that the solution means aggregation unit 332 performs an aggregation process of the solution means information when two or more pieces of solution means information acquired by the solution means acquisition unit 138 match the solution means aggregation condition.

[0275] The solution aggregation unit 332 may perform the aggregation process of the solution information by, for example, passing the solution aggregation prompt to the generation AI. The solution aggregation unit 332, for example, substitutes two or more pieces of solution information to be aggregated into the original variable part of the solution aggregation prompt to create a solution aggregation prompt. Next, the solution aggregation unit 332, for example, passes the solution aggregation prompt to the generation AI and obtains one piece of solution information from the generation AI.

[0276] The solution aggregating unit 332 may perform aggregating processing of solution information by using, for example, the related word dictionary 311. For example, the solution aggregating unit 332 obtains a hypernym common to two or more pieces of solution information from the related word dictionary 311, and aggregates the two or more pieces of solution information into one hypernym.

[0277] The storage unit 31 and the related word dictionary 311 are preferably non-volatile recording media, but may also be realized as volatile recording media.

[0278] There is no restriction on the process by which information is stored in the storage unit 31, etc. For example, information may be stored in the storage unit 31, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 31, etc., or information inputted via an input device may be stored in the storage unit 31, etc.

[0279] The processing unit 33, the problem aggregating unit 331, and the solution means aggregating unit 332 can usually be realized by a processor, a memory, etc. The processing procedure of the processing unit 33, etc. is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit). The processor may be a CPU, MPU, GPU, etc., and the type is not important.

[0280] Next, an example of the operation of the patent information processing device 3 will be described with reference to the flowchart of Fig. 19. In the flowchart of Fig. 19, the description of the same steps as those in the flowchart of Fig. 4 will be omitted.

[0281] (Step S1901) The task aggregating unit 331 performs a task aggregating process. An example of the task aggregating process will be described with reference to the flowchart in FIG.

[0282] (Step S1902) The solution means aggregating unit 332 performs a solution means aggregating process. An example of the solution means aggregating process will be described with reference to the flowchart in FIG.

[0283] In the flowchart of FIG. 19, the process ends when the power is turned off or an interrupt occurs to end the process.

[0284] Next, an example of the task aggregation process in step S1901 will be described with reference to the flowchart in FIG.

[0285] (Step S2001) The issue aggregating unit 331 acquires issue information corresponding to each of two or more pieces of patent-specific information.

[0286] (Step S2002) The assignment aggregating unit 331 performs unique processing on the two or more pieces of assignment information acquired in step S2001.

[0287] (Step S2003) The assignment aggregating unit 331 acquires the number of pieces of assignment information as a result of the unique process in step S2002.

[0288] (Step S2004) The task aggregating unit 331 judges whether or not the number acquired in step S2003 satisfies the task aggregating condition. If it satisfies the task aggregating condition, the process proceeds to step S2005, and if it does not, the process returns to the upper process.

[0289] (Step S2005) The issue summarizing unit 331 assigns 1 to a counter i.

[0290] (Step S2006) The assignment aggregation unit 331 judges whether or not the i-th assignment information pair (two pieces of assignment information) exists among the two or more pieces of assignment information acquired. If the i-th assignment information pair exists, the process proceeds to step S2007, and if not, the process returns to step S2003.

[0291] (Step S 2007 ) The task summarizing unit 331 acquires a task summarizing prompt source from the prompt source storage unit 111 .

[0292] (Step S2008) The task summarizing unit 331 places the i-th set of task information in the variable portion of the task summarizing prompt source, and creates a task summarizing prompt.

[0293] (Step S2009) The task summarizing unit 331 passes the task summarizing prompt to the generation AI.

[0294] (Step S2010) The task aggregation unit 331 acquires information from the generation AI and uses the information to determine whether the two pieces of task information can be aggregated. If they can be aggregated, the process proceeds to step S2011. If they cannot be aggregated, the process proceeds to step S2012.

[0295] (Step S2011) The assignment aggregating unit 331 acquires one piece of assignment information that is the aggregation result of the i-th set of assignment information, and associates it with each of the two pieces of assignment information that constitute the i-th set of assignment information.

[0296] (Step S2012) The issue summarizing unit 331 increments the counter i by 1. The process returns to step S2006.

[0297] Next, an example of the solution means aggregation process in step S1902 will be described with reference to the flowchart in FIG.

[0298] (Step S2101) The solution means aggregating unit 332 acquires solution means information corresponding to each of two or more pieces of patent specifying information.

[0299] (Step S2102) The solution means aggregating unit 332 performs unique processing on the two or more pieces of solution means information acquired in step S2101.

[0300] (Step S2103) The solution aggregation unit 332 acquires the number of solution information pieces resulting from the unique process in step S2102.

[0301] (Step S2104) The solution means aggregation unit 332 judges whether or not the number acquired in step S2103 satisfies the solution means aggregation condition. If it satisfies the solution means aggregation condition, the process proceeds to step S2105, and if it does not, the process returns to the upper process.

[0302] (Step S2105) The solution means aggregating unit 332 assigns 1 to a counter i.

[0303] (Step S2106) The solution aggregation unit 332 judges whether or not the i-th set of solution information (two pieces of solution information) exists among the two or more pieces of solution information acquired. If the i-th set of solution information exists, the process proceeds to step S2107, and if not, the process returns to step S2103.

[0304] (Step S2107) The solution aggregation unit 332 acquires a solution aggregation prompt source from the prompt source storage unit 111.

[0305] (Step S2108) The solution means aggregating unit 332 places the i-th set of solution means information in the variable part of the solution means aggregating prompt source, and creates a solution means aggregating prompt.

[0306] (Step S2109) The solution summarizing unit 332 passes the solution summarizing prompt to the generation AI.

[0307] (Step S2110) The solution aggregation unit 332 acquires information from the generation AI, and uses the information to determine whether the two solution information can be aggregated. If it can be aggregated, the process proceeds to step S2111, and if it cannot be aggregated, the process proceeds to step S2112.

[0308] (Step S2111) The solution means aggregating unit 332 acquires one solution means information that is an aggregation result of the i-th solution means information set, and associates it with each of two solution means information constituting the i-th solution means information set.

[0309] (Step S2112) The solution means aggregating unit 332 increments the counter i by 1. The process returns to step S2106.

[0310] As described above, according to this embodiment, by appropriately using the generation AI, important information, namely issue information, can be obtained from patent information.

[0311] Furthermore, according to this embodiment, by appropriately using the generation AI, important solution information can be obtained from patent information.

[0312] In addition, according to this embodiment, by appropriately using the generation AI, more appropriate problem information can be obtained from patent information.

[0313] Furthermore, according to this embodiment, by appropriately using the generation AI, more appropriate solution information can be obtained from patent information.

[0314] Furthermore, according to this embodiment, by appropriately using the generation AI, appropriate problem information for the patent information can be obtained from two or more pieces of problem candidate information.

[0315] Furthermore, according to this embodiment, by appropriately using the generation AI, appropriate problem information for patent information can be obtained.

[0316] In addition, according to this embodiment, the generation AI can be appropriately used to reduce the number of patent information whose subject information is "other."

[0317] Furthermore, according to this embodiment, by appropriately using the generation AI, it is possible to obtain an appropriate number of types of problem information for two or more patent information without the need to prepare problem candidate information.

[0318] Furthermore, according to this embodiment, by appropriately using the generation AI, appropriate solution information for the patent information can be obtained from two or more candidate solution information.

[0319] Also, Fig. 22 shows the appearance of a computer that executes the program described in this specification to realize the patent information processing device 1 of the various embodiments described above. The above-mentioned embodiments can be realized by computer hardware and a computer program executed thereon. Fig. 22 is an overview of this computer system 300, and Fig. 23 is a block diagram of the system 300.

[0320] In FIG. 22, a computer system 300 includes a computer 301 including a CD-ROM drive, a keyboard 302, a mouse 303, and a monitor 304.

[0321] 23, computer 301 includes, in addition to CD-ROM drive 3012, MPU 3013, bus 3014 connected to CD-ROM drive 3012 etc., ROM 3015 for storing programs such as a boot-up program, RAM 3016 connected to MPU 3013 for temporarily storing instructions of application programs and providing temporary storage space, and hard disk 3017 for storing application programs, system programs, and data. Although not shown here, computer 301 may further include a network card for providing connection to a LAN.

[0322] A program for causing the computer system 300 to execute the functions of the patent information processing device 1 of the above-mentioned embodiment may be stored on a CD-ROM 3101, inserted into the CD-ROM drive 3012, and further transferred to the hard disk 3017. Alternatively, the program may be transmitted to the computer 301 via a network (not shown) and stored on the hard disk 3017. The program is loaded into the RAM 3016 when executed. The program may be loaded directly from the CD-ROM 3101 or the network.

[0323] The program does not necessarily include an operating system (OS) or a third party program that causes the computer 301 to execute the functions of the patent information processing device 1 of the above-mentioned embodiment. The program only needs to include an instruction portion that calls appropriate functions (modules) in a controlled manner to obtain a desired result. How the computer system 300 operates is well known, and a detailed description will be omitted.

[0324] In addition, in the above program, the steps of transmitting information and receiving information do not include processing performed by hardware, such as processing performed by a modem or interface card in the transmitting step (processing that is performed only by hardware).

[0325] The program may be executed by a single computer or a plurality of computers. That is, the program may be executed by a centralized processing or a distributed processing.

[0326] Furthermore, in each of the above embodiments, it goes without saying that two or more communication means present in one device may be physically realized by one medium.

[0327] In each of the above embodiments, each process may be realized by centralized processing in a single device, or may be realized by distributed processing in a plurality of devices.

[0328] The present invention is not limited to the above-described embodiment, and various modifications are possible, and it goes without saying that these modifications are also included within the scope of the present invention. [Industrial Applicability]

[0329] As described above, the patent information processing device 1 of the present invention has the effect of being able to obtain important information from patent information by appropriately using generation AI, and is useful as a server that automatically creates patent maps. [Explanation of symbols]

[0330] A, B Information Systems 1, 3 Patent information processing device 2 Terminal Equipment 11, 31 Storage area 12 Reception 13, 33 Processing section 14 Output section 21 Terminal storage section 22 Terminal Reception 23 Terminal Processing Section 24 Terminal transmitter 25 Terminal receiving section 26 Terminal Output Section 111 Prompt Source Storage Unit 112 Project Candidate Management Department 113 Solution candidate management department 121 Information Reception Department 130 Patent Information Acquisition Department 131 Pre-processing section 132 Solution Preprocessing Unit 133 Pre-judgment processing unit 134 Task prompt acquisition section 135 Assignment acquisition department 136 Issue Judgment Department 137 Solution prompt acquisition unit 138 Solution acquisition part 139 Solution Judgment Department 13A Other judgment section 13B Map composition section 141 Task Output Section 142 Solution output unit 311 Related Words Dictionary 331 Issue Collection Department 332 Solution Aggregation Department

Claims

1. A computer that can access a prompt source storage unit that stores a problem prompt source from which a prompt for acquiring problem information related to a problem of a patent is obtained, an information receiving unit that receives patent identification information that identifies the patent for which problem information is to be obtained; a patent information acquisition unit that acquires patent information relating to the content of a patent corresponding to the patent identification information; an assignment prompt acquisition unit that acquires an assignment prompt for acquiring assignment information by using the assignment prompt source and the patent information; a task acquisition unit that transfers the task prompt acquired by the task prompt acquisition unit to a generation AI and acquires task information from the generation AI; A program for causing the assignment acquisition unit to function as an assignment output unit that outputs the assignment information acquired.

2. The prompt source storage unit includes: A solution prompt source from which a prompt for obtaining solution information related to a solution for solving the problem of the patent is obtained is stored; a solution prompt acquisition unit that acquires a solution prompt for acquiring solution information by using the solution prompt source and the patent information; a solution acquisition unit that transfers the solution prompt acquired by the solution prompt acquisition unit to a generation AI and acquires solution information from the generation AI; The program according to claim 1 , further causing the computer to function as a solution output unit that outputs the solution information acquired by the solution acquisition unit.

3. The computer, The task information acquired by the task acquisition unit further functions as a task determination unit that determines whether or not the task information satisfies a task condition that is a condition that the task information should satisfy; The task acquisition unit If the task determination unit determines that the task condition is not satisfied, the task determination unit again passes the task prompt to the generation AI and obtains task information from the generation AI. The task output unit:

2. The program according to claim 1, for causing the computer to function as outputting the assignment information that satisfies the assignment conditions.

4. The computer, a solution determination unit that determines whether the solution information acquired by the solution acquisition unit satisfies a solution condition that the solution information should satisfy; The solution acquisition unit, When the solution determination unit determines that the solution condition is not satisfied, the solution prompt is again passed to the generation AI, and solution information is obtained from the generation AI; The solution output unit is 3. The program according to claim 2, for causing the computer to function as outputting the solution information that satisfies the solution condition.

5. The source of the task prompt is: This is information that is used to obtain a prompt for obtaining any one of two or more pieces of candidate assignment information that are candidates for assignment information, The task prompt acquisition unit A program as described in claim 1, which causes the computer to function as obtaining a prompt for obtaining any one of the two or more candidate problem information using the problem prompt source and the patent information.

6. The task prompt acquisition unit If the assignment information does not correspond to any of the two or more assignment candidate information that are candidates for the assignment information, an assignment prompt is obtained to obtain assignment information indicating "others"; The task acquisition unit 6. A program as claimed in claim 5, for causing the computer to function as if there were a case where "other" assignment information was acquired, which is assignment information that does not correspond to any of the two or more pieces of assignment candidate information.

7. The information receiving unit is Accept two or more patent specific information, The task prompt acquisition unit For each of the two or more patent-specific information, obtain a problem prompt for obtaining problem information using the problem prompt source and the patent information; The task acquisition unit For each of the two or more pieces of patent specific information, the problem prompt acquired by the problem prompt acquisition unit is passed to a generation AI, and problem information is acquired from the generation AI. The computer The task acquisition unit further functions as a task aggregation unit that acquires one or more pieces of task information using two or more different pieces of task information acquired by the task acquisition unit; The task output unit:

2. The program according to claim 1, for causing the computer to function as outputting the one or more pieces of assignment information acquired by the assignment aggregation unit, rather than the assignment information acquired by the assignment acquisition unit.

8. The solution prompt source is: The information is used to obtain a prompt for obtaining any one of two or more pieces of candidate solution information that are candidates for solution information, The solution means prompt acquisition unit 3. A program as claimed in claim 2, for causing the computer to function as a program for obtaining a prompt for obtaining any one of the two or more candidate solution means information by using the solution means prompt source and the patent information.

9. The solution means prompt acquisition unit When the solution information does not correspond to any of the two or more candidate solution information, a solution prompt is obtained for obtaining solution information indicating "others". The solution acquisition unit, 9. The program according to claim 8, for causing the computer to function as if there were a case where "other" solution information, which is solution information that does not correspond to any of the two or more candidate solution information, was acquired.

10. The information receiving unit is Accept two or more patent specific information, The solution means prompt acquisition unit For each of the two or more patent-specific information, a solution prompt for acquiring the solution information is obtained using the solution prompt source and the patent information; The solution acquisition unit, For each of the two or more pieces of patent specific information, the solution prompt acquired by the solution prompt acquisition unit is passed to a generation AI, and the computer is caused to function as an entity that acquires solution information from the generation AI; The computer, The number of pieces of “other” solution information is acquired, and the number of pieces of “other” solution information is further made to function as an other determination unit that determines whether or not the number satisfies a retry condition; The task acquisition unit The program described in claim 9, for causing the computer to function as follows: when the other judgment unit determines that the retry condition is satisfied, the solution prompt is again passed to the generation AI for the patent specific information corresponding to the ``other'' solution information, and the solution information is obtained from the generation AI.

11. The information receiving unit is Accept two or more patent specific information, The solution means prompt acquisition unit For each of the two or more patent-specific information, a solution prompt for acquiring the solution information is obtained using the solution prompt source and the patent information; The solution acquisition unit, For each of the two or more pieces of patent specific information, the solution prompt acquired by the solution prompt acquisition unit is passed to a generation AI, and the computer is caused to function as an entity that acquires solution information from the generation AI; The computer, a solution aggregation unit that acquires one or more pieces of solution information by using two or more different pieces of solution information acquired by the solution acquisition unit; The solution output unit is 3. The program according to claim 2, for causing the computer to function as outputting the one or more pieces of solution information acquired by the solution aggregation unit, rather than the solution information acquired by the solution acquisition unit.

12. The computer, a problem pre-processing unit that acquires problem information using patent information related to the content of the patent corresponding to the patent identification information; The task pre-processing unit further functions as a judgment pre-processing unit that judges whether the task information acquired by the task pre-processing unit satisfies an AI processing condition for acquiring task information from the generation AI without using the task information; The task acquisition unit A program as described in claim 1, for causing the computer to function as follows: when the pre-judgment processing unit determines that the AI ​​processing condition is satisfied, the task prompt acquired by the task prompt acquisition unit is passed to a generation AI, and task information is acquired from the generation AI.

13. The computer, a solution pre-processing unit that acquires solution information using patent information related to the content of the patent corresponding to the patent specifying information; The solution information acquired by the solution pre-processing unit is further made to function as a judgment pre-processing unit that judges whether or not the solution information acquired by the solution pre-processing unit satisfies an AI processing condition for acquiring the solution information from the generated AI without using the solution information; The solution acquisition unit, A program as described in claim 2, for causing the computer to function as follows: when the pre-judgment processing unit determines that the AI ​​processing condition is satisfied, the solution prompt acquired by the solution prompt acquisition unit is passed to the generation AI, and solution information is acquired from the generation AI.

14. The information receiving unit is Accept two or more patent specific information, The task prompt acquisition unit For each of the two or more patent-specific information, obtain a problem prompt for obtaining problem information using the problem prompt source and the patent information; The solution means prompt acquisition unit For each of the two or more patent-specific information, a solution prompt for acquiring the solution information is obtained using the solution prompt source and the patent information; The task acquisition unit For each of the two or more pieces of patent specific information, the problem prompt acquired by the problem prompt acquisition unit is passed to a generation AI, and problem information is acquired from the generation AI; The solution acquisition unit, For each of the two or more pieces of patent specific information, the solution prompt acquired by the solution prompt acquisition unit is passed to a generation AI, and the computer is caused to function as an entity that acquires solution information from the generation AI; The computer, a map construction unit that constructs a patent map having a problem axis having two or more pieces of problem information acquired by the problem acquisition unit and a solution axis having two or more pieces of solution information acquired by the solution acquisition unit; 3. The program according to claim 2, further functioning as a map output unit that outputs the patent map constructed by the map construction unit.

15. a prompt source storage unit for storing a prompt source from which a prompt for acquiring problem information related to a patent problem is acquired; an information receiving unit that receives patent identification information that identifies the patent for which problem information is to be obtained; a patent information acquisition unit that acquires patent information relating to the content of a patent corresponding to the patent identification information; an assignment prompt acquisition unit that acquires an assignment prompt for acquiring assignment information by using the assignment prompt source and the patent information; a task acquisition unit that transfers the task prompt acquired by the task prompt acquisition unit to a generation AI and acquires task information from the generation AI; A patent information processing device comprising: a task output unit that outputs the task information acquired by the task acquisition unit.

16. A patent information processing method realized by a prompt source storage unit in which a problem prompt source from which a prompt for acquiring problem information related to a patent is acquired is stored, an information receiving unit, a patent information acquisition unit, a problem prompt acquisition unit, a problem acquisition unit, and a problem output unit, an information receiving step in which the information receiving unit receives patent identification information that identifies a patent for which problem information is to be obtained; a patent information acquisition step in which the patent information acquisition unit acquires patent information relating to the content of the patent corresponding to the patent identifying information; an assignment prompt acquisition step in which the assignment prompt acquisition unit acquires an assignment prompt for acquiring assignment information by using the assignment prompt source and the patent information; a task acquisition step in which the task acquisition unit passes the task prompt acquired by the task prompt acquisition unit to a generation AI and acquires task information from the generation AI; The patent information processing method includes a task output step in which the task output unit outputs the task information acquired by the task acquisition unit.

Images