Information processing device, information processing method, and information processing program

Abstract

This technology enhances user satisfaction with the results of assessments regarding the possibility of acquiring intellectual property rights. [Solution] The information processing device includes: a related technology acquisition unit that acquires related technical information related to the intellectual property to be analyzed based on information about the intellectual property to be analyzed described in natural language; a positive opinion generation unit that generates a positive opinion regarding the possibility of obtaining rights to the intellectual property to be analyzed using a first large-scale language model; a negative opinion generation unit that generates a negative opinion regarding the possibility of obtaining rights to the intellectual property to be analyzed using a second large-scale language model; and a conclusion generation unit that generates a conclusion regarding the possibility of obtaining rights to the intellectual property to be analyzed using a third large-scale language model.

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus, an information processing method, and an information processing program.

Background Art

[0002] Patent Document 1 discloses a technique for determining the possibility of obtaining rights regarding information related to intellectual property input by a user. In this technique, as a determination result, a rank indicating the possibility of obtaining the rights is presented to the user. Further, when the intellectual property is an invention, as a determination result, the degree of coincidence with similar documents for each constituent element of the invention is presented to the user.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the technique described in Patent Document 1, there may be cases where the user's acceptance of the determination result cannot be sufficiently obtained only by the rank of the possibility of obtaining rights, the degree of coincidence with similar documents, etc. as described above. Therefore, it is required to enhance the user's acceptance of the determination result.

[0005] The present disclosure has been made in view of the above problems, and an exemplary object thereof is to provide a technique for enhancing the user's acceptance of the determination result of the possibility of obtaining rights related to intellectual property.

Means for Solving the Problems

[0006] An information processing device relating to an exemplary aspect of this disclosure includes: related technology acquisition means for acquiring related technology information indicating related technologies related to intellectual property under analysis based on information under analysis in which the intellectual property under analysis is described in natural language; affirmative opinion generation means for generating an affirmative opinion regarding the possibility of obtaining rights to the intellectual property under analysis based on the information under analysis and the related technology information using a first large-scale language model; negative opinion generation means for generating a negative opinion regarding the possibility of obtaining rights based on the information under analysis and the related technology information using a second large-scale language model; and conclusion generation means for generating a conclusion regarding the possibility of obtaining rights based on the affirmative opinion and the negative opinion using a third large-scale language model.

[0007] An information processing method relating to an exemplary aspect of this disclosure includes: a related technology acquisition process in which at least one processor acquires related technology information indicating related technologies related to the intellectual property under analysis based on information under analysis in which the intellectual property under analysis is described in natural language; an affirmative opinion generation process in which the at least one processor generates an affirmative opinion regarding the possibility of obtaining rights to the intellectual property under analysis based on the information under analysis and the related technology information using a first large-scale language model; a negative opinion generation process in which the at least one processor generates a negative opinion regarding the possibility of obtaining rights based on the information under analysis and the related technology information using a second large-scale language model; and a conclusion generation process in which the at least one processor generates a conclusion regarding the possibility of obtaining rights based on the affirmative opinion and the negative opinion using a third large-scale language model.

[0008] An illustrative aspect of the present disclosure relates to an information processing program that causes at least one processor to function as an information processing device, and includes: related technology acquisition means for acquiring related technology information indicating related technologies related to intellectual property to be analyzed based on information to be analyzed in natural language sentences about the intellectual property to be analyzed; affirmative opinion generation means for generating an affirmative opinion regarding the possibility of obtaining rights to the intellectual property to be analyzed based on the information to be analyzed and the related technology information using a first large-scale language model; negative opinion generation means for generating a negative opinion regarding the possibility of obtaining rights based on the information to be analyzed and the related technology information using a second large-scale language model; and conclusion generation means for generating a conclusion regarding the possibility of obtaining rights based on the affirmative opinion and the negative opinion using a third large-scale language model. [Effects of the Invention]

[0009] One illustrative aspect of this disclosure is that it can provide a technology that enhances user satisfaction with the results of the determination of the possibility of acquiring intellectual property rights. [Brief explanation of the drawing]

[0010] [Figure 1] This is a block diagram showing the configuration of the information processing device related to this disclosure. [Figure 2] This is a flowchart showing the flow of the information processing method related to this disclosure. [Figure 3] This diagram schematically shows an overview of the information processing system related to this disclosure. [Figure 4] This is a block diagram showing the configuration of the information processing system related to this disclosure. [Figure 5] This is a flowchart showing the flow of the information processing method related to this disclosure. [Figure 6] This diagram schematically shows an example of the input screen for the analysis target related to this disclosure. [Figure 7] This diagram schematically shows an example of the keyword generation screen related to this disclosure. [Figure 8]This figure schematically shows an example of a prior art screen related to this disclosure. [Figure 9] This diagram schematically shows an example of a differential screen related to this disclosure. [Figure 10] This diagram schematically shows an example of both opinion screens related to this disclosure. [Figure 11] This diagram schematically shows another example of the two opinion screens related to this disclosure. [Figure 12] This diagram schematically shows an example of the conclusion screen related to this disclosure. [Figure 13] This diagram schematically shows an example of the final draft screen related to this disclosure. [Figure 14] This block diagram shows the hardware configuration of the computer that functions as each device related to this disclosure. [Modes for carrying out the invention]

[0011] The following are examples of embodiments of the present invention. However, the present invention is not limited to the exemplary embodiments shown below, and various modifications are possible within the scope of the claims. For example, embodiments obtained by appropriately combining some or all of the technologies (things or methods) employed in each of the exemplary embodiments shown below may also be included in the scope of the present invention. Furthermore, embodiments obtained by appropriately omitting some of the technologies employed in each of the exemplary embodiments shown below may also be included in the scope of the present invention. In addition, the effects mentioned in each of the exemplary embodiments shown below are examples of effects that can be expected in that exemplary embodiment and do not define the scope of the present invention. That is, embodiments that do not produce the effects mentioned in each of the exemplary embodiments shown below may also be included in the scope of the present invention.

[0012] [First Exemplary Embodiment] A first exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. This exemplary embodiment is a basic form for each of the exemplary embodiments described later. Note that the scope of application of each technology adopted in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technology adopted in this exemplary embodiment can be adopted in other exemplary embodiments included in the present disclosure as long as there are no particular technical obstacles. In addition, each technology shown in the drawings referred to for explaining this exemplary embodiment can also be adopted in other exemplary embodiments included in the present disclosure as long as there are no particular technical obstacles.

[0013] (Configuration of Information Processing Apparatus 1) The configuration of the information processing apparatus 1 will be described with reference to FIG. 1. FIG. 1 is a block diagram showing the configuration of the information processing apparatus 1. As shown in FIG. 1, the information processing apparatus 1 includes a related technology acquisition unit 11, an affirmative opinion generation unit 12, a negative opinion generation unit 13, and a conclusion generation unit 14. The related technology acquisition unit 11 is an example of a configuration that realizes related technology acquisition means. The affirmative opinion generation unit 12 is an example of a configuration that realizes affirmative opinion generation means. The negative opinion generation unit 13 is an example of a configuration that realizes negative opinion generation means. The conclusion generation unit 14 is an example of a configuration that realizes conclusion generation means.

[0014] The related technology acquisition unit 11 acquires related technology information indicating related technologies related to the intellectual property to be analyzed based on analysis target information in which the intellectual property to be analyzed is described in a natural language sentence. Here, the intellectual property to be analyzed is intellectual property that can be described in a natural language sentence. For example, the intellectual property to be analyzed may be an invention, a utility model, or a thesis, but is not limited thereto. In addition, the intellectual property to be analyzed may be intellectual property in any state such as under consideration, before filing, after filing, before examination, after examination, etc. Further, for example, the analysis target information may include information indicating a scope of rights (for example, claims), an outline, and part or all of a detailed description. The intellectual property to be analyzed may be input, for example, by a user's operation or may be input by being read from an arbitrary storage medium.

[0015] In addition, the related technology acquisition unit 11 may select related acquisition information based on the analysis target information from among a plurality of candidates for related technology information. Further, the related technology acquisition unit 11 may acquire related technology information designated by the user according to the analysis target information.

[0016] The positive opinion generation unit 12 generates a positive opinion regarding the patent acquisition possibility of the analysis target based on the analysis target information and the related technology information using a first large language model. For example, the patent acquisition possibility of the analysis target may include the possibility that one or both of novelty and inventiveness are recognized. Also, for example, the patent acquisition possibility may include the possibility of satisfying other requirements in addition to novelty and / or inventiveness. For example, the positive opinion is a natural language sentence indicating an opinion on the premise that the intellectual property of the analysis target has novelty and / or inventiveness with respect to the related technology. For example, the positive opinion may include a natural language sentence indicating that it has novelty and / or inventiveness and the basis therefor.

[0017] For example, when the analysis target information and the related technology information are input, the first large language model outputs a positive opinion regarding the patent acquisition possibility. Note that the information input to the first large language model includes at least the analysis target information and the related technology information, and other information may or may not be included.

[0018] For example, the first large language model may be a general large language model fine-tuned using positive case information. The positive case information includes, for example, a case of the analysis target information, a case of the related technology information, and a case of a positive opinion regarding the patent acquisition possibility indicated by the case of the analysis target information. Such positive case information may include information obtained regarding other intellectual properties whose patent acquisition possibility has been actually affirmed, or may include information generated for training.

[0019] Furthermore, for example, the first large-scale language model does not necessarily have to be fine-tuned and may be a general-purpose large-scale language model. In this case, for example, positive opinions may be output by in-context learning in which the first large-scale language model is input not only the information to be analyzed and related technical information but also the aforementioned positive case information.

[0020] The negative opinion generation unit 13 uses a second large-scale language model to generate a negative opinion regarding the possibility of obtaining rights to the intellectual property under analysis, based on the information under analysis and related technical information. Specific examples of the possibility of obtaining rights to the intellectual property under analysis are as described above. For example, the negative opinion is a natural language sentence that expresses an opinion based on the premise that the intellectual property under analysis lacks novelty and / or inventive step with respect to related technology. For example, the negative opinion may include a natural language sentence that states that it lacks novelty and / or inventive step, and the basis for this statement.

[0021] For example, the second large-scale language model, upon receiving the information to be analyzed and related technical information, outputs a negative opinion regarding the possibility of obtaining the rights. The information input to the second large-scale language model includes at least the information to be analyzed and related technical information, and may or may not include other information.

[0022] For example, the second large-scale language model may be a general-purpose large-scale language model that has been fine-tuned using negative case information. Negative case information includes, for example, cases of the information under analysis, cases of related technical information, and cases of negative opinions regarding the patentability of the intellectual property indicated by the case of the information under analysis. Such negative case information may include information obtained regarding other intellectual property whose patentability has actually been denied, or it may include information generated for training purposes.

[0023] Furthermore, for example, the second large-scale language model does not necessarily have to be fine-tuned and may be a general-purpose large-scale language model. In that case, the second large-scale language model may output negative opinions through in-context learning, in which the aforementioned negative case information is input in addition to the information to be analyzed and related technical information.

[0024] The conclusion generation unit 14 uses a third large-scale language model to generate a conclusion regarding the possibility of obtaining rights to the intellectual property under analysis, based on the affirmative and negative opinions. For example, the conclusion is a natural language sentence indicating which of the affirmative and negative opinions regarding the novelty and / or inventiveness of the intellectual property under analysis with respect to the related technology is valid. For example, the conclusion may include either an affirmative or negative opinion, and a natural language sentence indicating the reason for it.

[0025] For example, the third large-scale language model outputs a conclusion regarding the possibility of obtaining rights when it receives the information to be analyzed, related technical information, affirmative opinions, and negative opinions as input. The information input to the third large-scale language model includes at least affirmative and negative opinions, and may or may not include other information.

[0026] For example, the third large-scale language model may be a general-purpose large-scale language model that has been fine-tuned using case information for conclusions. The case information for conclusions may include, for example, business information relating to the business related to the intellectual property under analysis, and / or a patent portfolio related to the intellectual property under analysis. This allows for the generation of conclusions that take into account the business information and / or the patent portfolio. The case information for conclusions may also include, for example, cases of information under analysis, cases of related technical information, cases of affirmative opinions, cases of negative opinions, and cases of conclusions. Such case information for conclusions may be generated to include affirmative and negative opinions generated by the affirmative opinion generation unit 12 and the negative opinion generation unit 13 with respect to other intellectual property whose patentability has actually been affirmed or denied. Furthermore, such case information for conclusions may also be information generated for training purposes.

[0027] Furthermore, for example, the third large-scale language model does not necessarily have to be fine-tuned and may be a general-purpose large-scale language model. In that case, the conclusion may be output by in-context learning, in which the third large-scale language model is inputted with the information to be analyzed, related technical information, positive and negative opinions, as well as the aforementioned case information for conclusions.

[0028] Furthermore, if at least two of the first, second, and third large-scale language models are fine-tuned models, then these at least two models are different from each other. Also, for example, if at least two of the first, second, and third large-scale language models are general-purpose large-scale language models, then these at least two models may be the same or different.

[0029] (Effects of Information Processing Device 1) As described above, the information processing device 1 employs a configuration that includes: a related technology acquisition unit 11 that acquires related technology information indicating related technologies related to the intellectual property to be analyzed based on the information to be analyzed, which is described in natural language; an affirmative opinion generation unit 12 that generates an affirmative opinion regarding the possibility of acquiring rights to the intellectual property to be analyzed based on the information to be analyzed and the related technology information using a first large-scale language model; a negative opinion generation unit 13 that generates a negative opinion regarding the possibility of acquiring rights based on the information to be analyzed and the related technology information using a second large-scale language model; and a conclusion generation unit 14 that generates a conclusion regarding the possibility of acquiring rights based on the affirmative and negative opinions using a third large-scale language model. Therefore, with the information processing device 1, a conclusion is generated based on both affirmative and negative opinions regarding the possibility of acquiring rights to the intellectual property, which has the effect of increasing the user's satisfaction with the conclusion regarding the possibility of acquiring rights.

[0030] (Information processing method S1 flow) The flow of the information processing method S1 will be explained with reference to Figure 2. For example, if the information processing device 1 is equipped with at least one processor, the information processing device 1 executes the information processing method S1. Figure 2 is a flowchart showing the flow of the information processing method S1. As shown in Figure 2, the information processing method S1 includes a related technology acquisition process S11, a positive opinion generation process S12, a negative opinion generation process S13, and a conclusion generation process S14.

[0031] In the related technology acquisition process S11, at least one processor (for example, the related technology acquisition unit 11) acquires related technology information indicating related technologies related to the intellectual property to be analyzed, based on the analysis target information in which the intellectual property to be analyzed is described in natural language. Details of the related technology acquisition process S11 will be described in the same way as the details of the related technology acquisition unit 11 described above.

[0032] In the affirmative opinion generation process S12, at least one processor (e.g., the affirmative opinion generation unit 12) uses a first large-scale language model to generate an affirmative opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the information under analysis and related technical information. Details of the affirmative opinion generation process S12 will be described in the same way as the details of the affirmative opinion generation unit 12 described above.

[0033] In the negative opinion generation process S13, at least one processor (e.g., a negative opinion generation unit 13) uses a second large-scale language model to generate a negative opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the information under analysis and related technical information. Details of the negative opinion generation process S13 will be described in the same way as the details of the negative opinion generation unit 13 described above.

[0034] Furthermore, the positive opinion generation process S12 and the negative opinion generation process S13 are not limited to being executed in the order described above; they may also be executed in the reverse order, or some or all of the processes may be executed in parallel.

[0035] In conclusion generation process S14, at least one processor uses a third large-scale language model to generate a conclusion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the affirmative and negative opinions. Details of conclusion generation process S14 will be described in the same way as details of conclusion generation unit 14 described above.

[0036] (Effects of information processing method S1) As described above, the information processing method S1 employs a configuration that includes: a related technology acquisition process S11 in which at least one processor acquires related technology information indicating related technologies related to the intellectual property to be analyzed based on the information to be analyzed, which is described in natural language; a positive opinion generation process S12 in which at least one processor generates a positive opinion regarding the possibility of obtaining rights to the intellectual property to be analyzed, based on the information to be analyzed and the related technology information, using a first large-scale language model; a negative opinion generation process S13 in which at least one processor generates a negative opinion regarding the possibility of obtaining rights, based on the information to be analyzed and the related technology information, using a second large-scale language model; and a conclusion generation process S14 in which at least one processor generates a conclusion regarding the possibility of obtaining rights, based on the positive and negative opinions, using a third large-scale language model. Therefore, the same effects as the information processing device 1 can be obtained with the information processing method S1.

[0037] [Second exemplary embodiment] A second exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. Components having the same function as those described in the above-described exemplary embodiment are denoted by the same reference numerals, and their descriptions are omitted as appropriate. The scope of application of each technology adopted in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technology adopted in this exemplary embodiment can also be adopted in other exemplary embodiments included in this disclosure, to the extent that no particular technical problems arise. Furthermore, each technology shown in the drawings referenced to describe this exemplary embodiment can also be adopted in other exemplary embodiments included in this disclosure, to the extent that no particular technical problems arise.

[0038] (Overview of Information Processing System 100A) The information processing system 100A presents affirmative, negative, and conclusions regarding the novelty and inventive step (example of patentability) of the target invention (example of intellectual property) represented by the claim to be analyzed, based on the claim to be analyzed (example of information to be analyzed) and prior art documents (example of related technical information). Furthermore, if the conclusion does not meet predetermined conditions, the information processing device 1A generates an improvement proposal for the claim (example of improvement proposal information), and repeats the operation using this improvement proposal as a new claim to be analyzed, thereby presenting the user with an improvement proposal whose conclusion meets the predetermined conditions as the final proposal.

[0039] Figure 3 is a schematic diagram illustrating the overview of the information processing system 100A. As shown in Figure 3, in the information processing system 100A, keywords are generated from the claims to be analyzed using the large-scale language model LLM1, or summaries are generated using the large-scale language model LLM2. Note that both keywords and summaries may be generated, but the following explanation will focus on the configuration in which only one of them is generated. Furthermore, classifications such as IPC (International Patent Classification) are identified from the claims to be analyzed using the large-scale language model LLM3. Next, multiple candidates for prior art documents are obtained from database 3, which will be described later, using the keywords or summaries and the classifications. Next, the large-scale language model LLM4 is used to identify prior art documents to be compared with the claims to be analyzed from the multiple candidates. Next, the differences and commonalities between the claims to be analyzed and the prior art documents are analyzed using the large-scale language model LLM5. Finally, the large-scale language model LLM6 is used to generate an affirmative opinion regarding the novelty and inventive step of the subject invention indicated by the claims to be analyzed. Furthermore, the large-scale language model LLM7 is used to generate negative opinions regarding the novelty and inventive step of the subject invention. Next, the large-scale language model LLM8 is used to generate a conclusion on whether the positive or negative opinion is valid. If the conclusion does not meet the predetermined conditions (for example, the conclusion is not positive), the large-scale language model LLM9 is used to generate proposed improvements to the claim. Then, the above series of processes is repeated with the proposed improvements to the claim as the new claim to be analyzed.

[0040] (Configuration of Information Processing System 100A) The configuration of the information processing system 100A will be described with reference to Figure 4. Figure 4 is a block diagram showing the configuration of the information processing system 100A. As shown in Figure 4, the information processing system 100A includes an information processing device 1A, a large-scale language model storage device 2, a database 3, an input device 4, and a display device 5. The information processing device 1A is communicated with the large-scale language model storage device 2, the database 3, the input device 4, and the display device 5 via a network or peripheral device connection interface. Some or all of the information stored in the large-scale language model storage device 2 and the database 3 may also be stored in the storage unit 120 of the information processing device 1A. In addition, one or both of the input device 4 and the display device 5 may be built into the information processing device 1A instead of being connected to it. Furthermore, the input device 4 and the display device 5 may be connected to or built into a user terminal (not shown), and the user terminal may be communicated with the information processing device 1A via a network. Although Figure 4 shows one large-scale language model memory device 2, one database 3, one input device 4, and one display device 5, the information processing system 100A may include multiple instances of some or all of these devices.

[0041] (Large-scale language model memory 2) Large-scale language model memory device 2 stores the large-scale language models LLM1 to LLM9. Each of the large-scale language models LLM1 to LLM9 is a deep learning model generated to perform a natural language processing task. For example, each of the large-scale language models LLM1 to LLM9 is a model that performs a text generation task, taking a natural language sentence as input and outputting a generated natural language sentence. Each of the large-scale language models LLM1 to LLM9 may be a fine-tuned version of a general-purpose large-scale language model, or it may be a general-purpose large-scale language model. If at least one of the large-scale language models LLM1 to LLM9 is a general-purpose large-scale language model, in-context learning may be performed using that large-scale language model. Also, if at least two of the large-scale language models LLM1 to LLM9 are general-purpose large-scale language models, these two may be the same model or different models.

[0042] Large-scale language model LLM1 is an example of a fifth large-scale language model used for keyword generation. For example, when a claim to be analyzed is input to large-scale language model LLM1, it outputs keywords related to the claim to be analyzed. For example, large-scale language model LLM1 may be a general-purpose large-scale language model that has been fine-tuned for the technical field assumed to be covered by the claim to be analyzed. Alternatively, large-scale language model LLM1 may be a general-purpose large-scale language model. In that case, in addition to the claim to be analyzed, knowledge related to the field represented by the claim to be analyzed may be input to large-scale language model LLM1 to output keywords.

[0043] The large-scale language model LLM2 is an example of a large-scale language model used to generate summaries of information to be analyzed. For example, when the claims to be analyzed are input, the large-scale language model LLM2 outputs a summary of those claims. For example, the large-scale language model LLM2 may be fine-tuned for the technical field assumed to be covered by the claims to be analyzed. Alternatively, the large-scale language model LLM2 may be a general-purpose large-scale language model. In that case, the large-scale language model LLM2 may output a summary when, in addition to the claims to be analyzed, knowledge related to the field represented by the claims to be analyzed is input.

[0044] The large-scale language model LLM3 is an example of a seventh large-scale language model used to identify the classification of intellectual property indicated by the information being analyzed. For example, if the object of analysis is an invention, the aforementioned IPC is an example of "intellectual property classification," but it is not limited to this. For example, the large-scale language model LLM3 outputs a classification when the claims to be analyzed are input. For example, the large-scale language model LLM3 may be a general-purpose large-scale language model that has been fine-tuned using case information for classification identification. The case information for classification identification includes examples of claims to be analyzed and examples of classifications. The case information for classification identification may be generated, for example, based on published patent documents. Alternatively, the large-scale language model LLM3 may be a general-purpose large-scale language model. In that case, the large-scale language model LLM3 may output a classification by in-context learning, in which the above-mentioned case information for classification identification is input in addition to the claims to be analyzed.

[0045] The large-scale language model LLM4 is an example of a sixth large-scale language model used to select relevant technical information from multiple candidates. For example, when the large-scale language model LLM4 is given a claim to be analyzed and multiple patent documents as input, it outputs patent documents that are relevant to the claim to be analyzed as prior art documents. For example, the large-scale language model LLM4 may be a general-purpose large-scale language model that has been fine-tuned for the technical field assumed to be the claim to be analyzed. Alternatively, the large-scale language model LLM4 may be a general-purpose large-scale language model. In that case, in addition to the claim to be analyzed and multiple patent documents, knowledge related to the field indicated by the claim to be analyzed may be input to the large-scale language model LLM4, and the most relevant prior art documents may be output.

[0046] The large-scale language model LLM5 is used to generate differences and commonalities between the information to be analyzed and related technical information. For example, when the claims and prior art documents to be analyzed are input to the large-scale language model LLM5, it outputs the differences and commonalities between the claims and prior art documents to be analyzed. For example, the large-scale language model LLM5 may be a general-purpose large-scale language model that has been fine-tuned using case information that includes differences and commonalities. The case information that includes differences and commonalities includes examples of the claims to be analyzed, examples of the prior art documents, and examples of differences and commonalities between the two examples. The case information that includes differences and commonalities may be generated, for example, based on the history of published patent documents, or it may include information generated for training. Alternatively, the large-scale language model LLM5 may be a general-purpose large-scale language model. In that case, the large-scale language model LLM5 may output differences and commonalities through in-context learning, in which the above-mentioned case information that includes differences and commonalities is input in addition to the claims and prior art documents to be analyzed.

[0047] Large-scale language model LLM6 is an example of a first large-scale language model used to generate affirmative opinions. For example, when differences and similarities between the claims under analysis and prior art documents are input, large-scale language model LLM6 outputs an affirmative opinion regarding the novelty and inventive step of the subject invention represented by the claims under analysis. The affirmative opinion includes the opinion that the invention is novel and inventive, and the basis for that opinion. For example, large-scale language model LLM6 may be a general-purpose large-scale language model that has been fine-tuned using positive case information. Positive case information includes examples of differences and similarities between the claims under analysis and the prior art documents, and examples of affirmative opinions regarding the novelty and inventive step of the subject invention represented by the claims under analysis. Positive case information may be generated, for example, based on the history of published patent documents, or it may include information generated for training. Large-scale language model LLM6 may also be a general-purpose large-scale language model. In that case, the large-scale language model LLM6 may output a positive opinion through in-context learning, in which the aforementioned positive case information is input in addition to the differences and commonalities between the claims and prior art documents under analysis.

[0048] The large-scale language model LLM7 is an example of a second large-scale language model used to generate negative opinions. For example, when the differences and similarities between the claims under analysis and the prior art documents are input, the large-scale language model LLM7 outputs negative opinions regarding the novelty and inventive step of the invention represented by the claims under analysis. The negative opinions may include opinions that the invention lacks novelty and inventive step, along with the reasons for this opinion. Alternatively, the negative opinions may include opinions that the invention is novel but lacks inventive step, along with the reasons for this opinion. For example, the large-scale language model LLM7 may be a general-purpose large-scale language model that has been fine-tuned using negative case information. The negative case information includes examples of differences and similarities between the claims under analysis and the prior art documents, as well as examples of negative opinions regarding the novelty and inventive step of the claims under analysis. The negative case information may be generated, for example, based on the history of published patent documents, or it may include information generated for training. The large-scale language model LLM7 may also be a general-purpose large-scale language model. In that case, the large-scale language model LLM7 may output negative opinions through in-context learning, in which the aforementioned negative case information is input in addition to the differences and similarities between the claims and prior art documents under analysis.

[0049] The large-scale language model LLM8 is an example of a third large-scale language model used to generate conclusions. For example, the large-scale language model LLM8 outputs conclusions when given claims, prior art documents, affirmative and negative opinions to be analyzed as input. The conclusions include which of the affirmative and negative opinions is valid and the reasoning behind it. For example, the large-scale language model LLM8 may be a general-purpose large-scale language model that has been fine-tuned using case information for conclusions. The case information for conclusions may include, as mentioned above, business information relating to the business related to the intellectual property to be analyzed, and / or a patent portfolio related to the intellectual property to be analyzed. This allows for the generation of conclusions that take into account the business information and / or the patent portfolio. The case information for conclusions includes examples of claims to be analyzed, examples of prior art documents, examples of affirmative opinions, examples of negative opinions, and examples of conclusions. The case information for conclusions may be generated, for example, based on the history of published patent documents, or it may include information generated for training. Furthermore, for example, the examples of affirmative and negative opinions in the case information for conclusions may be generated using large-scale language models LLM6 and LLM7, with the examples of claims and prior art documents under analysis as input. Alternatively, large-scale language model LLM8 may be a general-purpose large-scale language model. In that case, the conclusion may be output by in-context learning, with the above-mentioned case information for conclusions, in addition to the claims, prior art documents, affirmative and negative opinions under analysis, being input to large-scale language model LLM8.

[0050] The large-scale language model LLM9 is an example of a fourth large-scale language model used to generate improvement proposal information. For example, when the large-scale language model LLM9 is given the claims to be analyzed, prior art documents, differences and similarities, affirmative opinions, negative opinions, and conclusions as input, it outputs improvement proposals for the claims to be analyzed. For example, the large-scale language model LLM9 may be a general-purpose large-scale language model that has been fine-tuned using improvement case information. Improvement case information may include, for example, business information relating to the business related to the intellectual property to be analyzed, a patent portfolio related to the intellectual property to be analyzed, and / or strategic information relating to said patent portfolio. This allows important components of the business information, patent portfolio, and / or strategic information to be included in the improvement proposal information. Furthermore, the improvement case information includes examples of claims to be analyzed, examples of prior art documents, examples of differences and similarities, examples of affirmative opinions, examples of negative opinions, examples of conclusions, and examples of improvement proposals. Improvement case information may be generated, for example, based on the history information of published patent documents. For example, examples of improvement proposals may be generated based on amendments to the claims in the progress information. Furthermore, the improvement example information may include information generated for training. Also, the large-scale language model LLM9 may be a general-purpose large-scale language model. In that case, improvement proposals may be output by in-context learning, where the large-scale language model LLM9 is input with the claims to be analyzed, prior art documents, differences and similarities, positive opinions, negative opinions, and conclusions, in addition to the aforementioned improvement example information.

[0051] (Database 3) Database 3 stores the search targets for related technical information. For example, Database 3 may store multiple patent documents as search targets. Alternatively, Database 3 may store each of the multiple patent documents in a manner that allows searching based on the similarity of features. For example, each patent document may be associated with a vector representation that shows the features of that patent document. The vector representation may be information obtained by converting at least a part of each patent document (e.g., claims and abstract) into a vector format using an embedding model. In this case, each patent document is indexed in this vector format. The conversion to vector format and indexing processes may be performed in advance by the related technical information acquisition unit 11, or by an external device of the information processing device 1A. Note that the search targets for related technical information stored in Database 3 may include not only patent documents but also non-patent documents.

[0052] (Input device 4 and display device 5) The input device 4 is configured to receive input to the information processing device 1A, and may include, for example, an input device such as a keyboard, mouse, touch panel, camera, or microphone. The display device 5 is configured to display the screen output from the information processing device 1A, and may include, for example, a display. The input device 4 and the display device 5 may also be integrally formed as a touch panel or the like.

[0053] (Configuration of Information Processing Device 1A) As shown in Figure 4, the information processing device 1A includes a control unit 110 and a storage unit 120. The control unit 110 controls all parts of the information processing device 1A. The storage unit 120 stores various data and programs that the control unit 110 references.

[0054] The control unit 110 includes, in addition to the related technology acquisition unit 11, the positive opinion generation unit 12, the negative opinion generation unit 13, and the conclusion generation unit 14 provided by the information processing device 1, an improvement unit 15, a dual opinion presentation unit 16, a final proposal presentation unit 17, an analysis target acquisition unit 18, and a difference generation unit 19. The improvement unit 15 is an example of a configuration that realizes the improvement means. The dual opinion presentation unit 16 is an example of a configuration that realizes the dual opinion presentation means. The final proposal presentation unit 17 is an example of a configuration that realizes the final proposal presentation means.

[0055] The analysis target acquisition unit 18 acquires the claims to be analyzed (an example of information to be analyzed). The claims to be analyzed may be acquired, for example, based on user operations using the input device 4.

[0056] The related technology acquisition unit 11 is configured in the same manner as in Exemplary Embodiment 1, and is configured as follows: The related technology acquisition unit 11 may generate keywords or summaries from the claims to be analyzed (an example of information to be analyzed) using the large-scale language model LLM1 or LLM2, and acquire prior art documents (an example of related technical information) from the database 3 using the generated keywords or summaries. Alternatively, the related technology acquisition unit 11 may acquire multiple candidates for prior art documents (an example of related technical information) and select one of the multiple candidates as the prior art document (an example of related technical information) using the large-scale language model LLM4. Alternatively, the related technology acquisition unit 11 may identify the classification of the invention to be analyzed (an example of intellectual property to be analyzed) using the large-scale language model LLM3, and acquire prior art documents (an example of related technical information) using the identified classification.

[0057] For example, the related technology acquisition unit 11 includes a keyword generation unit 111, a summary generation unit 112, a classification identification unit 113, a candidate acquisition unit 114, and a related technology selection unit 115.

[0058] The keyword generation unit 111 generates keywords from the claims to be analyzed using the large-scale language model LLM1. Details of the large-scale language model LLM1 are as described above. This makes it possible to retrieve multiple candidate prior art documents from the database 3 based on their similarity to the keywords. Alternatively, the keyword generation unit 111 may retrieve keywords entered by the user instead of generating keywords from the claims to be analyzed using the large-scale language model LLM1, or in addition to doing so.

[0059] The summary generation unit 112 generates a summary from the claims to be analyzed using the large-scale language model LLM2. Details of the large-scale language model LLM2 are as described above. This makes it possible to retrieve multiple candidate prior art documents from database 3 based on their similarity to the summary.

[0060] Furthermore, to obtain multiple candidate prior art documents, the user may be able to choose whether to use keywords or abstracts. If the user has not selected whether to use keywords or abstracts, one of a predetermined set (e.g., keywords) may be used, and the other (e.g., abstracts) may be optional and selectable based on user input. Alternatively, the choice between keywords and abstracts may be automatically determined based on predetermined conditions, without user input. For example, if the input claim to be analyzed is longer than a threshold, an abstract may be generated; otherwise, keywords may be generated.

[0061] The classification identification unit 113 uses the large-scale language model LLM3 to identify the classification of the claim to be analyzed. Details of the large-scale language model LLM3 are as described above. This makes it possible to narrow down multiple candidate prior art documents based on their classification.

[0062] The candidate acquisition unit 114 uses the generated keywords or abstracts and the identified classifications to acquire multiple candidate prior art documents from the database 3. For example, the candidate acquisition unit 114 generates vector representations that show the characteristics of the generated keywords or abstracts using an embedding model. The candidate acquisition unit 114 also identifies multiple patent documents stored in the database 3 whose similarity between the vector representation of the keywords or abstracts and the vector representation of the patent documents is above a threshold. The candidate acquisition unit 114 then acquires those patent documents that match the identified classifications as multiple candidate prior art documents. This makes it possible to acquire more appropriate patent documents from the database 3 as multiple candidate prior art documents compared to simply using keywords or abstracts. If the candidate acquisition unit 114 acquires only one candidate (for example, one patent document with a similarity to the keywords or abstract above a threshold, or one patent document that matches the identified classification, etc.), the processing by the related technology selection unit 115, which will be described next, can be omitted.

[0063] The related technology selection unit 115 uses the large-scale language model LLM4 to select one of several candidate prior art documents as a prior art document. The selected prior art document may be one or more. Details of the large-scale language model LLM4 are as described above. This makes it possible to identify a prior art document that is appropriate as a comparison target with the claim under analysis from among several candidate prior art documents obtained based on similarity with keywords or abstracts and classification. The related technology selection unit 115 may also select a candidate selected by the user from among several candidates as a prior art document. Furthermore, if the user instructs the related technology selection unit 115 to have the computer select the prior art document, the related technology selection unit 115 may use the large-scale language model LLM4 to select a prior art document from among several candidates.

[0064] The difference generation unit 19 generates differences and commonalities between the claims under analysis and the prior art documents using the large-scale language model LLM5. Details of the large-scale language model LLM5 are as described above.

[0065] The affirmative opinion generation unit 12 is configured in the same manner as in Exemplary Embodiment 1, and is configured as follows: The affirmative opinion generation unit 12 uses the large-scale language model LLM6 to generate an affirmative opinion regarding the novelty and inventive step of the invention under analysis, based on the differences and commonalities between the claims under analysis and the prior art documents. Details of the large-scale language model LLM6 are as described above.

[0066] The negative opinion generation unit 13 is configured in the same manner as in Exemplary Embodiment 1, and is configured as follows. The negative opinion generation unit 13 uses the large-scale language model LLM7 to generate negative opinions regarding the novelty and inventive step of the branched invention based on the differences and commonalities between the claims and prior art documents under analysis. Details of the large-scale language model LLM7 are as described above.

[0067] The conclusion generation unit 14 is configured in the same manner as in Exemplary Embodiment 1, and is configured as follows: The conclusion generation unit 14 uses the large-scale language model LLM8 to generate conclusions regarding the novelty and inventive step of the subject invention based on the claims under analysis, prior art documents, affirmative and negative opinions.

[0068] The improvement unit 15 uses the large-scale language model LLM9 to generate a proposed claim improvement (an example of proposed improvement information) that shows an improvement to the claim to be analyzed (an example of information to be analyzed). Details of the large-scale language model LLM9 are as described above. The related technology acquisition unit 11, the affirmative opinion generation unit 12, the negative opinion generation unit 13, and the conclusion generation unit 14 function again with the proposed claim improvement as a new claim to be analyzed. For example, the improvement unit 15 may generate a proposed claim improvement if the conclusion does not satisfy the predetermined conditions described later. Alternatively, for example, the improvement unit 15 may generate a proposed claim improvement if the user instructs an improvement to the claim, regardless of whether the conclusion satisfies the predetermined conditions. As a result, the generation of proposed claim improvements is repeated recursively, making it possible to create proposed claim improvements while gradually increasing the possibility of novelty and inventive step being affirmed.

[0069] Both opinion presentation units 16 present the user with both positive and negative opinions. This improves the user's acceptance of the conclusion compared to simply presenting the conclusion.

[0070] The final draft presentation unit 17 presents the proposed claim improvement (an example of proposed improvement information) to the user as the final claim (an example of final draft information) if the conclusion generated using the proposed claim improvement as the claim to be analyzed satisfies predetermined conditions. The predetermined conditions may, for example, be that an affirmative opinion regarding novelty and inventive step is deemed valid. Alternatively, the predetermined conditions may be that an affirmative opinion regarding novelty at least is deemed valid. In other words, for example, if a negative opinion stating that the claim has novelty but lacks inventive step is deemed valid, the predetermined conditions may be considered satisfied. However, the predetermined conditions are not limited to these. This makes it possible to present the user with a final claim that has a higher probability of being affirmed for novelty and inventive step, thereby improving the user's satisfaction. (Information processing method S1A flow) The information processing device 1A, configured as described above, executes the information processing method S1A. Figure 5 is a flowchart showing the flow of the information processing method S1A. As shown in Figure 5, the information processing method S1A includes steps S101 to S113.

[0071] In step S101, the analysis target acquisition unit 18 acquires the claim to be analyzed.

[0072] Figure 6 is a schematic diagram showing an example of the analysis target input screen displayed on the display device 5 in step S101. As shown in Figure 6, screen example G1 includes a claim input area G11 and an operation object G12. The claim input area G11 accepts the input of a natural language sentence indicating the claim to be analyzed. The input natural language sentence is displayed in the claim input area G11. Although screen example G1 shows an example where one claim has been entered, multiple claims may be entered. The multiple claims may be in a parallel relationship or in a referencing relationship. For example, when an operation on the operation object G12 is accepted, the next step S102 is executed.

[0073] Steps S102 to S104 are an example of the related technology acquisition process. In step S102, the related technology acquisition unit 11 generates keywords or summaries from the claims to be analyzed. For example, when generating keywords, the keyword generation unit 111 generates keywords from the claims to be analyzed using the large-scale language model LLM1. Also, for example, when generating summaries, the summary generation unit 112 generates summaries from the claims to be analyzed using the large-scale language model LLM2. As previously mentioned, the choice between generating keywords and summaries will not be repeated in detail. Screen example G1 shows an example in which keywords are generated using a natural language sentence entered in the claim input area G11 as the claim to be analyzed.

[0074] Figure 7 is a schematic diagram showing an example of the generated keyword screen displayed on the display device 5 in step S102. As shown in Figure 7, the example screen G2 includes a keyword area G21, a search count setting area G22, a search target setting area G23, and an operation object G24. The keyword area G21 shows the keywords generated from the claims to be analyzed. In this example, four keywords are generated. The search count setting area G22 accepts an operation to set the number of prior art documents to be obtained as candidates. In this example, it is set to five. That is, five candidates are obtained as candidates for prior art documents. The search target setting area G23 accepts an operation to set the search target for the candidates for prior art documents. In this example, it is possible to select whether or not to set the patent documents recorded in the database 3 as search targets by the year of disclosure, and 2022 and 2023 are set as search targets. For example, when an operation on the operation object G24 is accepted, the next steps S103 to S105 are executed. Note that if a summary is generated instead of keywords, screen example G2 will include the area where the summary is displayed instead of the keyword area G21.

[0075] In step S103, the classification identification unit 113 uses the large-scale language model LLM3 to identify the classification of the invention represented by the claim to be analyzed.

[0076] In step S104, the candidate acquisition unit 114 retrieves multiple patent documents from database 3 based on their similarity to keywords or abstracts. The candidate acquisition unit 114 also acquires those patent documents that match the specified classification as multiple candidates for prior art documents. If only one candidate is obtained in step S104, that candidate is designated as the prior art document, and the next step S105 is omitted.

[0077] In step S105, the related technology selection unit 115 uses the large-scale language model LLM4 to select prior art documents from among multiple candidates. As mentioned above, instead of using the large-scale language model LLM4 to select prior art documents from multiple candidates, the related technology selection unit 115 may perform the selection based on user input. Prior art documents that are not selected from among the multiple candidates may be used for further fine-tuning of some or all of the large-scale language models LLM1 to LLM9, or for in-context learning, etc.

[0078] Figure 8 is a schematic diagram showing an example of a prior art screen displayed on the display device 5 in step S105. As shown in Figure 8, screen example G3 includes a prior art document area G31 and an operation object G32. The prior art document area G31 shows an overview of the prior art document selected by the related technology selection unit 115. In screen example G3, bibliographic information is displayed as an overview, but the prior art document area G31 may also include other information (e.g., abstract, independent claims, etc.). For example, when an operation on the operation object G32 is accepted, the next step S106 is executed.

[0079] In step S106, the difference generation unit 19 uses the large-scale language model LLM5 to generate differences and similarities between the claims and prior art documents to be analyzed. The difference generation unit 19 also presents the generated differences and similarities to the user, for example, by displaying them on the display device 5. If multiple claims are input as the claims to be analyzed, the difference generation unit 19 may generate differences and similarities for each claim.

[0080] Figure 9 is a schematic diagram showing an example of a difference screen displayed on the display device 5 in step S106. As shown in Figure 9, the example screen G4 includes a common area G41, a difference area G42, and an operation object G43. The common area G41 includes the common points of the claim and prior art document to be analyzed. The difference area G42 includes the differences between the claim and prior art document to be analyzed. For example, when an operation on the operation object G43 is accepted, the next steps S107 to S109 are executed.

[0081] Step S107 is an example of the affirmative opinion generation process. In step S107, the affirmative opinion generation unit 12 uses the large-scale language model LLM6 to generate an affirmative opinion regarding the novelty and inventive step of the subject invention as indicated by the claims to be analyzed, based on the claims to be analyzed and the prior art documents. If multiple claims are input as the claims to be analyzed, the affirmative opinion generation unit 12 may generate an affirmative opinion for each claim.

[0082] Step S108 is an example of the negative opinion generation process. In step S108, the negative opinion generation unit 13 uses the large-scale language model LLM7 to generate negative opinions regarding the novelty and inventive step of the subject invention as indicated by the claims to be analyzed, based on the claims to be analyzed and the prior art documents. If multiple claims are input as the claims to be analyzed, the negative opinion generation unit 13 may generate a negative opinion for each claim.

[0083] Furthermore, the execution order of steps S107 and S108 is not limited to the order described above; they may be executed in the reverse order, and some or all of them may be executed in parallel.

[0084] Step S109 is an example of the process of presenting both opinions. In step S109, the opinion presentation unit 16 presents the positive opinion and the negative opinion to the user, for example, by displaying them on the display device 5.

[0085] Figure 10 is a schematic diagram showing an example of both opinion screens displayed on the display device 5 in step S109. As shown in Figure 10, screen example G9 includes an affirmative opinion area G91 and operation objects G92 and G93. The affirmative opinion area G91 includes an affirmative opinion regarding novelty and an affirmative opinion regarding inventive step. The affirmative opinion includes the statement, "The subject invention is novel," and the statement, "...(omitted)...not disclosed," which is the basis for that statement. The affirmative opinion also includes the statement, "The subject invention has an inventive step," and the statement, "...(omitted)...not easily conceivable," which is the basis for that statement. If multiple claims are entered as the claims to be analyzed, the affirmative opinions may be classified and displayed for each claim. Operation object G92 accepts an operation to instruct the display of a negative opinion. When an operation is accepted for operation object G92, screen example G9 transitions to screen example G10, which will be described next.

[0086] Figure 11 is a schematic diagram showing another example of the two opinion screens displayed on the display device 5 in step S109. As shown in Figure 11, screen example G10 includes a negative opinion area G101 and operation objects G102 and G93. The negative opinion area G101 includes negative opinions regarding novelty and negative opinions regarding inventive step. The negative opinion includes a sentence stating "The subject invention lacks novelty" and a sentence indicating the basis for that opinion, "Element A: ~(omitted)~...". The negative opinion also includes a sentence stating "The subject invention lacks inventive step" and a sentence indicating the basis for that opinion, "It is considered that A and B are ~(omitted)~". If multiple claims are entered as the claims to be analyzed, the negative opinions may be classified and displayed for each claim. Operation object G102 accepts an operation to instruct the display of positive opinions. When an operation is accepted for operation object G102, screen example G10 transitions to the aforementioned screen example G9.

[0087] As described above, by allowing screen examples G9 and G10 to be switched between each other, the user can review both the positive and negative opinions regarding the novelty and inventive step of the invention. Alternatively, instead of switching between the positive opinion and both opinions as in screen examples G9 and G10, both opinions may be displayed on a single screen.

[0088] When an operation on the operation object G93 is accepted in screen example G9 or G10, the next step S110 is executed.

[0089] Step S110 is an example of the conclusion generation process. In step S110, the conclusion generation unit 14 uses the large-scale language model LLM8 to generate conclusions regarding the novelty and inventive step of the subject invention as presented in the claims under analysis, based on the claims under analysis, prior art documents, affirmative and negative opinions. The conclusion generation unit 14 also presents the generated conclusions to the user, for example, by displaying them on the display device 5.

[0090] In step S111, the control unit 110 determines whether the conclusion satisfies predetermined conditions. As mentioned above, the predetermined conditions may be that the positive opinion is deemed valid with respect to novelty and inventive step, or at least that the positive opinion is deemed valid with respect to novelty. The case where Yes is determined in step S111 will be described later. If No is determined in step S111, the next step S112 is executed.

[0091] Step S112 is an example of the improvement process. In step S112, the improvement unit 15 uses the large-scale language model LLM9 to generate proposed claim improvements based on the claims to be analyzed, prior art documents, differences and similarities, negative opinions, positive opinions, and conclusions. The improvement unit 15 also presents the generated proposed claim improvements to the user, for example, by displaying them on the display device 5.

[0092] Next, the control unit 110 repeats the process from step S102, treating the proposed claim improvement as a new claim to be analyzed.

[0093] Furthermore, if the decision is made to be Yes in step S111, step S113 is executed. In step S113, the final proposal presentation unit 17 presents to the user, for example, by displaying on the display device 5, the most recent claim improvement proposal, which is the claim under analysis whose conclusion satisfies predetermined conditions, as the final claim proposal.

[0094] Figure 12 is a schematic diagram showing an example of a conclusion screen that indicates a conclusion that has been determined to satisfy predetermined conditions in step S111. This conclusion screen is also an example of a conclusion screen that is displayed on the display device 5 in step S110. As shown in Figure 12, the example screen G13 includes a conclusion area G131 and a justification area G132. In this example, the conclusion area G131 contains the conclusion that the affirmative opinion is valid. The justification area G132 contains the text that provides the justification.

[0095] Figure 13 schematically shows an example of the final draft screen displayed on the display device 5 in step S113. As shown in Figure 13, the example screen G14 includes the final draft area G141. In the final draft area G141, the changes to the original claim being analyzed are displayed in a recognizable manner (in this example, with an underline). Note that the recognizable manner of the changes is not limited to underlining; it may also include a marker or the like.

[0096] (Effects of the information processing device 1A and the information processing method S1A) As described above, the information processing device 1A further includes an improvement unit 15 that generates improvement proposal information showing improvement proposals for the information to be analyzed using a fourth large-scale language model, and the related technology acquisition unit 11, the positive opinion generation unit 12, the negative opinion generation unit 13, the conclusion generation unit 14, and the improvement unit 15 are configured to further function as the information to be analyzed. Therefore, with the information processing device 1A, in addition to the effects achieved by the information processing device 1, the repeated generation of improvement proposal information can be used to gradually increase the possibility of acquiring intellectual property rights indicated by the information to be analyzed.

[0097] Furthermore, in the information processing device 1A, the related technology acquisition unit 11 employs a configuration in which it generates keywords from the information to be analyzed using a fifth large-scale language model and acquires related technology information from a database using the generated keywords. Therefore, in addition to the effects achieved by the information processing device 1, the information processing device 1A provides the effect of being able to acquire more appropriate related technology information.

[0098] Furthermore, in the information processing device 1A, the related technology acquisition unit 11 acquires multiple candidates for related technical information and uses a sixth large-scale language model to select one of the candidates as the related technical information. Therefore, in addition to the effects achieved by the information processing device 1, the information processing device 1A can acquire more appropriate related technical information.

[0099] Furthermore, in the information processing device 1A, the related technology acquisition unit 11 employs a configuration in which it identifies the classification of the intellectual property to be analyzed using a seventh large-scale language model and acquires related technical information using the identified classification. As a result, it is possible to acquire more appropriate related technical information.

[0100] Furthermore, the information processing device 1A employs a configuration that further includes both positive and negative opinion presentation units 16 that present both positive and negative opinions to the user. Therefore, in addition to the effects achieved by the information processing device 1, the information processing device 1A provides the effect of further improving the user's satisfaction with the conclusion by allowing the user to recognize both positive and negative opinions.

[0101] Furthermore, the information processing device 1A is configured to include a final proposal presentation unit 17 that presents the proposed improvement information to the user as final proposal information when the conclusion generated using the proposed improvement information as the information to be analyzed satisfies predetermined conditions. Therefore, in addition to the effects achieved by the information processing device 1, the information processing device 1A can provide the user with proposed improvements to the claims to increase the likelihood of obtaining patent rights.

[0102] (modified version) In the exemplary embodiment 2, the intellectual property to be analyzed is not limited to inventions. For example, other intellectual property described in natural language, such as inventions, may be applied as the intellectual property to be analyzed. Furthermore, the criteria for patentability are not limited to both novelty and inventive step, but either one may be applied. Furthermore, in addition to novelty and / or inventive step, other requirements may be applied as criteria for patentability. Furthermore, the information to be analyzed is not limited to claims. For example, information including summaries, detailed descriptions, or ideas may be applied as the information to be analyzed, either in place of or in addition to claims. Furthermore, non-patent literature may be applied as related technical information, not limited to patent literature. Furthermore, there may be more than one prior art document.

[0103] [Examples of implementation using software] Some or all of the functions of the devices constituting the information processing devices 1, 1A, and the information processing system 100A (hereinafter also referred to as "the above devices") may be implemented by hardware such as integrated circuits (IC chips) or by software.

[0104] In the latter case, each of the above devices is implemented, for example, by a computer that executes instructions for a program, which is software that realizes each function. An example of such a computer (hereinafter referred to as Computer C) is shown in Figure 14. Figure 14 is a block diagram showing the hardware configuration of Computer C, which functions as each of the above devices.

[0105] Computer C comprises at least one processor C1 and at least one memory C2. Memory C2 stores a program P that causes computer C to operate as each of the above-mentioned devices. In computer C, processor C1 reads program P from memory C2 and executes it, thereby realizing each of the above-mentioned devices.

[0106] For processor C1, for example, a CPU (Central Processing Unit), GPU (Graphic Processing Unit), DSP (Digital Signal Processor), MPU (Micro Processing Unit), FPU (Floating Point Number Processing Unit), PPU (Physics Processing Unit), TPU (Tensor Processing Unit), quantum processor, microcontroller, or a combination thereof can be used. For memory C2, for example, flash memory, HDD (Hard Disk Drive), SSD (Solid State Drive), or a combination thereof can be used.

[0107] Computer C may also be equipped with RAM (Random Access Memory) for loading program P at runtime and for temporarily storing various data. Furthermore, computer C may be equipped with communication interfaces for sending and receiving data with other devices. Additionally, computer C may be equipped with input / output interfaces for connecting input / output devices such as keyboards, mice, displays, and printers.

[0108] Furthermore, program P can be recorded on a non-temporary, tangible recording medium M that is readable by computer C. Such a recording medium M could be, for example, tape, disk, card, semiconductor memory, or programmable logic circuitry. Computer C can acquire program P via such a recording medium M. Program P can also be transmitted via a transmission medium. Such a transmission medium could be, for example, a communication network or broadcast waves. Computer C can also acquire program P via such a transmission medium.

[0109] Furthermore, each of the above functions of each of the above devices may be implemented by a single processor in a single computer, by multiple processors in a single computer working together, or by multiple processors in each of multiple computers working together. In addition, the programs for implementing each of the above functions in each of the above devices may be stored in a single memory in a single computer, distributed and stored in multiple memories in a single computer, or distributed and stored in multiple memories in each of multiple computers.

[0110] [Additional Note A] This disclosure includes the technologies described in the following appendices. However, the present invention is not limited to the technologies described in the following appendices, and various modifications are possible within the scope of the claims.

[0111] (Note A1) A means for acquiring related technology information that indicates related technologies related to the intellectual property to be analyzed, based on the information to be analyzed which is described in natural language, A positive opinion generation means that generates a positive opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the analysis target information and the related technical information, using a first large-scale language model, A negative opinion generation means that generates a negative opinion regarding the possibility of obtaining the rights based on the information to be analyzed and the related technical information using a second large-scale language model, A conclusion generation means that uses a third large-scale language model to generate a conclusion regarding the possibility of obtaining the rights based on the affirmative and negative opinions, An information processing device equipped with the following features.

[0112] (Appendix A2) The system further comprises an improvement means that generates improvement proposal information showing improvement proposals for the information to be analyzed, using a fourth large-scale language model. The aforementioned means for acquiring related technologies, the means for generating positive opinions, the means for generating negative opinions, the means for generating conclusions, and the means for improvement further function as the proposed improvement information as the information to be analyzed. The information processing device described in Appendix A1.

[0113] (Note A3) The aforementioned related technology acquisition means generates keywords or summaries from the information to be analyzed using a fifth large-scale language model, and acquires the related technology information from a database using the generated keywords or summaries. The information processing device described in Appendix A1 or A2.

[0114] (Note A4) The related technology acquisition means acquires a plurality of candidates for the related technical information and uses a sixth large-scale language model to select one of the plurality of candidates as the related technical information. An information processing device as described in any one of the appendices A1 to A3.

[0115] (Note A5) The aforementioned related technology acquisition means uses a seventh large-scale language model to identify the classification of the intellectual property to be analyzed, and uses the identified classification to acquire the related technology information. An information processing device described in any one of the appendices A1 through A4.

[0116] (Note A6) The system further includes means for presenting both the positive and negative opinions to the user. An information processing device as described in any one of the appendices A1 to A5.

[0117] (Note A7) The system further includes a final proposal presentation means that presents the aforementioned improvement proposal information to the user as final proposal information when the conclusion generated using the aforementioned improvement proposal information as the information to be analyzed satisfies predetermined conditions, The information processing device described in Appendix A2.

[0118] [Additional Notes B] This disclosure includes the technologies described in the following appendices. However, the present invention is not limited to the technologies described in the following appendices, and various modifications are possible within the scope of the claims.

[0119] (Note B1) At least one processor performs a related technology acquisition process to acquire related technology information indicating related technologies related to the intellectual property under analysis, based on the information under analysis, in which the intellectual property under analysis is described in natural language. The at least one processor performs an affirmative opinion generation process that uses a first large-scale language model to generate an affirmative opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the information under analysis and the related technical information. The at least one processor performs a negative opinion generation process that generates a negative opinion regarding the possibility of obtaining the rights based on the information to be analyzed and the related technical information, using a second large-scale language model. The at least one processor performs a conclusion generation process that uses a third large-scale language model to generate a conclusion regarding the possibility of obtaining the rights based on the affirmative and negative opinions, An information processing method that includes this.

[0120] (Note B2) The at least one processor further includes an improvement process that uses a fourth large-scale language model to generate improvement proposal information indicating improvement proposals for the information to be analyzed, The at least one processor further functions as the information to be analyzed, in the process of acquiring the relevant technology, generating positive opinions, generating negative opinions, generating conclusions, and improving the improvement process. The information processing method described in Appendix B1.

[0121] (Note B3) In the aforementioned related technology acquisition process, the at least one processor generates keywords or summaries from the information to be analyzed using a fifth large-scale language model, and acquires the related technology information from a database using the generated keywords or summaries. The information processing method described in Appendix B1 or B2.

[0122] (Note B4) In the aforementioned related technology acquisition process, the at least one processor acquires a plurality of candidates for the related technical information and uses a sixth large-scale language model to select one of the plurality of candidates as the related technical information. The information processing method described in any one of the appendices B1 to B3.

[0123] (Note B5) In the aforementioned related technology acquisition process, the at least one processor uses a seventh large-scale language model to identify the classification of the intellectual property to be analyzed, and uses the identified classification to acquire the related technology information. The information processing method described in any one of the appendices B1 to B4.

[0124] (Note B6) The at least one processor further includes both opinion presentation processing, which presents the positive opinion and the negative opinion to the user. The information processing method described in any one of the appendices B1 through B5.

[0125] (Note B7) The at least one processor further includes a final proposal presentation process in which, if the conclusion generated using the proposed improvement information as the information to be analyzed satisfies predetermined conditions, the proposed improvement information is presented to the user as the final proposal information. The information processing method described in Appendix B2.

[0126] [Additional Note C] This disclosure includes the technologies described in the following appendices. However, the present invention is not limited to the technologies described in the following appendices, and various modifications are possible within the scope of the claims.

[0127] (Note C1) A program that makes a computer function as an information processing device. The aforementioned computer, A means for acquiring related technology information that indicates related technologies related to the intellectual property to be analyzed, based on the information to be analyzed which is described in natural language, A positive opinion generation means that generates a positive opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the analysis target information and the related technical information, using a first large-scale language model, A negative opinion generation means that generates a negative opinion regarding the possibility of obtaining the rights based on the information to be analyzed and the related technical information using a second large-scale language model, A conclusion generation means that uses a third large-scale language model to generate a conclusion regarding the possibility of obtaining the rights based on the affirmative and negative opinions, An information processing program that functions as such.

[0128] (Note C2) The aforementioned computer, A fourth large-scale language model is used to further function as an improvement means for generating improvement proposal information that shows proposed improvements to the information to be analyzed. The aforementioned means for acquiring related technologies, the means for generating positive opinions, the means for generating negative opinions, the means for generating conclusions, and the means for improvement further function as the proposed improvement information as the information to be analyzed. The information processing program described in Appendix C1.

[0129] (Note C3) The aforementioned related technology acquisition means generates keywords or summaries from the information to be analyzed using a fifth large-scale language model, and acquires the related technology information from a database using the generated keywords or summaries. The information processing program described in Appendix C1 or C2.

[0130] (Note C4) The related technology acquisition means acquires a plurality of candidates for the related technical information and uses a sixth large-scale language model to select one of the plurality of candidates as the related technical information. An information processing program described in any one of the appendices C1 to C3.

[0131] (Note C5) The aforementioned related technology acquisition means uses a seventh large-scale language model to identify the classification of the intellectual property to be analyzed, and uses the identified classification to acquire the related technology information. An information processing program described in any one of the appendices C1 to C4.

[0132] (Appendix C6) The aforementioned computer, To further function as both means of presenting the aforementioned positive and negative opinions to the user, An information processing program described in any one of the appendices C1 to C5.

[0133] (Note C7) The aforementioned computer, When the conclusion generated using the aforementioned improvement proposal information as the information to be analyzed satisfies predetermined conditions, the improvement proposal information is further made to function as a final proposal presentation means, which presents the said improvement proposal information to the user as the final proposal information. The information processing program described in Appendix C2.

[0134] [Additional Note D] This disclosure includes the technologies described in the following appendices. However, the present invention is not limited to the technologies described in the following appendices, and various modifications are possible within the scope of the claims.

[0135] (Note D1) It comprises at least one processor, and the at least one processor is A related technology acquisition process that acquires related technology information indicating related technologies related to the intellectual property to be analyzed, based on the information to be analyzed which is described in natural language, A first large-scale language model is used to generate an affirmative opinion generation process that generates an affirmative opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the information under analysis and the related technical information. A negative opinion generation process that generates a negative opinion regarding the possibility of obtaining the rights based on the information to be analyzed and the related technical information using a second large-scale language model, A conclusion generation process that uses a third large-scale language model to generate a conclusion regarding the possibility of obtaining the rights based on the affirmative and negative opinions, An information processing device that performs this task.

[0136] The information processing device may also include memory. Furthermore, the memory may store a program that causes at least one processor to execute each of the aforementioned processes.

[0137] (Note D2) The aforementioned at least one processor, Further improvement processing is performed using a fourth large-scale language model to generate improvement proposal information that shows proposed improvements to the information to be analyzed. The aforementioned related technology acquisition process, the affirmative opinion generation process, the negative opinion generation process, the conclusion generation process, and the improvement process further function as the proposed improvement information as the information to be analyzed. The information processing device described in Appendix D1.

[0138] (Note D3) In the aforementioned related technology acquisition process, the at least one processor generates keywords or summaries from the information to be analyzed using a fifth large-scale language model, and acquires the related technology information from a database using the generated keywords or summaries. The information processing device described in Appendix D1 or D2.

[0139] (Note D4) In the aforementioned related technology acquisition process, the at least one processor acquires a plurality of candidates for the related technical information and uses a sixth large-scale language model to select one of the plurality of candidates as the related technical information. An information processing device as described in any one of the appendices D1 to D3.

[0140] (Note D5) In the aforementioned related technology acquisition process, the at least one processor uses a seventh large-scale language model to identify the classification of the intellectual property to be analyzed, and uses the identified classification to acquire the related technology information. An information processing device as described in any one of the appendices D1 to D4.

[0141] (Note D6) The aforementioned at least one processor, Further, the process of presenting both the positive and negative opinions to the user is performed. An information processing device as described in any one of the appendices D1 to D5.

[0142] (Note D7) The aforementioned at least one processor, If the conclusion generated using the aforementioned improvement proposal information as the information to be analyzed satisfies predetermined conditions, a final proposal presentation process is further executed to present the said improvement proposal information to the user as the final proposal information. The information processing device described in Appendix D2.

[0143] [Additional Note E] This disclosure includes the technologies described in the following appendices. However, the present invention is not limited to the technologies described in the following appendices, and various modifications are possible within the scope of the claims.

[0144] (Note E1) A program that makes a computer function as an information processing device. To the aforementioned computer, A related technology acquisition process that acquires related technology information indicating related technologies related to the intellectual property to be analyzed, based on the information to be analyzed which is described in natural language, A first large-scale language model is used to generate an affirmative opinion generation process that generates an affirmative opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the information under analysis and the related technical information. A negative opinion generation process that generates a negative opinion regarding the possibility of obtaining the rights based on the information to be analyzed and the related technical information using a second large-scale language model, A conclusion generation process that uses a third large-scale language model to generate a conclusion regarding the possibility of obtaining the rights based on the affirmative and negative opinions, A non-temporary recording medium that stores an information processing program that executes such a program. [Explanation of symbols]

[0145] 1. 1A Information Processing Device 100A Information Processing System 2. Large-scale language model storage 3 Databases 4 Input devices 5 Display device 11. Related Technology Acquisition Department 12 Positive opinion generation section 13 Negative opinion generation section 14 Conclusion generation part 15 Improvement Department 16. Both opinion presentation sections 17 Final proposal presentation section 18. Analysis Target Acquisition Section 19 Difference generation part 110 Control Unit 111 Keyword Generation Unit 112 Summary generator 113 Classification Specification Department 114 Candidate acquisition section 115 Related Technology Selection Section 120 Storage section C1 Processor C2 Memory

Claims

1. A means for acquiring related technology information that indicates related technologies related to the intellectual property to be analyzed, based on the information to be analyzed which is described in natural language, A positive opinion generation means that generates a positive opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the information under analysis and the related technical information, using a first large-scale language model, A negative opinion generation means that generates a negative opinion regarding the possibility of obtaining the rights based on the information to be analyzed and the related technical information using a second large-scale language model, A conclusion generation means that uses a third large-scale language model to generate a conclusion regarding the possibility of acquiring the rights based on the affirmative and negative opinions, An information processing device equipped with the following features.

2. The system further comprises an improvement means that generates improvement proposal information showing improvement proposals for the information to be analyzed, using a fourth large-scale language model. The aforementioned means for acquiring related technologies, the means for generating positive opinions, the means for generating negative opinions, the means for generating conclusions, and the means for improvement further function as the proposed improvement information as the information to be analyzed. The information processing apparatus according to claim 1.

3. The aforementioned related technology acquisition means generates keywords or summaries from the information to be analyzed using a fifth large-scale language model, and acquires the related technology information from a database using the generated keywords or summaries. The information processing apparatus according to claim 1 or 2.

4. The related technology acquisition means acquires a plurality of candidates for the related technology information and uses a sixth large-scale language model to select one of the plurality of candidates as the related technology information. The information processing apparatus according to claim 1 or 2.

5. The aforementioned related technology acquisition means uses a seventh large-scale language model to identify the classification of the intellectual property to be analyzed, and uses the identified classification to acquire the related technology information. The information processing apparatus according to claim 1 or 2.

6. The system further includes means for presenting both the positive and negative opinions to the user. The information processing apparatus according to claim 1 or 2.

7. The system further includes a final proposal presentation means that presents the aforementioned improvement proposal information to the user as final proposal information when the conclusion generated using the aforementioned improvement proposal information as the information to be analyzed satisfies predetermined conditions, The information processing apparatus according to claim 2.

8. At least one processor performs a related technology acquisition process to acquire related technology information indicating related technologies related to the intellectual property under analysis, based on the information under analysis, in which the intellectual property under analysis is described in natural language. The at least one processor performs an affirmative opinion generation process that uses a first large-scale language model to generate an affirmative opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the information under analysis and the related technical information. The at least one processor performs a negative opinion generation process that generates a negative opinion regarding the possibility of obtaining the rights based on the information to be analyzed and the related technical information, using a second large-scale language model. The at least one processor performs a conclusion generation process that uses a third large-scale language model to generate a conclusion regarding the possibility of obtaining the rights based on the affirmative and negative opinions, An information processing method that includes this.

9. An information processing program that causes at least one processor to function as an information processing device, A means for acquiring related technology information that indicates related technologies related to the intellectual property to be analyzed, based on the information to be analyzed which is described in natural language, A positive opinion generation means that generates a positive opinion regarding the possibility of acquiring rights to the intellectual property under analysis, based on the information under analysis and the related technical information, using a first large-scale language model, A negative opinion generation means that generates a negative opinion regarding the possibility of obtaining the rights based on the information to be analyzed and the related technical information using a second large-scale language model, A conclusion generation means that uses a third large-scale language model to generate a conclusion regarding the possibility of acquiring the rights based on the affirmative and negative opinions, An information processing program that functions as such.

Images