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

Abstract

To accurately investigate the relationship between patent documents and the relevant technical information. [Solution] Server 14 extracts the feature portion of the invention described in the independent claims within the patent document. Server 14 inputs an instruction to the generation model 20 to investigate the relationship between the feature portion and the technical information, and obtains the investigation results output from the generation model 20. Server 14 outputs the investigation results.

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 assisting in the understanding of claims in a patent document.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, consider the case of using generative AI (Artificial Intelligence) when investigating whether there is technology related to a patent (or where it exists) in the target technical information.

[0005] In such a case, for example, first, the content of the claims described in the claims of the patent document and the target technical information are input into the generative AI. Next, for example, an instruction sentence instructing to investigate the relevance between the content of the claims and the target technical information is input into the generative AI. Specifically, an instruction sentence for investigating whether there is information related to the content of the claims in the target technical information (or where the information related to the content of the claims exists in the technical information) is input into the generative AI. The generative AI outputs an investigation result regarding whether there is information related to the content of the claims in the target technical information (or where the information related to the content of the claims exists in the technical information) according to the instruction sentence. In this way, it is possible to investigate the relevance between the content of the claims and the target technical information using the generative AI.

[0006] However, claims in patent documents often describe both characteristic and non-characteristic aspects of the invention. Therefore, if the content of the claims is directly input into a generation AI, the AI ​​may search for whether or not the non-characteristic aspects of the invention exist within the target technical information. In particular, independent claims in patent documents often describe the prerequisites for establishing the invention, so if the content of independent claims is directly input into a generation AI, the accuracy of the search results may be poor.

[0007] The technology disclosed in Patent Document 1 above is a technology for assisting in the understanding of claims within a patent document, and is not a technology for investigating the relationship between a patent document and the technical information in question.

[0008] This disclosure is made in view of the above points and aims to provide an information processing device, an information processing method, and an information processing program that can accurately investigate the relationship between a patent document and the technical information in question. [Means for solving the problem]

[0009] An information processing device according to a first aspect of this disclosure includes: an extraction unit that extracts feature portions of an invention described in an independent claim within a patent document; an acquisition unit that inputs an instruction statement to the generation model instructing it to investigate the relationship between the feature portions and technical information, thereby acquiring the investigation results output from the generation model; and an output unit that outputs the investigation results. According to the information processing device according to a first aspect of this disclosure, the relationship between a patent document and the target technical information can be investigated with high accuracy.

[0010] An information processing device according to a second aspect of this disclosure is an information processing device according to a first aspect, wherein the acquisition unit inputs the description of at least a portion of the dependent claims in the patent document to the generation model, thereby acquiring the investigation results output from the generation model. According to the information processing device according to the second aspect of this disclosure, by considering the description of the dependent claims, the relationship between the patent document and the target technical information can be investigated with greater accuracy.

[0011] An information processing device according to a third aspect of this disclosure is an information processing device according to a first or second aspect, wherein the extraction unit acquires the feature portion output from the generation model by inputting an instruction statement to the generation model instructing the extraction of the feature portion, and the acquisition unit acquires the investigation result output from the generation model by inputting at least one of an instruction statement to the generation model instructing the generation model to investigate whether or not technology related to the feature portion exists in the technical information and an instruction statement to investigate the location in the technical information in which technology related to the feature portion exists. According to the information processing device according to a third aspect of this disclosure, it is possible to accurately investigate whether or not an invention described in an independent claim in a patent document exists in the technical information in question. Furthermore, according to the information processing device according to a third aspect of this disclosure, it is possible to accurately investigate the location in the technical information in which technology related to the feature portion exists.

[0012] The information processing method according to the fourth aspect of this disclosure is an information processing method in which a computer performs processing by inputting an instruction statement into a generation model that instructs the extraction of characteristic portions of the invention described in an independent claim within a patent document and to investigate the relationship between the characteristic portions and technical information, thereby obtaining the investigation results output from the generation model and outputting the investigation results. According to the information processing method according to the fourth aspect of this disclosure, the relationship between a patent document and the target technical information can be investigated with high accuracy.

[0013] The information processing program according to the fifth aspect of this disclosure is an information processing program that causes a computer to execute a process by inputting an instruction statement into a generation model that instructs the program to extract the characteristic portion of the invention described in an independent claim within a patent document and to investigate the relationship between the characteristic portion and technical information, thereby obtaining the investigation results output from the generation model and outputting the investigation results. According to the information processing program according to the fifth aspect of this disclosure, the relationship between a patent document and the target technical information can be investigated with high accuracy. [Effects of the Invention]

[0014] This disclosure provides an information processing device, an information processing method, and an information processing program that can accurately investigate the relationship between a patent document and the technical information it concerns. [Brief explanation of the drawing]

[0015] [Figure 1] This figure shows an example of the functional configuration of an information processing system according to an embodiment of the disclosed technology. [Figure 2] Block diagram showing the server hardware configuration. [Figure 3] This flowchart shows the flow of information processing by the server. [Modes for carrying out the invention]

[0016] Hereinafter, an example of an embodiment of this disclosure will be described with reference to the drawings. In each drawing, identical or equivalent components and parts are given the same reference numerals. Also, the dimensional ratios in the drawings are exaggerated for illustrative purposes and may differ from the actual ratios.

[0017] FIG. 1 is a block diagram showing an example of the functional configuration of an information processing system 10 according to an embodiment. As shown in FIG. 1, the information processing system 10 includes a user terminal 12 and a server 14 which is an example of an information processing apparatus. The user terminal 12 and the server 14 are communicably connected via a network 16 such as the Internet, for example.

[0018] The information processing system 10 according to the present embodiment is a system for investigating the relevance between the invention described in the claims in a patent document and the technical information targeted. The patent document refers to a document related to a patent such as a patent publication or a patent gazette. A patent document generally includes claims, a detailed description of the invention, an abstract, and drawings.

[0019] The server 14 causes a generation model, which is a known generation AI, to read a patent document and technical information to be compared with the patent document, gives a predetermined instruction text (prompt) to the generation model, and causes the generation model to output the correspondence between the invention described in the claims in the patent document and the technical information.

[0020] The technical information is at least one of text information, image information, and video information in which technical matters are expressed. The text information is information in which technology is expressed by text. For example, a patent document, a technical specification, a technical standard, a product manual, and a product pamphlet are examples of the text information. The image information is information in which technology is expressed by an image. For example, a drawing of a product in which technology is embodied is an example of the image information. The video information is information in which technology is expressed by a video. For example, a video when the operation of a product in which technology is embodied is photographed is an example of the video information.

[0021] As described above, when investigating the relevance between a patent document and the target technical information using a generation AI, for example, first, the content of the claims in the patent document and the target technical information are input to the generation AI.

[0022] Here, for example, consider the case where the content of the claim is "A control device for controlling X, the control device including A, B, and C." In such a case, "A control device for controlling X" is a part called the preamble, which is not the characteristic part of the invention but corresponds to a prerequisite part for the invention to be established. Also, for example, among the components A, B, and C that are components of the invention, A and C may not be the characteristic parts of the invention. Therefore, it is possible to assume that only the component B of the invention among the content of the claim "A control device for controlling X, the control device including A, B, and C." is the key part and the characteristic part of the invention. Or it is also possible to assume that a part of component A of the invention and a part of component B of the invention are the key part and the characteristic part of the invention.

[0023] In such a case, if the content of the claim "A control device for controlling X, the control device including A, B, and C." is directly input into the generative AI, for example, the generative AI may search for the "control device for controlling X" which is the preamble configuration in the technical information without searching for the component B which is the characteristic part of the invention in the technical information, and output the investigation result that there is a description of the technology related to the claim in the technical information. Such a state can also be said to be a state where noise is included in the input to the generative AI, and as a result, noise is also included in the investigation result, deteriorating the accuracy of the investigation result.

[0024] Therefore, in this embodiment, before causing the generative AI to investigate the relevance between the patent document and the technical information, the characteristic part of the invention described in the independent claim is extracted. And in this embodiment, by inputting the characteristic part of the invention extracted from the independent claim into the generative AI, the relevance between the content of the claim and the technical information is investigated using the generative AI. As a result, the noise included in the information input to the generative AI is reduced, and the noise included in the investigation result is also reduced, improving the accuracy of the investigation result. The following will specifically explain.

[0025] (User terminal) The user terminal 12 is operated by the user. The user terminal 12 exchanges information with the server 14 in response to the user's operations.

[0026] (server) Functionally, as shown in Figure 1, the server 14 comprises a data storage unit 140, a generation model storage unit 142, a reception unit 144, an extraction unit 146, an acquisition unit 148, and an output unit 150.

[0027] The data storage unit 140 stores various data received by the reception unit 144 and data used in the processing described later. For example, the data storage unit 140 stores patent document data, relevant technical information, and other data.

[0028] The generation model storage unit 142 stores a generation model 20, which is an example of a generation AI. The generation model 20 is a pre-trained model that has been machine-learned to generate and output text or images in response to input. For example, the generation model 20 may use any of the GPT series (e.g., GPT-3, GPT-4), BERT, or RoBERTa as a model that generates text according to input. Note that the generation model 20 may be stored on a device different from the server 14. In this case, the server 14 uses the generation model 20 by accessing the other device on which the generation model 20 is running. In this embodiment, the case in which the generation model 20 is stored in the generation model storage unit 142, which is an example of a storage unit of the server 14, will be described as an example.

[0029] The reception unit 144 receives input data from an external source. For example, the reception unit 144 receives various data transmitted from the user terminal 12. Alternatively, for example, the reception unit 144 receives various data downloaded from an external database (not shown) in accordance with instructions transmitted from the user terminal 12. For example, the reception unit 144 receives the patent number of a patent document transmitted from the user terminal 12 as various data, and downloads the patent publication data corresponding to that patent number from an external database (not shown). Also, for example, the reception unit 144 receives the relevant technical information as various data. Then, for example, the reception unit 144 stores the various data in the data storage unit 140.

[0030] The extraction unit 146 extracts the characteristic features of the invention described in the independent claims within the patent document. Specifically, the extraction unit 146 inputs an instruction statement to the generation model 20 instructing it to extract the characteristic features of the invention described in the independent claims, thereby obtaining the characteristic features output from the generation model 20.

[0031] For example, the instruction text includes the description of an independent claim. For example, the extraction unit 146 inputs the instruction text [<Insert description of independent claim> Extract the key part (or feature part) of the invention in the above claims] into the generation model 20, thereby obtaining the feature part of the invention related to the independent claim output from the generation model 20. The generation model 20 outputs information such as [Point 1: XXX, Point 2: YYY].

[0032] The acquisition unit 148 inputs an instruction statement to the generation model 20 instructing it to investigate the relationship between the feature portion extracted by the extraction unit 146 and the target technical information stored in the data storage unit 140, thereby acquiring the investigation results output from the generation model 20.

[0033] For example, the acquisition unit 148 inputs an instruction statement to the generation model 20 instructing it to investigate whether the technology associated with the feature portion extracted by the extraction unit 146 exists in the target technical information stored in the data storage unit 140. The generation model 20 outputs the results of the investigation into whether the technology associated with the feature portion exists in the target technical information.

[0034] Alternatively, for example, the acquisition unit 148 inputs an instruction statement to the generation model 20 that instructs it to investigate the locations where technologies related to the characteristic portion of the target technical information exist. The generation model 20 outputs the results of the investigation regarding the locations where technologies related to the characteristic portion of the target technical information exist.

[0035] For example, the instruction text includes a description of the feature part. For example, the acquisition unit 148 inputs the instruction text [<Insert the key point of the invention (description of the feature part)><Insert technical information similar to the key point of the invention (description of the feature part)>In the above technical information, is there a description similar to a part of the key point of the invention through inference and broad interpretation?] into the generation model 20, thereby acquiring the investigation results output from the generation model 20. The generation model 20 outputs information such as [{Judgment: Yes, Reason: AAA, Original text of the description: WWW}, {Judgment: No, Reason: BBB, Original text of the description: ZZZ}].

[0036] Furthermore, when the acquisition unit 148 investigates the relationship between the feature portion extracted by the extraction unit 146 and the target technical information stored in the data storage unit 140, it may vectorize the feature portion and the technical information before investigating the relationship between the feature portion and the technical information. For example, the acquisition unit 148 vectorizes the text corresponding to the feature portion using a known vectorization technique, and also vectorizes the information present in each part of the technical information using a known vectorization technique. For example, if the technical information is text information, each sentence written in each paragraph is vectorized. Also, for example, if the technical information is image information, each partial image of each part that makes up the image is vectorized. Also, for example, if the technical information is video information, each frame image that makes up the video is vectorized. Then, the acquisition unit 148 calculates the similarity between the vector of the feature portion and each vector of the information present in each part of the technical information, and inputs only the technical information of the parts where the similarity is above a predetermined threshold into the generation model 20 as the technical information to be investigated. For example, in the instruction above, the part that says "<Insert technical information similar to the key points of the invention (description of the feature)>" can be modified to insert only the technical information where the similarity is above a predetermined threshold. This eliminates the need to investigate all the information present in each part of the technical information, instead only investigating the technical information where the similarity to the feature is above a predetermined threshold, which can result in improved accuracy of the investigation results.

[0037] The output unit 150 outputs the survey results obtained by the acquisition unit 148.

[0038] The survey results output from the output unit 150 are transmitted to the user terminal 12.

[0039] The user operates the user terminal 12 and checks the survey results displayed on the display unit (not shown) of the user terminal 12.

[0040] The user terminal 12 and server 14 can be implemented by a computer 50, for example, as shown in Figure 2. The computer 50 that implements the user terminal 12 and server 14 includes a CPU 51, a memory 52 as a temporary storage area, and a non-volatile storage unit 53. The computer 50 also includes an input / output interface (I / F) 54 to which input / output devices (not shown) are connected, and a read / write (R / W) unit 55 that controls the reading and writing of data to the recording medium 59. The computer 50 also includes a network I / F 56 that connects to a network such as the Internet. The CPU 51, memory 52, storage unit 53, input / output I / F 54, R / W unit 55, and network I / F 56 are connected to each other via a bus 57.

[0041] The storage unit 53 can be implemented using a Hard Disk Drive (HDD), Solid State Drive (SSD), flash memory, etc. The storage unit 53, as a storage medium, stores programs that enable the computer to function. The CPU 51 reads the programs from the storage unit 53, loads them into memory 52, and sequentially executes the processes contained within the programs.

[0042] Therefore, each functional configuration of the server 14 is realized by the CPU 51 reading and executing an information processing program stored in the memory 52 or storage unit 53.

[0043] Next, the operation of the information processing system 10 of this embodiment will be described.

[0044] Figure 3 is a flowchart showing the flow of information processing by the server 14. The CPU 11 reads the information processing program from the storage unit 53, loads it into memory 52, and executes it, thereby performing the information processing.

[0045] In step 100, the extraction unit 146 reads the patent document data stored in the data storage unit 140. Then, in step 100, the extraction unit 146 extracts the feature portions of the invention described in the independent claims within the patent document. Specifically, in step 100, the extraction unit 146 inputs an instruction statement to the generation model 20 instructing it to extract the feature portions of the invention described in the independent claims, thereby obtaining the feature portions output from the generation model 20.

[0046] In step 102, the acquisition unit 148 inputs an instruction statement to the generation model 20 that instructs it to investigate the relationship between the feature portion extracted in step 100 and the target technical information stored in the data storage unit 140.

[0047] In step 104, the acquisition unit 148 acquires the survey results output from the generation model 20.

[0048] In step 106, the output unit 150 outputs the survey results obtained in step 104.

[0049] As described above, the server of the information processing system according to the embodiment extracts the feature portion of the invention described in the independent claims within the patent document. The server inputs an instruction statement to the generation model instructing it to investigate the relationship between the feature portion and the technical information, and obtains the investigation results output from the generation model. The server outputs the investigation results. This makes it possible to accurately investigate the relationship between the patent document and the target technical information.

[0050] Furthermore, the server inputs an instruction to the generative model that instructs it to extract the feature portion, thereby obtaining the feature portion output from the generative model. The server then inputs at least one of the following instructions to the generative model: an instruction to investigate whether the technology related to the feature portion exists within the target technical information, and an instruction to investigate the location within the technical information where the technology related to the feature portion exists. The server obtains the investigation results output from the generative model. This allows for an accurate investigation of whether the invention described in the independent claims within the patent document exists within the target technical information. It also allows for an accurate investigation of the location within the technical information where the technology related to the feature portion exists.

[0051] While embodiments of the present disclosure have been described in detail above with reference to the attached drawings, the technical scope of the present disclosure is not limited to these examples. The embodiments described above are illustrative and do not limit the technical scope of the present disclosure. It is clear to any person with ordinary skill in the art of the present disclosure that various modifications or alterations may occur within the scope of the technical idea set forth in the claims, and these modifications or alterations will naturally also fall within the technical scope of the present disclosure.

[0052] For example, the following variations can be given.

[0053] <Example 1> In the above embodiment, we have described an example in which the investigation results output from the generation model are obtained by inputting an instruction statement into the generation model that instructs the model to extract the characteristic parts of the invention described in the independent claims and to investigate the relationship between those characteristic parts and the technical information, but we are not limited to this. Since the content of independent claims is often abstract, the accuracy of the investigation results may be poor if only the independent claims are input into the generation model. For this reason, for example, the investigation results output from the generation model may be obtained by inputting the description of at least a part of the dependent claims in the patent document into the generation model. Since dependent claims are often more specific than independent claims, they may contain more specific descriptions regarding the characteristic parts of the invention. For this reason, by inputting the description of at least a part of the dependent claims in the patent document into the generation model, the relationship between the patent document and the target technical information can be investigated with greater accuracy. On the other hand, if the entire specification portion of the patent document is input into the generation model 20, it is assumed that the accuracy of the investigation results will be low because it contains many parts that are different from the characteristic parts of the invention.

[0054] <Modification 2> In the above embodiment, the example described was when the target technical information is image information or video information, and that such information is directly input to the generation model 20, but it is not limited to this. For example, when the target technical information is image information or video information, the image information or video information may be first converted into text using known technology (for example, by using the generation model 20 to convert the image information or video information into text) and then input to the generation model 20.

[0055] Furthermore, the effects described in the above embodiments are descriptive or illustrative, and are not limited to those described in the above embodiments. In other words, the technology relating to this disclosure may produce other effects that would be obvious to a person of ordinary skill in the art of this disclosure from the descriptions in the above embodiments, in addition to or in lieu of the effects described in the above embodiments.

[0056] Furthermore, the information processing that the CPU reads and executes in each of the above embodiments may be executed by various processors other than the CPU. Examples of such processors include PLDs (Programmable Logic Devices) such as FPGAs (Field-Programmable Gate Arrays) whose circuit configuration can be changed after manufacturing, and dedicated electrical circuits that are processors with circuit configurations specifically designed to execute specific processing, such as ASICs (Application Specific Integrated Circuits). In addition, the information processing may be executed by one of these various processors, or by a combination of two or more processors of the same or different types (for example, multiple FPGAs, and a combination of a CPU and an FPGA). More specifically, the hardware structure of these various processors is an electrical circuit that combines circuit elements such as semiconductor elements.

[0057] Furthermore, although the embodiments described above describe a configuration in which the information processing program is pre-stored (installed) in the memory unit, the invention is not limited thereto. The program may be provided in a form recorded on a non-transitory recording medium such as a CD-ROM (Compact Disk Read Only Memory), DVD-ROM (Digital Versatile Disk Read Only Memory), and USB (Universal Serial Bus) memory. Alternatively, the program may be provided in a form that is downloaded from an external device via a network. This disclosure may also be applied to program products. [Explanation of symbols]

[0058] 10. Information Processing Systems 14 Servers 20 Generative Models

Claims

1. An extraction unit that extracts the feature portion of the invention described in the independent claims within the patent document, An acquisition unit obtains the results of an investigation output from the generation model by inputting an instruction statement into the generation model that instructs it to investigate the relationship between the aforementioned feature portion and technical information. An output unit that outputs the results of the investigation, Information processing device including

2. The acquisition unit acquires the investigation results output from the generation model by inputting at least a portion of the description of the dependent claims in the patent document into the generation model. The information processing apparatus according to claim 1.

3. The extraction unit inputs an instruction statement to the generation model that instructs it to extract the feature portion, thereby acquiring the feature portion output from the generation model. The acquisition unit inputs at least one of the following to the generation model: an instruction to investigate whether or not technology related to the feature portion exists within the technical information, and an instruction to investigate the location within the technical information where technology related to the feature portion exists; thereby acquiring the investigation results output from the generation model. The information processing apparatus according to claim 1 or claim 2.

4. Extract the feature portion of the invention described in the independent claims within the patent document, By inputting an instruction statement into the generative model that instructs it to investigate the relationship between the aforementioned feature portion and the technical information, the investigation results output from the generative model are obtained. Output the results of the above investigation. An information processing method in which a computer performs the processing.

5. Extract the feature portion of the invention described in the independent claims within the patent document, By inputting an instruction statement into the generative model that instructs it to investigate the relationship between the aforementioned feature portion and the technical information, the investigation results output from the generative model are obtained. Output the results of the above investigation. An information processing program that causes a computer to perform a task.

Images