Information processing device, information processing method, computer program, and non-transitory computer-readable medium

Abstract

The present invention provides an information processing device, an information processing method, a computer program, and a non-transitory computer-readable medium that can reduce the time and cost required to come up with a name for a product or service related to an invention / idea described in a patent / utility model document. [Solution] The information processing method acquires a patent / utility model document including a first description of the problem that the invention / utility model is intended to solve, a second description of the effect of the invention / utility model, and a third description of the configuration of the invention / utility model, generates a prompt based on the patent / utility model document to cause a language model to generate a sentence including at least one candidate name for a product or service related to the invention / utility model described in the patent / utility model document, sends the prompt to the language model, and receives from the language model the sentence generated by the language model based on the prompt.

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, an information processing method, a computer program, and a non-transitory computer-readable medium. [Background technology]

[0002] Patent Document 1 discloses a trademark utilization support device that takes into consideration the utilization of a user's trademark and presents which trademark should be applied for and what application content should be applied for. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2023-059865 Summary of the Invention [Problem to be solved by the invention]

[0004] However, coming up with names for products and services often requires a great deal of time and cost, and therefore there is a need for an efficient method of coming up with names for products and services.

[0005] The present disclosure aims to provide an information processing device, an information processing method, a computer program, and a non-transitory computer-readable medium that can reduce the time and cost required to come up with a name for a product or service related to an invention / idea described in a patent / utility model document. [Means for solving the problem]

[0006] In order to achieve the above object, an information processing device according to one aspect includes: a processor; a memory that stores at least one instruction executable by the processor, the information processing device is capable of communicating with a language model; When at least one of the instructions is executed by the processor, the information processing device: Obtain a patent / utility model document including a first description of the problem that the invention / device is intended to solve, a second description of the effect of the invention / device, and a third description of the configuration of the invention / device; generating a prompt for causing the language model to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document based on the patent / utility model document; sending the prompt to the language model; The sentence generated by the language model based on the prompt is received from the language model.

[0007] In addition, an information processing method according to one aspect for achieving the above object includes: 1. An information processing method executed by an information processing device capable of communicating with a language model, comprising: Obtaining a patent / utility model document including a first description of a problem to be solved by the invention / device, a second description of an effect of the invention / device, and a third description of a configuration of the invention / device; generating a prompt based on the patent / utility model document to cause the language model to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document; sending the prompt to the language model; receiving from the language model the sentence generated by the language model based on the prompt.

[0008] Furthermore, a computer program according to one aspect for achieving the above object comprises: 1. A computer program comprising at least one instruction for execution by a processor of an information processing device in communication with a language model, the computer program comprising: At least one of the instructions, when executed by the processor, causes the information processing device to: Obtain a patent / utility model document including a first description of the problem that the invention / device is intended to solve, a second description of the effect of the invention / device, and a third description of the configuration of the invention / device; generating a prompt based on the patent / utility model document to cause the language model to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document; sending the prompt to the language model; The sentence generated by the language model based on the prompt is received from the language model.

[0009] In addition, a non-transitory computer-readable medium according to one aspect for achieving the above object includes: The above computer program is stored.

[0010] According to the information processing device having the above configuration, a prompt is generated based on a patent / utility model document to cause a language model to generate sentences including at least one candidate name for a product or service related to the invention / invention described in the patent / utility model document. The information processing device then transmits the generated prompt to the language model and receives from the language model sentences generated by the language model based on the prompt. In this way, the information processing device having the above configuration can receive sentences including candidate names for a product or service related to the invention / invention described in the patent / utility model document from the language model simply by acquiring the patent / utility model document. Therefore, by using sentences including at least one candidate name generated by the language model, a user can reduce the time and cost required to devise a name for a product or service related to the invention / invention described in the patent / utility model document. The information processing method, computer program, and non-transitory computer-readable medium having the above configuration can also achieve similar effects. [Effects of the Invention]

[0011] According to the present disclosure, it is possible to provide an information processing device, an information processing method, a computer program, and a non-transitory computer-readable medium that can reduce the time and cost required to come up with a name for a product or service related to an invention / idea described in a patent / utility model document. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating a language model, an information processing device, and a user terminal device according to an embodiment of the present disclosure (hereinafter referred to as this embodiment). [Figure 2] FIG. 2 is a flowchart illustrating an information processing method performed in the first embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of an input screen in the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of a prompt generated by the control unit of the information processing device in the first embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of a sentence displayed on the display unit of the user terminal device in the first embodiment. [Figure 6] FIG. 6 is a flowchart illustrating an information processing method performed in the second embodiment. [Figure 7] FIG. 7 is a diagram illustrating an example of an input screen in the second embodiment. [Figure 8] FIG. 8 is a diagram illustrating an example of a prompt generated by the control unit of the information processing device in the second embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of a sentence displayed on the display unit of the user terminal device in the second embodiment. [Figure 10] FIG. 10 is a flowchart illustrating an information processing method performed in the third embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of an input screen in the third embodiment. [Figure 12] FIG. 12 is a flowchart illustrating an information processing method executed in the fourth embodiment. [Figure 13]FIG. 13 is a diagram illustrating an example of an input screen in the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, an example of an embodiment of the present disclosure will be described with reference to the drawings.

[0014] (Overall composition) An information processing device 20 according to this embodiment will be described with reference to FIG. 1. FIG. 1 is a diagram illustrating a language model 10, an information processing device 20, and a user terminal device 30 according to an embodiment of the present disclosure. The language model 10 and the information processing device 20 can communicate with each other, for example, via the Internet. From the viewpoint of security, it is preferable that the information processing device 20 communicates with the language model 10 in a secure communication environment. The information processing device 20 and the user terminal device 30 can communicate with each other via wired or wireless communication.

[0015] The language model 10 is, for example, a large-scale language model constructed from a large-scale network trained with a large amount of text data. The language model 10 is constructed, for example, by a neural network. A neural network is a mathematical model that represents the characteristics of brain function through computer simulation. A neural network is, for example, a model in which artificial neurons (nodes) forming a network through synaptic connections change the strength of synaptic connections through learning, thereby acquiring problem-solving capabilities. The neural network includes, for example, an input layer, a hidden layer, and an output layer. The neural network acquires problem-solving capabilities through, for example, deep learning. Deep learning is performed using, for example, a Transformer including a Generative Pretrained Transformer (GPT), a Bidirectional Encoder Representations from Transformers (BERT), a Bidirectional and Auto-regressive Transformer (BART), etc., or a Recurrent Neural Network (RNN). The language model 10 is constructed by performing deep learning using, for example, a transformer.

[0016] Transformers are primarily used in the field of natural language processing. For example, when a Transformer has more than 100 billion parameters, it can perform natural language processing with high accuracy. A Transformer is an Encoder-Decoder model that uses only attention. Because a Transformer does not need to process time-series data sequentially, it can be parallelized more than a recurrent neural network, thereby reducing training time.

[0017] The language model 10 has previously learned a large amount of text data and is capable of solving various language processing tasks. Examples of language processing tasks include sentence generation and question answering. For example, when generating sentences, the language model 10 tokenizes the input sentence and converts it into a continuous numerical representation (multidimensional vector). Next, the encoder of the language model 10 analyzes the converted numerical representation and converts it into an intermediate representation. The decoder of the language model 10 then converts the intermediate representation back into a numerical representation, converts the tokens back into sentences, and outputs them. In this way, the language model 10 generates sentences based on the input sentences.

[0018] The information processing device 20 is, for example, a server such as a cloud server or an edge server. A cloud server is a server provided via the Internet. An edge server is a server provided via a network (for example, a local area network (LAN)) in an area closer to the user than the Internet. As illustrated in FIG. 1, the information processing device 20 includes an input unit 21, a control unit 22, and an output unit 23. These are connected to each other via a bus 24 so as to be able to communicate with each other.

[0019] The input unit 21 receives, for example, an instruction signal generated by the user terminal device 30 and a patent / utility model document input to the user terminal device 30 from the user terminal device 30. Details of the instruction signal generated by the user terminal device 30 will be described later. The patent / utility model document may be, for example, a document such as a specification prepared for a patent application or utility model registration application, or a publication related to the patent application or utility model registration application. The patent / utility model document may include, for example, a first description of the problem to be solved by the invention / device, a second description of the effects of the invention / device, and a third description of the configuration of the invention / device. However, the patent / utility model document may also include, in addition to the first, second, and third descriptions, a fourth description of a mode for implementing the invention / device. The first description may include, for example, a description of the problem to be solved by the invention / device. The second description may include, for example, a description of the effects of the invention / device. The third description may include, for example, at least one of a description of a means for solving the problem and a description of the scope of the patent / utility model registration claim. The fourth description includes, for example, a description of a mode for carrying out the invention / device. The input unit 21 transmits the input instruction signal and patent / utility model document to the control unit 22. In this specification, "invention / device" means an invention and / or device.

[0020] For example, a sentence generated by the language model 10 is input to the input unit 21 from the language model 10. The input unit 21 transmits the input sentence to the control unit 22.

[0021] The control unit 22 receives the patent / utility model document and sentences transmitted from the input unit 21, thereby acquiring the patent / utility model document and sentences generated by the language model 10 input to the user terminal device 30. The control unit 22 includes, as hardware components, a memory 221 and a processor 222. The memory 221 includes, for example, a read-only memory (ROM) storing various computer programs and a random access memory (RAM) having multiple work areas storing various computer programs executed by the processor 222. The ROM stores, for example, an artificial intelligence (AI) program for executing information processing, which will be described later. Such an AI program is, for example, a program (trained model) constructed by supervised machine learning or unsupervised machine learning. The processor 222 includes, for example, at least one of a central processing unit (CPU), a micro processing unit (MPU), a graphics processing unit (GPU), and a tensor processing unit (TPU). The CPU may be configured with multiple CPU cores. The GPU may be configured with multiple GPU cores. The processor 222 is configured to load a specified computer program from various computer programs stored in the ROM onto the RAM, and to execute information processing, which will be described later, in cooperation with the RAM.

[0022] In this embodiment, a non-transitory computer-readable medium may be used as a memory. A non-transitory computer-readable medium refers to any type of physical memory (such as RAM or ROM) that can store information or data that can be read by a processor. A non-transitory computer-readable medium may store instructions for execution by one or more processors. The term "non-transitory computer-readable medium" includes tangible items and excludes carrier waves and transitory signals (i.e., non-transitory). Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, programmable ROMs (PROMs), erasable PROMs (EPROMs), and flash ROMs).

[0023] The control unit 22 generates a prompt based on the patent / utility model document to cause the language model 10 to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document. The prompt generated by the control unit 22 will be described in detail later. The control unit 22 transmits the generated prompt to the language model 10. The language model 10 generates a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document based on the prompt generated by the control unit 22. The language model 10 transmits the generated sentence to the control unit 22. The control unit 22 receives the sentence generated by the language model 10. The control unit 22 transmits the sentence received from the language model 10 to the output unit 23.

[0024] The output unit 23 outputs the text received from the control unit 22 to the user terminal device 30 .

[0025] The user terminal device 30 is an electronic device operated by a user, such as a desktop PC, a notebook PC, a tablet terminal, a smartphone, etc. The user terminal device 30 includes an input unit 31 and a display unit 32.

[0026] The input unit 31 is configured to accept a user's input operation on the user terminal device 30 and to generate an electrical signal corresponding to the input operation. The input unit 31 is, for example, an input device such as a keyboard electrically connected to the user terminal device 30, or a touch panel superimposed on the display unit 32 of the user terminal device 30. The user, for example, inputs a patent / utility model document and performs operations on the input unit 31 to cause the language model 10 to generate a sentence including at least one candidate name related to a product or service related to the invention / concept described in the patent / utility model document. Upon the above input and operation, the input unit 31 generates an instruction signal to cause the language model 10 to generate the sentence, and then inputs the instruction signal and the patent / utility model document to the input unit 21 of the information processing device 20.

[0027] The display unit 32 is, for example, a liquid crystal display, etc. The display unit 32 is configured to display text output from the output unit 23 of the information processing device 20.

[0028] (First Example) Next, a first embodiment will be described with reference to Figs. 2 to 5. Fig. 2 is a flowchart illustrating an information processing method executed in the first embodiment. Fig. 3 is a diagram illustrating an input screen in the first embodiment. Fig. 4 is a diagram illustrating a prompt generated by the control unit 22 of the information processing device 20 in the first embodiment. Fig. 5 is a diagram illustrating a sentence displayed on the display unit 32 of the user terminal device 30 in the first embodiment.

[0029] In this embodiment, the specification and claims (an example of a patent utility model document) prepared for a patent application as illustrated in Fig. 3 are input to the input unit 31 of the user terminal device 30. The specification and claims input to the input unit 31 in this embodiment include a first description of the problem to be solved by the invention, a second description of the effects of the invention, a third description of the configuration of the invention, and a fourth description of the mode for carrying out the invention. As illustrated in Fig. 3, the invention related to the specification and claims is an invention related to a communication terminal (an example of a product).

[0030] As illustrated in Fig. 2, the input of a patent / utility model document and an operation for causing the language model 10 to generate sentences including at least one candidate name related to a product or service relating to the invention / device described in the patent / utility model document are performed on the input unit 31 (STEP 01). After the user inputs the specification and claims illustrated in Fig. 3, the user presses the start button B1. When the input and operation are performed, the input unit 31 transmits to the information processing device 20 an instruction signal for causing the language model 10 to generate sentences including at least one candidate name related to a communication terminal relating to the invention described in the input specification and claims, as well as the input specification and claims.

[0031] 2, the information processing device 20 acquires a patent / utility model document by inputting the patent / utility model document from the user terminal device 30 to the input unit 21 of the information processing device 20 (STEP 02). In this embodiment, the information processing device 20 acquires the specification and claims prepared for the patent application as shown in FIG.

[0032] As illustrated in FIG. 2, the control unit 22 determines whether the patent utility model document includes a fourth description (STEP 03). If the control unit 22 determines that the patent utility model document includes a fourth description (YES in STEP 03), the control unit 22 generates a prompt based on the patent utility model document, including a first portion P1 (see FIG. 4) based on the first and fourth descriptions, a second portion P2 (see FIG. 4) based on the second and fourth descriptions, and a third portion P3 (see FIG. 4) based on the third and fourth descriptions (STEP 04). On the other hand, if the control unit 22 determines that the patent utility model document does not include a fourth description (NO in STEP 03), the control unit 22 generates a prompt based on the patent utility model document, including a first portion P1 based on the first description, a second portion P2 based on the second description, and a third portion P3 based on the third description (STEP 05). In this embodiment, the specification acquired by the information processing device 20 includes the fourth description. Therefore, the control unit 22 determines that the specification and claims include the fourth description (YES in STEP 03), and generates a prompt based on the patent utility model document, including a first part P1 based on the first description and the fourth description, a second part P2 based on the second description and the fourth description, and a third part P3 based on the third description and the fourth description (STEP 04).

[0033] Here, an example of the process by which the control unit 22 generates a prompt will be described in detail. The control unit 22 extracts, from the patent and utility model document, a description written in the section on the problem to be solved by the invention as a first description. The control unit 22 extracts, from the patent and utility model document, a description written in the section on the effects of the invention as a second description. The control unit 22 extracts, from the patent and utility model document, a description written in the first claim (i.e., claim 1) as a third description. The control unit 22 extracts, from the patent and utility model document, a description on a mode for carrying out the invention as a fourth description.

[0034] Next, the control unit 22 generates a prompt by replacing at least one word included in at least one of the first description, the second description, and the third description based on the fourth description and then changing each description into a sentence in bulleted format. The first description, the second description, and the third description are often abstract, while the fourth description is often more specific than the first description, the second description, and the third description. Therefore, the control unit 22 can generate a prompt including a specific description by replacing at least one word included in at least one of the first description, the second description, and the third description based on the fourth description. The control unit 22 may also generate a prompt by rewriting at least one of the first description, the second description, and the third description based on the fourth description and then changing each description into a sentence in bulleted format. In other words, the control unit 22 may not only replace words, but also rewrite at least one of the first description, the second description, and the third description. In this case, the control unit 22 can generate a prompt including a specific description by rewriting at least one of the first description, the second description, and the third description based on the fourth description.

[0035] The inventors have realized that inputting a prompt including a specific description into the language model 10 causes the language model 10 to generate a more accurate completion for the prompt than inputting a prompt including an abstract description into the language model 10. A completion is a response from the language model 10 to the prompt. Therefore, in this embodiment, a completion is a sentence generated by the language model 10 based on the prompt. Based on this realization, in this embodiment, when a fourth description is included in a patent utility model document, the control unit 22 is designed to generate a prompt including a first portion P1 based on the first and fourth descriptions, a second portion P2 based on the second and fourth descriptions, and a third portion P3 based on the third and fourth descriptions.

[0036] In this example, based on the description of the mode for carrying out the invention, the control unit 22 replaces "communication terminal" in the first, second, and third descriptions with "mobile phone," "audio input unit" in the third description with "microphone," "audio output unit" in the third description with "speaker," "photographing unit" in the third description with "camera," and "input unit" in the third description with "push button." The control unit 22 also converts the description in the "Problem to be Solved by the Invention" column into a bulleted list format by inserting line breaks based on periods. Note that in this example, because the description in the "Advantages of the Invention" column contains only one sentence, the control unit 22 does not convert the description in the "Advantages of the Invention" column into a bulleted list format.

[0037] As illustrated in FIG. 4 , the prompt generated in this embodiment includes a first portion P1 based on the first and fourth descriptions, a second portion P2 based on the second and fourth descriptions, and a third portion P3 based on the third and fourth descriptions. The first, second, and third portions P1, P2, and P3 are each composed of itemized sentences. In this embodiment, the third description is the description of the first claim (i.e., claim 1), so the third portion P3 is based on the portion of the claims corresponding to the first claim (i.e., claim 1). The prompt also includes a command statement based on the instruction signal and a detailed description based on the patent utility model document. The command statement and the detailed description are written separately. The command statement is essentially unchanged regardless of the content input. The command statement commands the system to devise candidate names based on the detailed description. Based on this command, the control unit 22 outputs the detailed description of the content illustrated in FIG. 4.

[0038] Returning to FIG. 2 , STEP 06 will be described. The control unit 22 transmits the prompt generated in STEP 04 or STEP 05 to the language model 10 (STEP 06). In this embodiment, since the control unit 22 executes STEP 05, the control unit 22 transmits a prompt including a first portion P1 based on the first description and the fourth description, a second portion P2 based on the second description and the fourth description, and a third portion P3 based on the third description and the fourth description to the language model 10 (STEP 06). Upon receiving the prompt from the control unit 22, the language model 10 generates a sentence including at least one candidate name related to the product related to the invention described in the patent / utility model document based on the prompt. The language model 10 transmits the generated sentence to the information processing device 20.

[0039] The control unit 22 receives a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document from the language model 10 (STEP 07). In this embodiment, the control unit 22 receives a sentence including at least one candidate name for a product related to the invention described in the patent / utility model document from the language model 10. The control unit 22 transmits the received sentence to the output unit 23.

[0040] The output unit 23 outputs a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document to the user terminal device 30 (STEP 08). In this embodiment, the control unit 22 outputs a sentence including at least one candidate name for a product related to the invention described in the patent / utility model document to the user terminal device 30. As a result, the display unit 32 of the user terminal device 30 displays a sentence such as the one shown in FIG. 5.

[0041] As illustrated in FIG. 5 , in this embodiment, the sentence generated by the language model 10 includes multiple candidate names. In this embodiment, candidate names such as “clear communicator,” “bimodal communicator,” and “intelligent communicator” are presented. For example, in this embodiment, the language model 10 recognizes from the received prompt that it is requested to devise candidate names for a product related to an invention related to a “mobile phone,” and devise candidate names including the word “communicator,” which has a high relevance to “mobile phone.” By visually checking the sentence displayed on the display unit 32, the user can recognize candidate names for the product related to the invention described in the patent / utility model document. In other words, the user can obtain candidate names for the product related to the invention described in the patent / utility model document simply by inputting the patent / utility model document into the input unit 31 of the user terminal device 30, thereby efficiently devising product names.

[0042] The information processing device 20 having the above configuration generates a prompt for causing the language model 10 to generate sentences including at least one candidate name for a product according to the invention described in the patent / utility model document, based on the patent / utility model document. The information processing device 20 then transmits the generated prompt to the language model 10 and receives from the language model 10 sentences generated by the language model 10 based on the prompt. In this manner, the information processing device 20 can receive sentences including candidate names for a product according to the invention described in the patent / utility model document from the language model 10 simply by acquiring the patent / utility model document. Therefore, by using sentences including at least one candidate name generated by the language model 10, a user can reduce the time and cost required to devise a name for a product according to the invention described in the patent / utility model document. The information processing method, computer program, and non-transitory computer-readable medium having the above configuration can also achieve similar effects.

[0043] Furthermore, according to the information processing device 20 having the above configuration, the first part P1, the second part P2, and the third part P3 included in the prompt are composed of sentences in bulleted format. When the prompt is written in bulleted format, it is easier to receive a more appropriate answer from the language model. Therefore, according to the information processing device 20, it is easier to receive a sentence including more accurate candidate names from the language model 10.

[0044] Furthermore, in the information processing device 20 having the above configuration, the first portion P1 is based on the first description and the fourth description, the second portion P2 is based on the second description and the fourth description, and the third portion P3 is based on the third description and the fourth description. The first, second, and third descriptions are often abstract, while the fourth description is more specific than the first, second, and third descriptions. The inventors have noticed that the language model 10 generates a more accurate completion for a prompt when a prompt including a specific description is input to the language model 10 than when a prompt including an abstract description is input to the language model 10. Therefore, the information processing device 20 can generate a more appropriate prompt, and as a result, can receive a sentence including a more accurate candidate name from the language model 10.

[0045] Furthermore, the information processing device 20 having the above configuration replaces at least one word included in the first description, the second description, and the third description based on the fourth description. Therefore, the information processing device 20 generates a prompt including a specific description. Therefore, the information processing device 20 can generate a more appropriate prompt, and as a result, can receive a sentence including more accurate candidate names from the language model.

[0046] Furthermore, according to the information processing device 20 having the above configuration, the third part P3 is based on the part corresponding to the first claim (i.e., claim 1) among the claims, so that a concise prompt can be generated.

[0047] (Second Example) Next, a second embodiment will be described with reference to FIGS. 6 to 9. In this embodiment, parts similar to those in the first embodiment will be described using the same reference numerals as in the first embodiment, and descriptions of overlapping parts will be omitted where appropriate. FIG. 6 is a flowchart illustrating an information processing method executed in the second embodiment. FIG. 7 is a diagram illustrating an input screen in the second embodiment. FIG. 8 is a diagram illustrating a prompt generated by the control unit 22 of the information processing device 20 in the second embodiment. FIG. 9 is a diagram illustrating a sentence displayed on the display unit 32 of the user terminal device 30 in the second embodiment. This embodiment differs from the first embodiment in that not only the patent / utility model document but also predetermined conditions are input into the input unit 31 of the user terminal device 30. The predetermined conditions are set by the user. For example, the predetermined condition is that the candidate name includes at least one word set by the user.

[0048] In this embodiment, the information processing method illustrated in FIG. 6 is executed. The control unit 22 generates a prompt based on the patent utility model document and predetermined conditions. In this embodiment, the prompt also includes an instruction sentence and a detailed description, but the instruction sentence in this embodiment is based on an instruction signal and predetermined conditions. In this embodiment, the instruction sentence and the detailed description are written separately from each other.

[0049] As illustrated in FIG. 6, the input unit 31 inputs a patent / utility model document and predetermined conditions, and performs an operation to cause the language model 10 to generate a sentence including at least one candidate name related to a product or service related to the invention / device described in the patent / utility model document (STEP 11). After inputting the specification and claims illustrated in FIG. 7 and the predetermined conditions, the user presses the start button B2. Upon performing the input and operation, the input unit 31 transmits to the information processing device 20 an instruction signal to cause the language model 10 to generate a sentence including at least one candidate name related to a communication terminal related to the invention described in the input specification and claims, the input specification and claims, and the predetermined conditions. Note that the predetermined condition in this embodiment is that the candidate name must include "smart" (an example of a word set by the user).

[0050] 6, the information processing device 20 acquires the patent utility model document and the predetermined conditions by inputting the patent utility model document and the predetermined conditions into the input unit 21 of the information processing device 20 from the user terminal device 30 (STEP 12). In this embodiment, the information processing device 20 acquires the specification and claims prepared for the patent application, as well as the predetermined conditions, as shown in FIG.

[0051] As illustrated in FIG. 6, STEP 13 is the same as STEP 03 in the first embodiment. If the control unit 22 determines that the patent utility model document contains a fourth description (YES in STEP 13), it generates a prompt including a first portion P1 based on the first and fourth descriptions, a second portion P2 based on the second and fourth descriptions, and a third portion P3 based on the third and fourth descriptions, based on the patent utility model document and predetermined conditions (STEP 14). On the other hand, if the control unit 22 determines that the patent utility model document does not contain a fourth description (NO in STEP 13), it generates a prompt including a first portion P1 based on the first description, a second portion P2 based on the second description, and a third portion P3 based on the third description, based on the patent utility model document and predetermined conditions (STEP 15). In this embodiment, the specification acquired by the information processing device 20 contains a fourth description. Therefore, the control unit 22 determines that the specification and claims include the fourth description (YES in STEP 13), and generates a prompt including a first part P1 based on the first description and the fourth description, a second part P2 based on the second description and the fourth description, and a third part P3 based on the third description and the fourth description based on the patent utility model document and specified conditions (STEP 14).

[0052] Here, the prompt generated in this embodiment will be described with reference to Figure 8. As shown in Figure 8, the prompt generated in this embodiment differs from the prompt generated in the first embodiment in that the command statement contains a sentence based on a predetermined condition (specifically, the sentence "Please include the word 'smart' in the candidate names"). In other respects, the prompt generated in this embodiment is similar to the prompt generated in the first embodiment, so a detailed description will be omitted.

[0053] As shown in Fig. 6, steps 16 to 18 are the same as steps 06 to 08 in the first embodiment. However, in this embodiment, the display unit 32 of the user terminal device 30 displays the text shown in Fig. 9.

[0054] As illustrated in FIG. 9, in this embodiment, the sentence generated by the language model 10 also includes multiple candidate names. However, the multiple candidate names in this embodiment include the word "smart" set by the user. Therefore, in this embodiment, candidate names including "smart," such as "smart communicator," "smart multi-communicator," and "smart communication assist," are presented. By visually checking the sentence displayed on the display unit 32, the user can recognize candidate names related to the product according to the invention described in the patent utility model document that include the word "smart" set by the user.

[0055] In this embodiment, the information processing device 20 can also achieve the same effects as in the first embodiment.

[0056] Furthermore, according to the information processing device 20 having the above configuration, a prompt is generated based on the patent / utility model document and predetermined conditions set by the user, so that a sentence including a candidate name that is in line with the user's intention can be received from the language model 10.

[0057] Furthermore, according to the information processing device 20 having the above configuration, the predetermined condition is that the candidate name must include at least one word set by the user, so that it is possible to receive from the language model 10 a sentence including a candidate name that is in line with the user's desire to include a desired word.

[0058] Furthermore, according to the information processing device 20 having the above configuration, the prompt includes an imperative sentence based on a predetermined condition and a detailed description based on a patent / utility model document, with the imperative sentence and the detailed description written separately from each other. A prompt in which the imperative sentence and the detailed description are written separately from each other is more likely to receive a more appropriate answer from the language model 10 than a prompt that does not have this configuration. Therefore, according to the information processing device 20, it is possible to more easily receive a sentence including more accurate candidate names from the language model 10.

[0059] (Third Example) Next, a third embodiment will be described with reference to FIGS. 1, 3 to 5, 10, and 11. In this embodiment, parts similar to those in the first embodiment will be described using the same reference numerals, and overlapping parts will not be described again. FIG. 10 is a flowchart illustrating an information processing method executed in the third embodiment. FIG. 11 is a diagram illustrating an input screen in the third embodiment. This embodiment differs from the first embodiment in that case-specific information is entered into the input unit 31 of the user terminal device 30 instead of a patent or utility model document. The case-specific information is, for example, information for uniquely identifying a case related to a patent application or utility model registration application filed with the Japan Patent Office. Examples of the case-specific information include a patent application number, a utility model registration application number, a publication number, a patent number, and a utility model registration number. In this embodiment, the case-specific information is the patent application number, JP 2024-999999. It should be noted that the specification and claims of Japanese Patent Application No. 2024-999999 are the same as the specification and claims input to the input unit 31 of the user terminal device 30 in the first embodiment.

[0060] As indicated by the dashed line in FIG. 1 , in this embodiment, the information processing device 20 can communicate with an external database 40, for example, via the Internet. The external database 40 may be, for example, a database of the Japan Patent Office, a database of the United States Patent and Trademark Office, a database of the World Intellectual Property Organization (WIPO), or a commercial database provided by an intellectual property service provider. In this embodiment, the external database 40 is a database of the Japan Patent Office. For example, the Japan Patent Office database records application information related to patent applications and utility model registration applications filed with the Japan Patent Office. The application information includes various information such as patent application number, utility model registration application number, publication number, patent number, utility model registration number, specification, patent claims, and utility model registration claims. It is also assumed that the external database 40 records application information related to Japanese Patent Application No. 2024-999999. Furthermore, databases other than the Japan Patent Office database (for example, the United States Patent and Trademark Office database) also record application information related to patent applications and utility model registration applications in at least one country, just like the Japan Patent Office database. Note that in this embodiment, the term "country" is a concept that includes not only independent countries, but also countries and regions that are not internationally recognized, and unions such as the European Union.

[0061] 10 , the input unit 31 inputs case-specific information and performs an operation to cause the language model 10 to generate a sentence including at least one candidate name for a product or service related to an invention / device described in a patent / utility model document related to the case specified by the case-specific information (STEP 21). After inputting a patent application number (an example of case-specific information) shown in FIG. 11 as an example, the user presses the start button B3. When the input and operation are performed, the input unit 31 transmits to the information processing device 20 an instruction signal to cause the language model 10 to generate a sentence including at least one candidate name for a communication terminal related to an invention described in a patent / utility model document related to the case specified by the input patent application number, and the input patent application number.

[0062] 10, the information processing device 20 acquires case-specific information by inputting the case-specific information from the user terminal device 30 to the input unit 21 of the information processing device 20 (STEP 22). In this embodiment, the information processing device 20 acquires the patent application number of Japanese Patent Application No. 2024-999999 as the case-specific information.

[0063] The control unit 22 acquires the patent and utility model document from the external database 40 based on the case-specific information (STEP 23). In this embodiment, the case-specific information is the patent application number of Japanese Patent Application No. 2024-999999, so the control unit 22 acquires the patent and utility model document relating to Japanese Patent Application No. 2024-999999 from the external database 40. The specification and claims relating to Japanese Patent Application No. 2024-999999 are the same as the specification and claims exemplified in FIG. 3.

[0064] As illustrated in Fig. 10, steps 24 to 29 are the same as steps 03 to 08 in the first embodiment. Therefore, in this embodiment, the prompt generated by control unit 22 is the prompt illustrated in Fig. 4, and the text displayed on display unit 32 of user terminal device 30 is the text illustrated in Fig. 5.

[0065] In this embodiment, the information processing device 20 can also achieve the same effects as in the first embodiment.

[0066] Furthermore, the information processing device 20 having the above configuration acquires a patent / utility model document from the external database 40 based on a patent application number (an example of case-specific information). Based on the patent / utility model document, the information processing device 20 generates a prompt for causing the language model 10 to generate a sentence including at least one candidate name for a product related to the invention described in the patent / utility model document. The information processing device 20 then transmits the generated prompt to the language model 10 and receives from the language model 10 a sentence generated by the language model 10 based on the prompt. In other words, when a user simply inputs a patent application number into the input unit 31 of the user terminal device 30, the information processing device 20 generates a prompt for causing the language model 10 to generate a sentence including a candidate name for a product related to the invention described in the patent / utility model document associated with the patent application number. In this way, the information processing device 20 can reduce the burden on the user.

[0067] (Fourth Example) Next, a fourth embodiment will be described with reference to Figures 7 to 9, 12, and 13. In this embodiment, parts similar to those in the first to third embodiments will be described using the same reference numerals as those in the first to third embodiments, and descriptions of overlapping parts will be omitted as appropriate. Figure 12 is a flowchart illustrating an information processing method executed in the fourth embodiment. Figure 13 is a diagram illustrating an input screen in the fourth embodiment. This embodiment differs from the third embodiment in that not only the case-specific information but also predetermined conditions are input into the input unit 31 of the user terminal device 30. The predetermined conditions in this embodiment are the same as those in the second embodiment.

[0068] 12, the input unit 31 inputs case-specific information and predetermined conditions, and performs an operation to cause the language model 10 to generate a sentence including at least one candidate name for a product or service related to an invention / device described in a patent / utility model document related to the case specified by the case-specific information (STEP 31). After inputting the patent application number (an example of case-specific information) and the predetermined conditions shown in FIG. 13, the user presses the start button B4. When the input and operation are performed, the input unit 31 transmits to the information processing device 20 an instruction signal to cause the language model 10 to generate a sentence including at least one candidate name for a communication terminal related to an invention described in a patent / utility model document related to the case specified by the input patent application number, as well as the input patent application number and the predetermined conditions.

[0069] 12, the information processing device 20 acquires the case-specific information and the predetermined conditions by inputting the case-specific information and the predetermined conditions from the user terminal device 30 to the input unit 21 of the information processing device 20 (STEP 32). In this embodiment, the information processing device 20 acquires the patent application number (Patent Application No. 2024-999999) and the predetermined conditions shown in FIG.

[0070] As illustrated in Fig. 12, STEP 33 is the same as STEP 23 in the third embodiment. STEP 34 is the same as STEP 03 in the first embodiment. STEP 35 to STEP 39 are the same as STEP 14 to STEP 18 in the second embodiment. Therefore, in this embodiment, the prompt generated by control unit 22 is the prompt illustrated in Fig. 8, and the text displayed on display unit 32 of user terminal device 30 is the text illustrated in Fig. 9.

[0071] In this embodiment, the information processing device 20 can also achieve the same effects as those of the first to third embodiments.

[0072] The above-described embodiments are provided to facilitate understanding of the present disclosure, and are not intended to limit the present disclosure, which may be modified or improved without departing from the spirit and scope of the present disclosure.

[0073] In the above embodiment, the patent / utility model document includes a first description of the problem to be solved by the invention / device, a second description of the effect of the invention / device, a third description of the configuration of the invention / device, and a fourth description of the mode for implementing the invention / device, but the present disclosure is not limited thereto. For example, the patent / utility model document may not include the fourth description. In this case, the first part P1 of the prompt is based on the first description, the second part P2 is based on the second description, and the third part P3 is based on the third description.

[0074] In the above embodiment, when the patent utility model document includes a fourth description, the prompt generated by the control unit 22 includes a first part P1 based on the first and fourth descriptions, a second part P2 based on the second and fourth descriptions, and a third part P3 based on the third and fourth descriptions. However, the present disclosure is not limited to this. For example, when the patent utility model document includes a fourth description, the prompt generated by the control unit 22 may include a first part P1 based on the first description, a second part P2 based on the second description, and a third part P3 based on the third and fourth descriptions.

[0075] In the above embodiment, the control unit 22 of the information processing device 20 replaces at least one word included in the first description, the second description, and the third description based on the fourth description, but the present disclosure is not limited to this. For example, the control unit 22 may replace at least one word included in at least one of the first description, the second description, and the third description based on the fourth description.

[0076] In the above embodiment, the control unit 22 replaces at least one word included in the first description, the second description, and the third description based on the fourth description, but, for example, at least one of the first description, the second description, and the third description may be rewritten based on the fourth description. Even in this case, the control unit 22 can generate a prompt including a specific description, so that the information processing device 20 can receive a sentence including a more accurate candidate name from the language model.

[0077] In the above embodiment, the third part P3 is based on the part of the claims corresponding to the first claim (i.e., claim 1), but it may also be based on, for example, the part of the means for solving the problem (i.e., claim 1) or the part of the claims of the utility model registration corresponding to claim 1.

[0078] In the above embodiment, the third statement is a statement described in the first claim (i.e., claim 1) among the claims, but it may also be, for example, a statement described in all claims included in the claims, a statement regarding the scope of a utility model registration claim, or a statement regarding a means for solving a problem.

[0079] In the above embodiment, the third part P3 is based on the part corresponding to the first claim (i.e., claim 1) among the claims, but may also be based on parts corresponding to all claims described in the claims, or may also be based on the part corresponding to the first claim (i.e., claim 1) among the claims and parts corresponding to one or more claims that directly or indirectly cite the first claim.

[0080] In the above embodiment, the sentence generated by the language model 10 includes multiple candidate names, but it may also include only one candidate name.

[0081] In the above embodiment, the sentence generated by the language model 10 includes at least one candidate name for a product related to the invention described in the patent / utility model document, but the present disclosure is not limited thereto. For example, the sentence may include at least one candidate name for a service related to the invention described in the patent / utility model document, or at least one candidate name for a product or service related to the invention described in the patent / utility model document.

[0082] In the first and second embodiments, the specification and claims are input into the input unit 31 of the user terminal device 30, but for example, not only the specification and claims but also an abstract etc. may be input.

[0083] In the first and second embodiments, the specification and claims in Japanese are input into the input unit 31 of the user terminal device 30, but for example, the specification and claims in a language other than Japanese (e.g., English) may also be input.

[0084] In the second and fourth embodiments, the predetermined condition is that the candidate name contains "smart," but it may also be, for example, that the candidate name contains multiple words.

[0085] In the second and fourth embodiments, the predetermined condition is that the candidate name must contain "smart," but a word other than "smart" may also be set.

[0086] In the second and fourth embodiments, the predetermined condition is that the candidate name contains "smart," but the predetermined condition may be any other condition than the inclusion of any word.

[0087] In the third and fourth embodiments, the case-specific information is the patent application number, JP 2024-999999, but it may also be, for example, a utility model registration application number, publication number, patent number, or utility model registration number.

[0088] In the third and fourth embodiments, the external database 40 is a database of the Japan Patent Office, but it may also be, for example, a database of the United States Patent and Trademark Office, a commercial database provided by an intellectual property service provider, etc. For example, if the external database 40 is a database of the United States Patent and Trademark Office, English patent and utility model documents are obtained from the external database 40.

[0089] In the third and fourth embodiments, one item of case-specific information is input to the input unit 31 of the user terminal device 30, but a plurality of items of case-specific information may be input.

[0090] As described above, the present specification discloses the following: (1) a processor; a memory that stores at least one instruction executable by the processor, the information processing device is capable of communicating with a language model; When at least one of the instructions is executed by the processor, the information processing device: Obtain a patent / utility model document including a first description of the problem that the invention / device is intended to solve, a second description of the effect of the invention / device, and a third description of the configuration of the invention / device; generating a prompt for causing the language model to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document based on the patent / utility model document; sending the prompt to the language model; An information processing device that receives, from the language model, the sentence generated by the language model based on the prompt. (2) the prompt includes a first portion based on the first description, a second portion based on the second description, and a third portion based on the third description; The information processing device according to (1), wherein the first part, the second part, and the third part are composed of sentences in bulleted format. (3) The patent / utility model document includes a fourth description of a mode for carrying out the invention / device; the first portion is based on the first description and the fourth description; the second portion is based on the second description and the fourth description; The information processing device according to (2), wherein the third part is based on the third description and the fourth description. (4) An information processing device according to (3), which replaces at least one word contained in at least one of the first description, the second description, and the third description based on the fourth description, or rewrites at least one of the first description, the second description, and the third description based on the fourth description. (5) The third description includes at least one of a description of a means for solving the problem and a description of the scope of the patent / utility model registration claim; An information processing device described in any one of (2) to (4), wherein the third part is based on the means for solving the problem or the part corresponding to the first claim of the scope of the patent / utility model registration claims. (6) Obtain the specified conditions set by the user, The information processing device according to any one of (1) to (5), which generates the prompt based on the patent / utility model document and the predetermined conditions. (7) The information processing device according to (6), wherein the predetermined condition is that the candidate name includes at least one word set by the user. (8) The prompt includes a command based on the predetermined condition and a detailed description based on the patent and utility model document; The information processing device according to (6) or (7), wherein the command statement and the detailed description statement are written separately from each other. (9) An information processing method executed by an information processing device capable of communicating with a language model, comprising: Obtaining a patent / utility model document including a first description of a problem to be solved by the invention / device, a second description of an effect of the invention / device, and a third description of a configuration of the invention / device; generating a prompt based on the patent / utility model document to cause the language model to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document; sending the prompt to the language model; receiving from the language model the sentence generated by the language model based on the prompt. (10) A computer program comprising at least one instruction executed by a processor of an information processing device capable of communicating with a language model, At least one of the instructions, when executed by the processor, causes the information processing device to: Obtain a patent / utility model document including a first description of the problem that the invention / device is intended to solve, a second description of the effect of the invention / device, and a third description of the configuration of the invention / device; generating a prompt based on the patent / utility model document to cause the language model to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document; sending the prompt to the language model; receiving from the language model the sentence generated by the language model based on the prompt. (11) A non-transitory computer-readable medium on which the computer program described in (10) is stored. [Explanation of symbols]

[0091] 10: Language model 20: Information processing device 21: Input section 22: Control unit 23: Output section 24: Bus 30: User terminal device 31: Input section 32: Display section 40: External database 221: Memory 222: Processor

Claims

1. a processor; a memory that stores at least one instruction executable by the processor, the information processing device is capable of communicating with a language model; When at least one of the instructions is executed by the processor, the information processing device: Obtain a patent / utility model document including a first description of the problem that the invention / device is intended to solve, a second description of the effect of the invention / device, and a third description of the configuration of the invention / device; generating a prompt based on the patent / utility model document to cause the language model to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document; sending the prompt to the language model; An information processing device that receives, from the language model, the sentence generated by the language model based on the prompt.

2. the prompt includes a first portion based on the first description, a second portion based on the second description, and a third portion based on the third description; The information processing device according to claim 1 , wherein the first part, the second part, and the third part are composed of sentences in bulleted format.

3. The patent / utility model document includes a fourth description of a mode for carrying out the invention / device; the first portion is based on the first description and the fourth description; the second portion is based on the second description and the fourth description; The information processing device according to claim 2 , wherein the third portion is based on the third description and the fourth description.

4. 4. The information processing device of claim 3, wherein at least one word contained in at least one of the first description, the second description, and the third description is replaced based on the fourth description, or at least one of the first description, the second description, and the third description is rewritten based on the fourth description.

5. the third description includes at least one of a description of a means for solving the problem and a description of the scope of the patent / utility model registration claim; The information processing device according to claim 2 , wherein the third part is based on a part corresponding to the means for solving the problem or a first claim of the scope of claims for patent / utility model registration.

6. Obtain the specified conditions set by the user, The information processing device according to claim 1 or 2, wherein the prompt is generated based on the patent / utility model document and the predetermined condition.

7. The information processing apparatus according to claim 6 , wherein the predetermined condition is that the candidate name includes at least one word set by the user.

8. the prompt includes a command statement based on the predetermined condition and a detailed description based on the patent and utility model document; The information processing device according to claim 6 , wherein the command statement and the detailed explanation are written separately from each other.

9. 1. An information processing method executed by an information processing device capable of communicating with a language model, comprising: Obtaining a patent / utility model document including a first description of a problem to be solved by the invention / device, a second description of an effect of the invention / device, and a third description of a configuration of the invention / device; generating a prompt based on the patent / utility model document to cause the language model to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document; sending the prompt to the language model; receiving from the language model the sentence generated by the language model based on the prompt.

10. 1. A computer program comprising at least one instruction for execution by a processor of an information processing device in communication with a language model, the computer program comprising: At least one of the instructions, when executed by the processor, causes the information processing device to: Obtain a patent / utility model document including a first description of the problem that the invention / device is intended to solve, a second description of the effect of the invention / device, and a third description of the configuration of the invention / device; generating a prompt based on the patent / utility model document to cause the language model to generate a sentence including at least one candidate name for a product or service related to the invention / device described in the patent / utility model document; sending the prompt to the language model; receiving from the language model the sentence generated by the language model based on the prompt.

11. A non-transitory computer readable medium having stored thereon the computer program of claim 10.

Images