Document creation support device and program
By using specially learned and adjusted natural language processing language models to generate auxiliary information, the problem of restriction of auxiliary information in the prior art is solved, and high-quality patent office submission document production is achieved.
Patent Information
- Application Number
- CN202480004155.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-05-09
- Filing Date
- 2024-05-08
- Publication Date
- 2025-05-06
AI Technical Summary
In the prior art, when making a document submitted by the patent office, auxiliary information is limited to the correlation range before and after the correction, and it is difficult to provide high-quality auxiliary information.
By leveraging specially learned and adjusted natural language processing language models, various auxiliary information suitable for user intentions are generated to assist in the production of high-quality patent office filings.
It realizes the provision of various auxiliary information suitable for user intentions, significantly improving the efficiency and quality of producing high-quality patent office filing documents.
Smart Images

Figure CN119948516A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a document production assisting device and program. Background Art
[0002] Regarding technology for assisting in the preparation of documents submitted to a patent office, such as application documents for inventions or utility models, patent document 1 discloses a technology that associates a predetermined statement before correction and a revised statement in a statement of claim 1 that has already been prepared when the document maker later corrects it, stores statements of superordinate concepts or subordinate concepts of the statement before correction as related information in a storage unit, associates and stores a predetermined statement before correction and a revised statement in a statement of claim 2 that has already been prepared when the document maker later corrects it in a storage unit, thereby updating the related information, and outputs related information based on the updated related information for the statement input when preparing claim 3.
[0003] According to the technology of patent document 1, when preparing application documents for an invention or utility model, by outputting relevant information such as sentences of superordinate concepts or subordinate concepts of the sentence before correction as auxiliary information, the document maker can be assisted in preparing the application document by referring to the auxiliary information and using appropriate expressions.
[0004] Prior art literature Patent Literature Patent Document 1: Japanese Patent Application Laid-Open No. 2021-086425 Summary of the Invention
[0005] Problems to be solved by the invention However, the technique in Patent Document 1 uses information generated based on the correlation between the pre-correction and post-correction sentences as auxiliary information. Therefore, the auxiliary information is limited to the correlation between the pre-correction and post-correction sentences. Consequently, since application documents are prepared with reference to this limited auxiliary information, there is room for further improvement in terms of facilitating the production of high-quality application documents.
[0006] The present invention has been made in view of the above circumstances, and an object of the present invention is to easily create high-quality documents to be submitted to the patent office by providing various auxiliary information suitable for the user's intention.
[0007] Means used to solve problems To address the above-mentioned issues, the inventors conducted extensive research and discovered that this objective can be achieved by generating auxiliary information to assist in the preparation of patent office submissions using a language model that has been learned and adapted based on prompt content to specifically prepare patent office submissions for inventions and utility models. Furthermore, the inventors have completed the present invention. Specifically, the present invention provides the following.
[0008] The present invention is a document production auxiliary device, characterized by having: A document database storing document information of inventions and utility models in a standardized unified data format suitable for learning language models for natural language processing; and The auxiliary information generation unit generates auxiliary information for assisting in the preparation of the documents submitted to the patent office by using a language model based on the prompted content. The language model is a model that combines one or more of the natural language processing methods based on the literature database and is learned and adjusted to specifically prepare documents submitted to the patent office for inventions and utility models.
[0009] According to the features of the present invention, by utilizing a language model for natural language processing specifically for producing patent office submission documents for inventions and utility models, and generating auxiliary information corresponding to the prompted content, various auxiliary information suitable for the intention of the user who input the prompted content is provided, thereby making it easy to produce high-quality patent office submission documents.
[0010] Effects of the Invention According to the present invention, by providing various auxiliary information suitable for the user's intention, high-quality patent office filing documents can be easily produced. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Figure 1 This is an explanatory diagram showing the information flow of the document preparation support device according to the present embodiment.
[0012] Figure 2 This is a block diagram showing an example of the hardware configuration and software configuration of the document creation support device according to the present embodiment.
[0013] Figure 3 It is an explanatory diagram showing a display screen of a user interface.
[0014] Figure 4 It is an explanatory diagram showing a display screen of a user interface.
[0015] Figure 5 It is an explanatory diagram showing a display screen of a user interface.
[0016] Figure 6 It is an explanatory diagram showing a display screen of a user interface. DETAILED DESCRIPTION
[0017] Hereinafter, an example of an embodiment of the present invention will be described in detail with reference to the drawings.
[0018] (Document creation assisting device 1) like Figure 1 As shown, the document preparation support device 1 is configured to generate support information for supporting the preparation of documents submitted to the patent office by utilizing a language model of natural language processing used for preparing documents submitted to the patent office specifically for inventions and utility models.
[0019] As a specific example, the document preparation assistance device 1 includes: a document database that stores document storage information of inventions and utility models in a state standardized into a unified data format to adapt to the learning of language models for natural language processing; and an auxiliary information generation unit 115 that uses a language model based on the document database to combine one or more natural language processing methods according to the prompted content and performs learning and adjustment to specifically prepare patent office submission documents for inventions and utility models, thereby generating auxiliary information for assisting in the preparation of patent office submission documents.
[0020] Here, "invention and utility model document information" includes domestic documents such as published invention and utility model publications, patent publications (invention publications, utility model publications), foreign documents such as U.S. patent specifications and international publications, and non-patent documents such as domestic technical journals and pre-draft collections. Furthermore, "document information" also includes opinions, amendments, various trial decision documents (examination decision trials, adversary trials), opposition decision documents, and judgment publications. This document information is collected from various search websites, including J-PlatPat (registered trademark), the European Patent Office's patent information search website Espacenet (registered trademark), and the international patent information search website WIPO PatentScope, as well as the Intellectual Property High Court's trial precedent search website and the internal data of various patent offices.
[0021] "Natural language processing" enables computers to understand text or audio data written in natural language and perform processing appropriate to their purpose. Specifically, examples include morpheme analysis, which assigns information such as word classes by breaking down the smallest units of natural language into "morphemes," syntactic analysis, which analyzes the grammatical structure of natural language to clarify sentence structure and meaning, semantic analysis, which analyzes the meaning of natural language to understand the meaning of words or sentences and make logical judgments or inferences, contextual analysis, which understands natural language by considering the context surrounding sentences, and intention analysis, which extracts the speaker's or writer's intentions from conversations or texts written in natural language. Thus, "natural language processing" combines morpheme analysis, syntactic analysis, semantic analysis, contextual analysis, and intention analysis to process natural language, enabling the generation of auxiliary information to assist in the preparation of documents submitted to the Patent Office in accordance with this embodiment, machine translation, automatic summarization, question-answering systems, and voice recognition.
[0022] A language model is a type of probabilistic model used in natural language processing. It is used to probabilistically predict how likely a given word or sentence is to occur as part of natural language. Specifically, a language model calculates the probability of occurrence of a given word string or sentence, or compares the probability of occurrence of multiple word strings or sentences. When predicting the next word or sentence, the language model automatically generates the most likely word or sentence based on the context.
[0023] “Standardization to a unified data form” means that when processing large amounts of data such as patent documents, information stored in different forms for each document is converted into a common form and organized in a unified manner. For example, patent documents contain various information such as the name of the applicant or inventor, application number or publication number, technical field, document summary, claims, etc., but most of this information is stored in different forms for each country, each document, or each database. Therefore, in order to create a language model for natural language processing, it may be difficult to collect necessary information from multiple databases or process information in different forms, but by standardizing to a unified data form, data retrieval or analysis can be easily achieved. Moreover, when storing information in machine learning documents, the efficiency or accuracy of the processing can be improved by using standardized data as input. In addition, the so-called “standardization” is a processing method that eliminates data redundancy and maintains data integrity or consistency by organizing and structuring the data in the table of the database according to certain rules.
[0024] The "document database" stores document information for inventions and utility models in a standardized, unified data format, making it suitable for learning language models used in natural language processing. Specifically, the document database includes the specification database 121, which stores specification information for inventions and utility models, and the intermediate file database 122, which stores intermediate file information such as opinion information and amendments for inventions and utility models. Thus, even when document information is collected using specifications with different document formats, such as those from Japan and the United States, storing the specification data in standardized formats, such as "Problem Solved" and "Implementation Methods," in the various tables of the specification database 121 reduces data duplication and improves data integrity. Furthermore, by linking the tables in the specification database 121, necessary information can be efficiently retrieved.
[0025] "Patent office filings for inventions and utility models" are documents submitted to the patent office, and examples include specifications, amendments, opinions, requests for trial, and patent oppositions. In the field of natural language processing, a "language model" refers to a mathematical model used to learn language patterns and grammatical rules, thereby generating or understanding natural language. For example, a language model used in the preparation of patent office filings is a language model learned specifically for the preparation of patent office filings. It is used to generate patent application specifications or opinions, which are a type of patent office filing, based on the content provided, or to generate auxiliary information when preparing specifications or opinions.
[0026] The "learning" part of "Language Model Learning and Tuning" includes learning with a teacher (for example, predicting the next word) and learning without a teacher (for example, learning vector representations of words). Furthermore, "tuning" involves evaluating the learned language model and adjusting parameters such as weights as needed, or adjusting hyperparameters such as the number of hidden layers or units in the neural network, the learning rate, and the strength of normalization to improve model performance.
[0027] "Hints" are information that displays information such as content that user 4 should input or operation methods, and instructs user 4 on how to perform these operations. Examples of "support information to assist in the preparation of documents for patent office submission" include suggestions for appropriate wording or expression when entering necessary document information, corrections to grammar or expression, information useful for document preparation such as references to similar patent documents, and information or suggestions regarding the form or format of a document. These hints and support information are described in detail in the user interface described below.
[0028] According to the above structure, a language model of natural language processing is used to specifically prepare patent office submission documents for inventions and utility models, and auxiliary information corresponding to the prompted content is generated, thereby providing various auxiliary information suitable for the intention of user 4 who input the prompted content, so that high-quality patent office submission documents can be easily prepared.
[0029] In addition, if Figure 2 As shown, the document preparation support device 1 includes, in addition to the auxiliary information generation unit 115, an activation unit 111, a data collection unit 112, a model learning unit 113, an input processing unit 114, a terminal control unit 116, and a communication unit 13. Part or all of the auxiliary information generation unit 115, the activation unit 111, the data collection unit 112, the model learning unit 113, the input processing unit 114, and the terminal control unit 116 may be configured as either hardware or software and included in the control unit 11, which is a computer.
[0030] The communication unit 13 is connected to the user terminal 2 operated by the user 4 via the Internet 5, enabling data communication. Data communication between the communication unit 13 and the user terminal 2 is not limited to the Internet 5; it can also use a dedicated line, an information communication network such as a LAN (Local Area Network), or Bluetooth (a registered trademark), a wireless communication standard for short-range communication. Furthermore, to prevent the leakage of highly confidential documents filed with patent offices, data communication between the communication unit 13 and the user terminal 2 can also be performed via a dedicated line. Furthermore, when data communication between the communication unit 13 and the user terminal 2 is performed via the Internet 5, it is preferable that a VPN (Virtual Private Network) function be provided. When VPN functionality is incorporated into the communication unit 13 and the user terminal 2, data communications from the user terminal 2 are routed through the VPN, thereby preventing unauthorized external access and ensuring secure communication. The user terminal 2 is a typical information processing device such as a personal computer, laptop computer, smartphone, or tablet terminal.
[0031] The activation unit 111 is capable of communicating data only with authenticated user terminals 2. Specifically, the activation unit 111 verifies whether the activation signal transmitted from a specific user terminal 2 matches pre-set authentication information. Examples of authentication methods include those combining one or more authentication elements, such as password authentication, biometric authentication, one-time password authentication, and smart card authentication using an embedded IC chip. This allows the document preparation support device 1 to communicate data only with authenticated user terminals 2, making it difficult to leak highly confidential documents submitted to patent offices.
[0032] The data collection unit 112 collects document information on inventions and utility models from patent information search sites 6 and other websites, standardizes the collected information into a unified data format, and stores it in a document database. This allows the data collection unit 112 to collect and organize the data required for learning or adjusting the language model and store it in the document database. This enables the auxiliary information generation unit 115 to efficiently and accurately assist in the preparation of documents submitted to the patent office. Alternatively, the collection of document information can be performed by the administrator of the document preparation assistance device 1 using the administrator terminal 7 to access the patent information search site 6 and other websites.
[0033] The model learning unit 113 has the function of learning and adjusting the language model used for natural language processing based on the invention and utility model literature database. Thus, by learning and adjusting the language model to become a language model specifically designed for preparing documents submitted to the patent office, the model learning unit 113 can perform more advanced natural language processing and generate high-quality auxiliary information for the auxiliary information generation unit 115 to assist in preparing documents submitted to the patent office.
[0034] The input processing unit 114 has the function of converting prompt information indicating the content of a prompt transmitted from the user terminal 2 into information that is easily processed by the language model for natural language processing. The terminal control unit 116 has the function of setting the display screen of the user terminal 2 to a screen suitable for preparing documents for submission to the Patent Office based on data from the UI database 124 (user interface database 124). Details will be described later.
[0035] In addition, the document preparation assistance device 1 has an input device 15 connected to the input receiving unit 118 and a display device 14 connected to the display control unit 117. Examples of the input device 15 include a keyboard, a mouse, a touch panel, a voice input device, and the like. Examples of the display device 14 include a liquid crystal display device, and the like. Thus, the document preparation assistance device 1 can be composed of an information processing device such as a general personal computer, a laptop computer, a smartphone, or a tablet terminal. In addition, the document preparation assistance device 1 may also lack at least one of the input device 15 and the display device 14. In this case, since the input device 15 and the display device 14 are provided on an external terminal not shown, the document preparation assistance device 1 can be used as a document preparation assistance server. In addition, the document preparation assistance device 1 can also be used as a user terminal 2, and the user 4 can prepare documents to be submitted to the patent office in the document preparation assistance device 1 while receiving assistance information.
[0036] In addition, in this embodiment, the functions of the document preparation support device 1 are described as being installed in an information processing device. However, the present invention is not limited to this. The functions of the document preparation support device 1 can also be installed in cloud computing. In this case, cloud computing can be used to add computer resources as needed, thereby processing large amounts of document data, enabling faster and more efficient processing. Even if the number of users 4 increases significantly, processing capacity can be easily expanded to cope with it.
[0037] (Storage unit 12, etc.) like Figure 1 As shown, the instruction manual database 121, the intermediate document database 122 storing documents related to the review decisions such as opinions or amendments, and the UI database 124 are stored in the storage unit 12. The storage unit 12 is connected to the document production assistance device 1 in a manner that enables data communication. The storage unit 12 can be composed of a hard disk or a combination of a hard disk and a memory. In the case of a structure in which the hard disk and the memory are combined, part of the data or index used by the database is cached in the memory as needed, thereby speeding up access to the database. In addition, the storage unit 12 also stores a user database 123, which stores various information including identification information for determining the user 4 or authentication information for authenticating the user 4. The activation unit 111 performs authentication processing and other processing based on the data in the user database 123.
[0038] Furthermore, the storage unit 12 may be a data server connected to an information communication network such as the Internet 5, separate from the document creation support apparatus 1. Furthermore, the storage unit 12 may be composed of a plurality of data servers provided for each database.
[0039] (user interface) like Figure 3 As shown, the terminal control unit 116 has a function for setting the display screen of the user terminal 2 to a screen suitable for preparing documents for submission to the patent office based on the UI database 124. Specifically, the terminal control unit 116 has a function for controlling the display screen 21 of the user terminal 2 to display a prompt section 212 for receiving prompt information, an auxiliary information section 213 for displaying auxiliary information generated by the auxiliary information generation section 115, and a document generation section 211, which serves as an area for preparing documents for submission to the patent office. This allows the user 4 using the user terminal 2 to prepare documents for submission to the patent office while simultaneously checking the contents of the prompt section 212 and the auxiliary information in the auxiliary information section 213, making it easier to prepare documents for submission to the patent office. Furthermore, at the start of a document for submission to the patent office, such as a specification, the user terminal 2 can display document generation section 211 with items conforming to a standard format.
[0040] In addition, if Figure 4 As shown, the terminal control unit 116 has the following function: to control so that a prompt operation unit 214 is displayed on the display screen 21 of the user terminal 2, and the prompt operation unit 214 enables the user to specify fixed words such as similar words, antonyms, functions, etc. that represent predetermined indication content, and to input the information composed of the specified fixed words and the sentence specified by the file production unit 211 as content to the prompt unit 212.
[0041] As a specific example, if the "predetermined instruction content" includes multiple instruction elements such as similar words, superordinate concepts, antonyms, and functions, the fixed phrases containing these instruction elements are displayed as buttons, and a first instruction button 2141 for the fixed phrase with the similar word as the instruction content, a second instruction button 2142 for the fixed phrase with the superordinate concept as the instruction content, a third instruction button 2143 for the fixed phrase with the antonym as the instruction content, and a fourth instruction button 2144 for the fixed phrase with the function as the instruction content are displayed in the prompt operation unit 214. For example, if the sentence "A spring is provided between the top plate and the base" is described in the file creation unit 211, if the user 4 drags the phrase "spring" to the first instruction button 2141 while specifying it, or if the user 4 clicks the first instruction button 2141 while specifying it, the phrase "spring" is stored in the memory of the user terminal 2 and combined with the prompt data for the fixed phrase associated with the first instruction button 2141, which is pre-stored in the memory. Then, the synthesized data is displayed on the presentation unit 212 of the user terminal 2 .
[0042] As a result, prompt section 212 displays a message combining a sentence and a fixed phrase, "Please give examples of words similar to rubber," as prompt content. This prompt content is sent to document preparation assistance device 1, where auxiliary information generation section 115 generates "[Examples of similar words] spring, coil, damper" as auxiliary information. By sending this auxiliary information to user terminal 2 and displaying it on auxiliary information section 213, user 4 can continue writing the specification using examples of springs while visually viewing or copying and pasting the auxiliary information from auxiliary information section 213. Thus, user 4 using user terminal 2 can input information combining predetermined instructions and designated phrases into prompt section 212 when preparing documents for submission to the patent office, making it easier to prepare documents for submission to the patent office.
[0043] In addition, if Figure 5 and Figure 6As shown, the terminal control unit 116 has the following function: to control so that a prompt operation unit 214 is displayed in the display screen 21, and the prompt operation unit 214 enables the user to specify a fixed phrase representing a predetermined instruction content, and to input the specified fixed phrase and the information consisting of a sentence segment containing the sentence specified by the file production unit 211 as content to be input into the prompt unit 212.
[0044] As a specific example, if the sentence "A spring is provided between the top plate and the base" is described in the file creation unit 211, when user 4 drags the sentence "spring" to the fourth instruction button 2144 while specifying the sentence "spring," or when user 4 clicks the fourth instruction button 2144 while specifying the sentence "spring," the sentence containing the sentence "spring" is stored in the memory of the user terminal 2 and combined with the pre-stored prompt data for the fixed phrase associated with the fourth instruction button 2144. The combined data is then displayed on the prompt unit 212 of the user terminal 2.
[0045] As a result, the sentence "A spring is provided between the top plate and the base" and the sentence "Please give examples of words similar to 'rubber'" are displayed as prompt content in prompt section 212. This prompt content is transmitted to document preparation assistance device 1, and auxiliary information generation section 115 generates auxiliary information such as "A spring is provided between the top plate and the base." The spring's function is to generate tension between the top plate and the base, supporting the top plate. The spring can be deformed, such as by compression or extension, and its elastic force exerts pressure on the top plate..." This auxiliary information is transmitted to user terminal 2 and displayed on auxiliary information section 213. User 4 can continue writing the specification while visually viewing or copying and pasting the auxiliary information in auxiliary information section 213, while acting as a spring. Thus, when preparing documents for submission to the patent office, information combining predetermined instructions and a sentence containing a designated phrase is input to prompt section 212. This allows for appropriate responses based on the context of the sentence, making it easier to prepare documents for submission to the patent office.
[0046] Furthermore, the terminal control unit 116 may have a function of making the user interface of the user terminal 2 have the following configurations 1 to 6.
[0047] (Structure 1) While the user continues to write an article in the document creation unit 211, the moment a sentence, section, or paragraph is completed can be automatically detected. At these moments, the text information of the sentence, section, or paragraph is transmitted to the prompt unit 212. The document creation assistance device 1 infers the article following the text information and displays it as auxiliary information in the auxiliary information unit 213. At this time, the portion of the document creation unit 211 corresponding to the text information transmitted to the prompt unit 212 can also be highlighted. In this case, the user 4 can easily determine which sentence, section, or paragraph the auxiliary information corresponds to. Alternatively, instead of automatically detecting the moment a sentence, section, or paragraph is completed, a fifth indicator button can be provided, and when the fifth indicator button is clicked, the auxiliary information of the article inferred to follow the sentence, section, or paragraph is displayed.
[0048] (Structure 2) The document preparation unit 211 may also include a sixth instruction button for requesting advice on the preparation of a specification for an invention or utility model based on input invention information, such as an outline of the invention or invention documents. For example, when negotiation documents are input and the sixth instruction button is clicked, the document preparation assistance device 1 extracts and summarizes important points from the documents and displays them as supplementary information in the supplementary information unit 213. Furthermore, when the sixth instruction button is clicked, for example, the prompt unit 212 displays "Please read the following document 1. We will inform you of points we believe those skilled in the art are not aware of. (Line break x2) (Main body of document 1)" and the supplementary information unit 213 displays supplementary information that skips two lines for important points with potential novelty. Subsequently, when the sixth instruction button is clicked, the prompt unit 212 displays "Please read the following document 1. We will inform you of points we believe those skilled in the art are familiar with. (Line break x2) (Main body of document 1)" and the supplementary information unit 213 displays supplementary information that skips two lines for important points with potential inventiveness.
[0049] Furthermore, when user 4 considers inputting an inventive point while referring to the auxiliary information (important points) and subsequently clicks the sixth instruction button, the document preparation assistance device 1 extracts prior art documents from the specification database 121, identifies components that may possess novelty and inventiveness, and components suspected of being prior art, and displays each component as auxiliary information. Furthermore, in structure 2, to improve operability for user 4, a user interface is provided that allows the novelty extraction process, the inventiveness extraction process, and the component evaluation process to be performed sequentially by continuously clicking the sixth instruction button. However, this is not limited to this, and instruction buttons may also be provided for each evaluation process. Furthermore, when the sixth instruction button is clicked, multiple prompts indicating the novelty extraction process, the inventiveness extraction process, and the component evaluation process may be selectively displayed on the prompt section 212.
[0050] (Structure 3) A seventh instruction button for creating a claim scheme may also be provided in the document creation section 211. For example, when an invention point is entered and the seventh instruction button is subsequently clicked, a claim scheme is generated in the document creation assistance device 1 based on the specification database 121 and displayed as auxiliary information. Furthermore, when a claim scheme is entered and the seventh instruction button is clicked, the overall structure of the specification may also be displayed as auxiliary information. Specifically, when a claim scheme is entered and the seventh instruction button is clicked, the prompt section 212 displays "Please indicate the overall structure of the specification corresponding to the following claims. (Line break x2) (Claims)" and the auxiliary information section 213 displays auxiliary information indicating the overall structure of the specification. Furthermore, when the seventh instruction button is clicked, the prompt section 212 displays "Please indicate examples of the following components in the implementation. If there are any references to materials, please also inform me. (Line break x2) (Components indicated in the overall structure)" and the auxiliary information section 213 displays examples of the embodiment and references to the materials.
[0051] In addition, after the user terminal 2 creates a claim scheme, when the seventh indication button is clicked, "Please check whether the following claims are in a form containing symbols. (Line break x2) (Claims)" can be displayed on the prompt section 212, and the check result can be displayed on the auxiliary information section 213. Alternatively, "For each term in the following claims, please list candidates for sentences that become superordinate concepts. (Line break x2) (Claims)" can be displayed on the prompt section 212, and candidates for sentences that become superordinate concepts can be displayed on the auxiliary information section 213. In addition, instead of the seventh indication button, different indication buttons can be set according to the content of each prompt. In addition, when the seventh button is clicked, the content of multiple prompts can be selectively displayed.
[0052] (Structure 4) The document preparation unit 211 may also include an eighth instruction button for checking the manual's format, grammar, and other aspects, as well as for inconsistencies in terminology and symbols. For example, during or after the manual is prepared, clicking the eighth instruction button causes the document preparation support device 1 to perform a check on the manual's format, grammar, and other aspects, as well as for inconsistencies in terminology and symbols, with the results displayed as auxiliary information. Furthermore, corrective measures for various inspection items may also be displayed as auxiliary information.
[0053] (Structure 5) The document preparation section 211 may also include a ninth instruction button for checking whether a proposed correction or opinion can eliminate the Office's opinion (e.g., whether all Office's opinions have been addressed). Specifically, clicking the ninth instruction button displays "Please summarize and memorize the following Office's Action. (Line break x2) (Office's Action)" and "Please check whether the proposed opinion below can eliminate the Office's opinion notified in the Office's Action just memorized. Please also indicate your proposed correction. (Line break x2) (Office's Action)" on the prompt section 212, and displays the check results and proposed correction on the auxiliary information section 213.
[0054] (Structure 6) In the document production unit 211, a tenth instruction button can be set for generating a flowchart or block diagram or a data table scheme corresponding to the content of the input claim or a part of the description, and a tenth instruction button can be set for generating a claim scheme or a description scheme corresponding to the content of the flowchart or block diagram or a data table.
[0055] (Document creation assistance program) The activation unit 111, data collection unit 112, model learning unit 113, input processing unit 114, auxiliary information generation unit 115, terminal control unit 116, display control unit 117, and input reception unit 118 can be implemented by either hardware or software. Each of these units 111-118 constitutes at least a portion of the control unit 11. When each of these units 111-118 is implemented by software, the computer serving as the control unit 11 executes a document preparation support program. This program generates auxiliary information to assist in the preparation of documents submitted to the patent office by utilizing a language model for natural language processing, which is used to prepare documents specifically for inventions and utility models submitted to the patent office.
[0056] As a specific example, a document preparation support program causes a computer, i.e., a user terminal 2, which is a terminal device capable of accessing a document preparation support device 1, to execute a display processing step (S1). The document preparation support device 1 includes a document database storing information on documents related to inventions and utility models in a standardized data format suitable for learning a language model for natural language processing; and an auxiliary information generation unit 115 that generates auxiliary information for assisting in the preparation of patent office submission documents based on presented content using a language model that combines, learns, and adjusts one or more natural language processing methods based on the document database to specifically prepare documents for patent office submissions related to inventions and utility models. The display processing step (S1) causes the user terminal 2 to display a presentation unit 212 for receiving presented content, an auxiliary information unit 213 for displaying the auxiliary information generated by the auxiliary information generation unit 115, a document preparation unit 211 serving as a preparation area for patent office submission documents, and a presentation operation unit 214 for allowing a user 4 to specify a fixed phrase representing predetermined instruction content. According to the program, a user can prepare a patent office submission document while simultaneously confirming the content of the presentation unit and various auxiliary information in the auxiliary unit.
[0057] Furthermore, the document preparation support program may include a configuration that causes the computer of the user terminal 2 to execute a prompting operation step (S2), wherein, when a fixed word or phrase is specified in the prompting operation unit 214 while a phrase is specified in the document preparation unit 211, information combining the specified fixed word or phrase and the phrase specified by the document preparation unit 211 is input to the prompting unit 212. According to this program, in the preparation of documents for submission to the patent office, information combining the predetermined instruction content and the specified phrase is input to the prompting unit 212, thereby facilitating the preparation of high-quality documents for submission to the patent office.
[0058] Furthermore, the document creation support program may include a configuration that causes the computer of the user terminal 2 to execute processing steps for realizing various functions of the terminal control unit 116 .
[0059] The document preparation assistance program constructed as described above can assist user 4 in preparing documents for patent office submission in various forms with high accuracy, thereby enabling rapid and high-quality preparation of documents for patent office submission. Furthermore, simply by installing the program on an information processing device such as a personal computer or tablet terminal, the information processing device can function as document preparation assistance device 1. Furthermore, the program can be distributed in a state recorded on a recording medium such as a CDROM or USB memory, or distributed via a two-way communication network or communication line such as the Internet or a one-way communication network such as television broadcasting.
[0060] (Document creation assistance method) Some or all steps in the document preparation assistance program can also be implemented in hardware. Specifically, a user terminal 2 capable of accessing the document preparation assistance device 1 only needs to be configured to execute the document preparation assistance method. Specifically, the document preparation assistance method is executed by the user terminal 2 capable of accessing the document preparation assistance device 1 and includes a display processing step (S1). The document preparation assistance device 1 includes a document database storing information on invention and utility model documents in a standardized data format suitable for learning a language model for natural language processing; and an auxiliary information generation unit 115 that generates auxiliary information for assisting in the preparation of patent office submission documents based on the content of a prompt, utilizing a language model specifically learned and adjusted by combining one or more natural language processing methods based on the document database to generate patent office submission documents for inventions and utility models. Furthermore, the document preparation assistance method may include both a prompt operation step (S2) and processing steps for implementing the various functions of the terminal control unit 116.
[0061] While the above description uses a document preparation assistance device 1 that generates auxiliary information using a language model for natural language processing used specifically for preparing patent office filings for inventions and utility models, the present invention is not limited to this embodiment. Specifically, the terminal control unit 116 for configuring the user interface may be provided within the document preparation assistance device, which may generate auxiliary information using a language model for general natural language processing, such as Chat GPT. Alternatively, a document preparation assistance program or method may be applied to a user terminal 2 that has access to such a document preparation assistance device.
[0062] Specifically, a document creation assistance device capable of communicating data with an information source such as a website or database via the Internet 5 and generating auxiliary information based on the content of a prompt using a language model for general natural language processing may also include a terminal control unit 116 that sets the user interface of the user terminal 2. Furthermore, a document creation assistance program may be a program that causes the computer of the user terminal 2 to execute the various processing steps of this embodiment, and the user terminal 2 may be capable of communicating data with the document creation assistance device that generates auxiliary information using a language model for general natural language processing. A document creation assistance method may also be a method executed by the user terminal 2, and the user terminal 2 may be capable of communicating data with the document creation assistance device that generates auxiliary information using a language model for general natural language processing.
[0063] In addition, within the scope of the present invention, those skilled in the art can conceive of various variations and modifications. Therefore, these variations and modifications are understood to fall within the scope of the present invention. For example, with respect to the above-mentioned embodiment, those skilled in the art appropriately add, delete, or change the design of the constituent elements, or add, omit, or change the conditions of the process, as long as they possess the gist of the present invention, they are included within the scope of the present invention.
[0064] Description of Reference Numerals 1. Document production auxiliary device 11Control Unit 111 Activation Department 112 Data Collection Department 113 Model Learning Department 114 Input Processing Unit 115 Auxiliary Information Generation Unit 116 Terminal Control Department 117 Display control unit 118 Input receiving unit 12 Storage unit 121 Instruction Manual Database 122 intermediate file database 123 user database 124UI Database 13Ministry of Communications 14 Display device 15 Input devices 2 User Terminal 21 Display screen 211 Document Production Department 212 Tips Department 213 Auxiliary Information Department 214 prompt operation department 4 users 5. Internet 6 Patent Information Retrieval Sites 7. Administrator terminal.
Claims
1. A document production auxiliary device, characterized in that: have: A document database stores document storage information of inventions and utility models in a standardized unified data form to be suitable for learning language models for natural language processing; as well as The auxiliary information generation unit generates auxiliary information for assisting in the preparation of the documents submitted to the patent office by using a language model based on the prompted content. The language model is a model that combines one or more of the natural language processing methods based on the literature database and is learned and adjusted to specifically prepare documents submitted to the patent office for inventions and utility models.
2. The document preparation auxiliary device according to claim 1, characterized in that: The document production assistance device includes a terminal control unit, which controls a user terminal to display a prompt unit for accepting the prompt content, an auxiliary information unit that displays the auxiliary information generated in the auxiliary information generation unit, and a document production unit that serves as a production area for submitting documents to the patent office.
3. The document preparation auxiliary device according to claim 2, characterized in that: The terminal control unit controls to display a prompt operation unit in the user terminal, and the prompt operation unit enables the user to specify a fixed phrase representing a predetermined instruction content, and inputs information consisting of the specified fixed phrase and the sentence specified by the file creation unit as content to the prompt unit.
4. The document production assisting device according to claim 2, characterized in that: The terminal control unit controls so that a prompt operation unit is displayed in the user terminal, and the prompt operation unit enables the user to specify a fixed phrase representing a predetermined instruction content, and uses the information composed of the specified fixed phrase and a sentence segment containing the sentence specified by the file creation unit as content input to the prompt unit.
5. A program, characterized in that The computer of the terminal device capable of accessing the document production assisting device executes the following steps: A display processing step of causing the terminal device to display a prompting section for receiving the content of the prompt, an auxiliary information section for displaying the auxiliary information generated in the auxiliary information generating section, and a document generating section as a generating area for submitting documents to the patent office; as well as a prompt operation step, in which, when the fixed word is specified in the prompt operation section while the sentence is specified in the file creation section, information consisting of the specified fixed word and the sentence specified by the file creation section is input as content to the prompt section, Wherein, the document production auxiliary device comprises: A document database stores document storage information of inventions and utility models in a standardized unified data form to be suitable for learning language models for natural language processing; as well as The auxiliary information generation unit generates auxiliary information for assisting in the preparation of the documents submitted to the patent office by using a language model based on the prompted content. The language model is a model that combines one or more of the natural language processing methods based on the literature database and is learned and adjusted to specifically prepare documents submitted to the patent office for inventions and utility models.
Citation Information
Patent Citations
Method of supporting creation of application document for patent, program, and information processing system
JP2021086425A