system

Abstract

We provide the system. [Solution] Means for receiving patent specifications, A means for analyzing the aforementioned patent specification and extracting key sections, A means of searching for similar patents from existing patent databases, Means for evaluating novelty and inventiveness, A means of determining whether or not to grant special permission based on the evaluation results, A means for notifying the user of the evaluation results, A system that includes this.

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Existing patent-related operations, such as patent applications, patent infringement confirmation, patent clustering, and real-time patent monitoring, are very costly and time-consuming. Therefore, it is difficult to conduct appropriate patent management, especially in fields with rapid technological evolution, and there is often a risk of patent infringement. The purpose of the present invention is to solve these problems and provide a system for efficiently performing patent-related operations.

Means for Solving the Problems

[0005] The present invention provides a system that includes means for receiving and analyzing patent specifications to extract key sections and searching for similar patents in an existing patent database. It also includes means for evaluating novelty and inventiveness, determining whether or not to grant a patent based on the evaluation results, and notifying the user of the evaluation results.

[0006] Furthermore, it includes a patent clustering function that includes means for collecting patent data, extracting features, and clustering them, as well as means for labeling the clusters and visualizing the results. It also includes a patent infringement verification function that receives product specifications, analyzes them, compares them with a patent database, assesses patent infringement risk, generates a report of the results, and notifies the user.

[0007] In addition, it includes a patent monitoring function that accepts settings for monitoring targets, monitors the patent database in real time, evaluates and notifies of newly published patents, and continuously monitors the patent database. Through these means, it achieves increased efficiency in patent-related work and a reduction in the risk of patent infringement.

[0008] A "patent specification" is a document that describes the details of an invention and is submitted when applying for a patent.

[0009] "Reception" refers to the process by which a system acquires information from an external source, specifically the electronic acquisition of data.

[0010] "Analysis" is the process of examining data and information in detail and extracting its intended meaning.

[0011] A "key section" refers to the important parts of a patent specification that contain information necessary for patent examination.

[0012] A "patent database" is a data store that stores information about existing patents and is a system that allows users to search for patent information.

[0013] A "similar patent" refers to a patent that already exists and contains similar technical content to a new patent application.

[0014] "Novelty" refers to the fact that an invention is not included in the prior art and is publicly disclosed for the first time.

[0015] "Inventive step" refers to the fact that an invention is technically advanced compared to the prior art and is one of the requirements for patentability.

[0016] "Evaluation result" refers to the conclusion or judgment obtained through analysis and evaluation processes.

[0017] "User" refers to a person or organization that uses a system.

[0018] "Notification" is a process by which a system conveys information to a user, and is carried out through email or a dashboard.

[0019] "Collection" is a process of systematically collecting specific information.

[0020] "Feature" refers to the unique properties or functions of a patent or product, and is obtained through analysis and extraction. [[ID=二十九]]

[0021] "Clustering" is a process of grouping data based on specific criteria.

[0022] "Labeling" is a process of assigning an appropriate label (name) indicating the content to each cluster.

[0023] "Visualization" is a process of presenting data or information in a visually understandable form such as graphs or charts. <了

[0024] 完 "Product specification" is a document that describes the technical details and functions of a product.

[0025] "Verification" is a process of comparing different datasets to confirm matches or differences. <000009]1>

[0026] It should be noted that there seems to be an error in the tag "[[ID=二十九]]" in the original text. It may be a miswriting. I have translated it as it is for now. If this is not what you intended, please correct it and I will re-translate it accurately."Patent infringement risk" refers to the possibility that a certain product or technology infringes on the patents of others.

[0027] "Report" refers to a document that summarizes evaluation results and analysis results.

[0028] "Monitoring target" refers to a specific technical field or keyword that the system continuously monitors.

[0029] "Real-time" refers to the state where processing and analysis are performed with almost no time lag.

[0030] "Monitoring" is a process of continuously paying attention to specific information or data and checking for changes and new information.

Brief Explanation of Drawings

[0031] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0032] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0033] First, let's explain the terminology used in the following explanation.

[0034] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0035] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0036] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0037] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0038] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0039] [First Embodiment]

[0040] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0041] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0042] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0043] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0044] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0045] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0046] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0047] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0048] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0049] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0050] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0051] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0052] This invention is an advanced AI system for efficiently performing patent-related tasks. The system operates as follows:

[0053] 1. Automated patent determination

[0054] Subject: Server

[0055] The server receives the patent specification uploaded by the user. The received specification is analyzed by an NLP (Natural Language Processing) engine, and key sections (such as the invention title, background, purpose, and detailed description) are extracted. The server searches the patent database for similar existing patents and evaluates novelty and inventiveness. Based on the evaluation results, it determines the likelihood of patent granting and notifies the user of the result in a detailed report.

[0056] Specific example:

[0057] A user uploads a specification document for a new invention to the system. This invention includes a new type of battery technology. The server analyzes the specification document and searches for similar existing patents. As a result, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[0058] 2. Patent Clustering

[0059] Subject: Server

[0060] The server collects patent data related to a specific technical field. Next, it extracts key features (keywords and section summaries) from each patent and classifies them using a clustering algorithm. This groups patents based on specific technical fields or application areas. The clustering results are presented to the user visually.

[0061] Specific example:

[0062] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it more efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them to the user.

[0063] 3. Confirmation of patent infringement

[0064] Subject: Server

[0065] The server receives product specifications uploaded by the user and analyzes them. Based on the analysis results, it compares them with a patent database and assesses the potential patent infringement risk. The assessment results are generated in report format and notified to the user.

[0066] Specific example:

[0067] If a company developing a new smartphone wants to check whether its product infringes on another company's patents, the server analyzes the product specifications and compares them with existing patents. Based on the results, it determines if certain features pose a risk of infringing another company's patents. This result is then notified to the user, prompting them to take appropriate action.

[0068] 4. Real-time patent monitoring

[0069] Subject: Server

[0070] Users can configure the server with the technology fields and keywords they wish to monitor. Based on these settings, the server monitors the public patent database in real time. It evaluates newly published patent information, identifies patents of importance to the user, and notifies them. This process is continuous, ensuring users always stay up-to-date.

[0071] Specific example:

[0072] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[0073] This system streamlines the processes of patent application, patent infringement verification, patent clustering, and real-time patent monitoring, significantly improving a company's patent management operations.

[0074] The following describes the processing flow.

[0075] 1. Automated patent determination

[0076] Step 1:

[0077] The server receives patent specifications uploaded by the user to the system.

[0078] Step 2:

[0079] The server uses an NLP engine to perform text analysis on the received patent specification and extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[0080] Step 3:

[0081] The server accesses the patent database and searches for similar existing patents based on the content of the patent specifications.

[0082] Step 4:

[0083] The server uses an AI model to evaluate the novelty and inventiveness of the patent specification. This includes comparing the extracted features with similar patents.

[0084] Step 5:

[0085] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report of the results.

[0086] Step 6:

[0087] The server notifies the user of the generated evaluation results report. The user can view the report via a dashboard or email.

[0088] 2. Patent Clustering

[0089] Step 1:

[0090] The server collects patent data related to a specific technology field from a database.

[0091] Step 2:

[0092] The server uses an AI algorithm to extract key features (keywords and section summaries) from each patent.

[0093] Step 3:

[0094] The server clusters the patent data using a clustering algorithm (e.g., k-means clustering) based on the extracted features.

[0095] Step 4:

[0096] The server automatically assigns appropriate labels to each cluster. These labels include those based on specific technologies and functions.

[0097] Step 5:

[0098] The server visually displays clustering results on a dashboard, allowing users to easily understand and manage their patent portfolio.

[0099] 3. Confirmation of patent infringement

[0100] Step 1:

[0101] The server receives product specifications uploaded by the user to the system.

[0102] Step 2:

[0103] The server uses an NLP engine to perform text analysis on the received product specifications and extracts key functions and features.

[0104] Step 3:

[0105] The server compares the product specifications with the patent database to identify similar patents and potentially infringing patents.

[0106] Step 4:

[0107] The server uses an AI model to assess the risk of patent infringement and assigns a risk score to each patent.

[0108] Step 5:

[0109] The server generates a patent infringement risk report based on the evaluation results.

[0110] Step 6:

[0111] The server notifies the user of the generated report, enabling the user to review the risks and take appropriate action.

[0112] 4. Real-time patent monitoring

[0113] Step 1:

[0114] The user configures the server with the technical fields and keywords to be monitored.

[0115] Step 2:

[0116] The server monitors the patent database in real time based on configured conditions and collects newly published patent information.

[0117] Step 3:

[0118] The server analyzes the collected patent information, compares it against the set conditions, and identifies important patents.

[0119] Step 4:

[0120] The server will notify users of newly identified patents. Available notification methods include email and dashboard notifications.

[0121] Step 5:

[0122] The server continuously monitors the patent database, ensuring that users always have access to the latest patent information.

[0123] By following the steps outlined above, this system efficiently manages patent-related tasks, reducing the risk of patent infringement and enabling rapid access to patent information.

[0124] (Example 1)

[0125] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0126] Traditional patent-related work, including evaluating patent applications, analyzing patent clusters, and confirming patent infringement risks, was manual, time-consuming, labor-intensive, and lacked accuracy and consistency. Furthermore, evaluating novelty and inventiveness required specialized knowledge, making it difficult for companies to manage patents efficiently. In addition, real-time monitoring of patent information and access to the latest updates were insufficient.

[0127] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0128] In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for analyzing the specifications using a natural language processing engine, means for evaluating novelty and inventiveness using a machine learning model, and means for notifying the user of the generated report via a user interface or email. This enables the automation and streamlining of the patent application process.

[0129] The server also includes means for collecting patent data, means for extracting the features of the patents, means for clustering the patents, means for labeling the clusters, means for visualizing the clustering results, means for classifying the patents using a clustering algorithm, and means for providing the visualized clustering results through a user interface. This facilitates the organization and analysis of patent portfolios, enabling the formulation of patent strategies and appropriate decision-making.

[0130] Furthermore, the server includes means for receiving product specifications, means for analyzing the product specifications, means for comparing the product specifications with a patent database, means for evaluating patent infringement risk using a generated AI model, means for generating the evaluation results as a report, means for notifying the user of the evaluation results, and means for evaluating patent grantability based on customizable evaluation criteria. This enables rapid evaluation of patent infringement risk and appropriate response during product development.

[0131] A "patent specification" is a document submitted when filing a patent application, and it describes the details of the invention.

[0132] A "patent database" is a database that stores existing patent information and is used for searching and matching patents.

[0133] A "natural language processing engine" is a software module that provides technology for processing and analyzing human language using computers.

[0134] A "machine learning model" is an algorithm that learns from past data and uses that knowledge to make predictions and classifications.

[0135] A "user interface" is an interface through which a user interacts with a computer system, and includes graphical displays and methods of operation.

[0136] "Patent clustering" is a method of grouping and classifying related patents.

[0137] A "clustering algorithm" is an algorithm that automatically classifies data into meaningful groups (clusters).

[0138] A "generative AI model" is a model that allows artificial intelligence (AI) to automatically generate and analyze data, and it includes pre-trained algorithms that are specialized for a particular purpose.

[0139] "Evaluation criteria" refer to the standards and indicators used when evaluating patents or products.

[0140] A "report" is a document that details the results of analysis and evaluation.

[0141] "Patent infringement risk" refers to the possibility that a product or technology infringes on existing patent rights.

[0142] This invention is an AI system for efficiently performing patent-related tasks. This system provides functions for automatic patent identification, patent clustering, patent infringement verification, and real-time patent monitoring.

[0143] Patent automated determination

[0144] In the automated patent evaluation function, the user uploads a patent specification, and the server receives it. The server uses an NLP (Natural Language Processing) engine to analyze the patent specification and extract key sections (such as the invention title, background, purpose, and detailed description). Next, the server compares it with an existing patent database to evaluate the novelty and inventiveness of the patent. The evaluation results are notified to the user as an automatically generated report. This allows the user to check the likelihood of patent granting with high accuracy.

[0145] Specific example:

[0146] When a user uploads a patent specification for a new battery technology to the system, the server analyzes the specification, compares it with existing patents, and evaluates the novelty and inventiveness of the invention. The evaluation results are then notified to the user in the form of a report.

[0147] Example of a prompt:

[0148] "Please evaluate this patent specification from the perspectives of novelty and inventive step."

[0149] "Please analyze the patent specifications for the new battery technology and create a report evaluating its patentability."

[0150] Patent clustering

[0151] In the patent clustering function, the server collects patent data related to a specific technical field. The server extracts the characteristics of each patent (keywords and section summaries) and classifies the patents using a clustering algorithm. The clustering results are visualized, allowing users to efficiently manage their patent portfolios.

[0152] Specific example:

[0153] A company with a large amount of patent data related to electric vehicle technology uses the system to organize its patents into clusters. The server uses a clustering algorithm to create clusters such as battery technology, motor control, and charging infrastructure, and provides them in a visualized format on the user interface.

[0154] Example of a prompt:

[0155] "Please cluster patents related to electric vehicle technology."

[0156] "Classify patents in a specific technical field and visualize the clustered results."

[0157] Patent infringement confirmation

[0158] In the patent infringement verification function, the user uploads the specifications of the product under development to the server. The server analyzes the product specifications, compares them with an existing patent database, and assesses the potential patent infringement risk. The assessment results are generated as a report and notified to the user.

[0159] Specific example:

[0160] Companies developing new smartphones use the system to check for patent infringement risks. The server analyzes the product specifications, compares them with existing patents, assesses the infringement risk, and notifies the user of the results in a report.

[0161] Example of a prompt:

[0162] "Please check if these product specifications infringe on any other company's patent rights."

[0163] "Please analyze the specifications of the new smartphone and create a report assessing the patent infringement risk."

[0164] Real-time patent monitoring

[0165] In the real-time patent monitoring function, users configure the server with the technical fields and keywords to be monitored. The server monitors the patent database in real time based on the configured conditions and evaluates newly published patent information. When relevant patents are published, the user is immediately notified.

[0166] Specific example:

[0167] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[0168] Example of a prompt:

[0169] "Monitor new patents related to artificial intelligence technology in real time."

[0170] "Please collect newly published patent information based on specific keywords and notify us of any relevant information."

[0171] This system automates the patent application process, enabling efficient evaluation of patent novelty and inventiveness, patent clustering, patent infringement risk assessment, and real-time patent monitoring. This allows companies to significantly improve their patent management operations.

[0172] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0173] Step 1:

[0174] The user uploads the patent specification.

[0175] Input: The user selects a patent specification file (PDF or Word document) and clicks the upload button.

[0176] Specific actions:

[0177] The user selects a file and presses the upload button.

[0178] The server receives the uploaded files and temporarily stores them in the database.

[0179] Output: The patent specifications are saved on the server.

[0180] Step 2:

[0181] The server passes the patent specification to the NLP engine, which analyzes it and extracts the key sections.

[0182] Input: Saved patent specification file.

[0183] Specific actions:

[0184] The server calls a natural language processing engine (such as SpaCy or BERT) to analyze the document.

[0185] Using NLU (Natural Language Understanding) technology, sections such as "Title of Invention," "Background," "Purpose," and "Detailed Description" are extracted from the patent specification.

[0186] Output: The extracted section information is generated as structured data in JSON format.

[0187] Step 3:

[0188] The server searches existing patent databases based on the information in the patent specifications to find similar patents.

[0189] Input: Section information in JSON format.

[0190] Specific actions:

[0191] The server issues SQL or NoSQL queries to a patent database (e.g., the USPTO database).

[0192] Patents with high similarity are listed in a ranked format.

[0193] The search results are scored and sorted so that the most similar patents appear at the top.

[0194] Output: List and score of similar patents.

[0195] Step 4:

[0196] The server evaluates novelty and inventiveness.

[0197] Input: List and score of similar patents.

[0198] Specific actions:

[0199] The server uses machine learning models (e.g., random forests or support vector machines) to calculate evaluation scores for novelty and inventiveness.

[0200] The evaluation will take into account patent classification, cited documents, technical field, etc.

[0201] Output: Evaluation results for novelty and inventive step.

[0202] Step 5:

[0203] The server determines the likelihood of patent approval based on the evaluation results.

[0204] Input: Evaluation results for novelty and inventive step.

[0205] Specific actions:

[0206] The evaluation score is analyzed, and the likelihood of patent approval is assessed on a three-point scale: "high," "medium," and "low."

[0207] The results after the assessment will be compiled into a report.

[0208] Output: A detailed report on the likelihood of patent approval.

[0209] Step 6:

[0210] The server notifies the user of the evaluation results via the user interface or email.

[0211] Input: A report on the likelihood of patent approval.

[0212] Specific actions:

[0213] Generates reports in HTML and PDF formats.

[0214] The report will be sent to the user's email address or available for download on the dashboard.

[0215] Output: Reports accessible to the user.

[0216] Step 7:

[0217] The user sends a request to the server to cluster additional patent data.

[0218] Input: Conditions and keywords for the patent data you want to cluster.

[0219] Specific actions:

[0220] The user specifies the patent's technical field and keywords and submits a clustering request.

[0221] The server collects patent data based on specified conditions.

[0222] Output: Patent dataset for clustering.

[0223] Step 8:

[0224] The server uses a patent clustering algorithm to classify and visualize patents.

[0225] Input: Patent dataset for clustering.

[0226] Specific actions:

[0227] The server extracts features from the patent data and applies clustering algorithms (e.g., K-means, hierarchical clustering).

[0228] Assign an appropriate label to each classified cluster.

[0229] Visualize the clustering results and display them on the user interface.

[0230] Output: Visualized patent clustering results that can be viewed by the user.

[0231] Step 9:

[0232] The user uploads product specifications and initiates a process to assess patent infringement risk.

[0233] Input: Product specifications provided by the user.

[0234] Specific actions:

[0235] The user uploads the product specifications.

[0236] The server analyzes the uploaded product specifications.

[0237] Output: Analyzed product specification data.

[0238] Step 10:

[0239] The server analyzes the product specifications and compares them with the patent database.

[0240] Input: Analyzed product specification data.

[0241] Specific actions:

[0242] The server compares product specification data with the patent database.

[0243] The patent infringement risk is assessed based on the matching results.

[0244] Output: Patent infringement risk assessment results.

[0245] Step 11:

[0246] The server generates a report of the patent infringement risk assessment results and notifies the user.

[0247] Input: Patent infringement risk assessment results.

[0248] Specific actions:

[0249] Generate reports and provide them in HTML or PDF format.

[0250] The report will be sent to the user's email address or displayed on the dashboard.

[0251] Output: Report on the results of the patent infringement risk assessment.

[0252] In this way, the system efficiently supports patent-related tasks and provides advanced functions that enable users to manage patents quickly and accurately.

[0253] (Application Example 1)

[0254] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0255] Patent-related work requires a wide range of tasks, including evaluating novelty and inventiveness, confirming patent infringement risk, and clustering and visualizing patents, and efficiently performing each of these processes is difficult. Furthermore, risk assessment to avoid infringing on competitors' patents during new product development is crucial, but this also consumes significant time and resources. To address these challenges, a system that can streamline patent-related work in a centralized manner is needed.

[0256] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0257] In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for receiving product specifications and evaluating the risk of patent infringement based on those specifications, means for determining whether or not to grant a patent based on the evaluation results, means for notifying the user of the evaluation results, and means for evaluating the risk from the content of similar patents and providing it to the user visually. This enables centralized processing of patent-related tasks, allows for rapid evaluation of novelty and inventiveness, automatic determination of patent infringement risk, and visual management of patents using clustering, thereby significantly improving the efficiency of a company's product development process and patent management.

[0258] A "patent specification" is a document submitted when filing a patent application, and it includes the title of the invention, background, purpose, detailed description, and other information that describes the technical content of the invention.

[0259] "Analysis" is the process of analyzing input data and extracting specific information or patterns.

[0260] "Key sections" refer to important components within a patent specification, such as sections containing the invention's title, background, and detailed description.

[0261] A "patent database" is a data store where existing patent information is accumulated, and it is used for searching and matching patents.

[0262] "Similar patents" refer to existing patents that have similar technical content to the entered patent specifications or product specifications.

[0263] "Novelty" refers to the characteristic that indicates whether an invention differs from existing technology and is being publicly disclosed for the first time.

[0264] "Inventive step" is a characteristic that indicates the extent to which an invention has advanced compared to existing technology, in other words, whether or not there is a technological advancement.

[0265] "Evaluation results" refer to conclusions or opinions obtained based on an assessment of novelty, inventiveness, or patent infringement risk.

[0266] A "product specification" is a document that contains detailed descriptions of a product's design, functions, performance, and other aspects.

[0267] "Patent infringement risk" refers to the assessment of the likelihood that a particular product or technology infringes on a patent held by another company.

[0268] Clustering is a method of grouping data based on specific criteria to consolidate similar patents.

[0269] "Labeling" refers to assigning appropriate tags or names to each cluster.

[0270] "Visualization" refers to methods of displaying analysis results and evaluation results in the form of graphs, charts, and other diagrams to make them easier to understand.

[0271] "User" refers to a person who uses the system to perform patent-related tasks.

[0272] This invention provides an advanced AI system for efficiently performing patent-related tasks. The system integrates multiple functions, including receiving patent specifications, receiving product specifications, searching patent databases, risk assessment, clustering, and notifications.

[0273] server

[0274] The server receives patent specifications and product specifications from users, analyzes them, and extracts key sections. It uses a natural language processing (NLP) engine to perform document analysis and automatically extracts each section of the patent specification (such as the invention title, background, and detailed description). It also queries patent databases to search for similar patents and evaluates novelty and inventive step.

[0275] The server further analyzes the product specification and evaluates the patent infringement risk based on the specification. The risk evaluation is realized by comparing with the existing patent information. The evaluation result is notified to the user in a detailed report format.

[0276] Use a clustering algorithm to classify the collected patent data and group the patents based on specific technical fields or application fields. The clustering result is provided to the user using a visualization tool.

[0277] Terminal

[0278] The terminal functions as an interface for the user to input patent specifications and product specifications and receive the evaluation results and reports from the server. The user can access the system using devices such as smartphones, smart glasses, and head-mounted displays.

[0279] User

[0280] The user can upload the patent specification and product specification of a new invention to the system and receive the evaluation result. The user can judge the feasibility of patent application or evaluate the development risk of a new product based on the evaluation result.

[0281] The server executes the processing using the following specific hardware and software:

[0282] Hardware: Servers and cloud infrastructure (e.g., AWS (registered trademark), Google (registered trademark) Cloud)

[0283] Software: Python, requests library, BeautifulSoup, regular expressions, sklearn library, natural language processing (NLP) engine

[0284] Specific Example

[0285] The following shows specific examples:

[0286] Development of a new smartphone

[0287] A process in which a user uploads the specifications of a new smartphone to a system and the system evaluates the patent infringement risk based on it. When the system analyzes the specifications and searches for similar patents, a specific patent is evaluated as a high risk. This evaluation result is provided to the user as a detailed report, allowing the user to visually confirm which parts are judged to be at risk.

[0288] Examples of prompt sentences

[0289] "Please upload the specifications of the new product. Based on this, we will evaluate the patent infringement risk. Please clearly state the keywords and phrases to be used."

[0290] "To use the patent infringement risk assessment system, upload the product specifications, search the patent database based on the section information extracted using the NLP engine, and generate a risk assessment report. It is necessary to specify certain keywords and phrases."

[0291] The flow of a specific process in Application Example 1 will be described using FIG. 12.

[0292] Step 1:

[0293] The user uses a terminal to upload the specifications of a new product (text file) to the system. When the user uploads the specifications, the terminal sends the data to the server. The input data is the specifications of the new product (text format). The output data is the product specifications sent to the server.

[0294] Step 2:

[0295] The server analyzes the received product specifications. Specifically, it uses a natural language processing (NLP) engine to extract the main sections of the product specifications (e.g., technical specifications, feature descriptions, etc.). The input data is the uploaded product specifications. The output data is the extracted section information (in text format).

[0296] Step 3:

[0297] The server searches the patent database based on the extracted section information. Here, it uses the patent database API to query for similar patents. The input data is the extracted section information. The output data is similar patent information obtained from the patent database.

[0298] Step 4:

[0299] The server uses the acquired similar patent information to assess patent infringement risk. A specific algorithm is used to calculate the likelihood that a new product infringes existing patents. The input data is similar patent information. The output data is the patent infringement risk assessment result.

[0300] Step 5:

[0301] The server generates a detailed report based on the risk assessment results. This report includes detailed information about the patent infringement risk and its rationale. The report is generated in a visually easy-to-understand format. The input data is the patent infringement risk assessment results. The output data is a detailed risk assessment report (PDF or HTML format).

[0302] Step 6:

[0303] The server notifies the user of the generated risk assessment report. Specifically, it sends the report to the terminal, allowing the user to view it through the interface. The input data is the detailed risk assessment report. The output data is the report information notified to the user.

[0304] Step 7:

[0305] The user browses the risk assessment report sent from the server using the terminal. As a result, the user can check the patent infringement risk of the new product and take necessary measures. The input data is the notified risk assessment report. The output data is the report information browsed by the user.

[0306] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform specific processing using the user's emotion.

[0307] The present invention combines an advanced AI system for efficiently performing patent-related operations and an emotion engine for recognizing the user's emotion. The system operates as follows.

[0308] 1. Patent automatic determination

[0309] Subject: Server

[0310] The server receives the patent specification uploaded by the user to the system. Next, the received patent specification is analyzed by the NLP engine, and the main sections (title of the invention, background, purpose, detailed description, etc.) are extracted. The server accesses the patent database and searches for similar existing patents based on the content of the patent specification. The server uses the AI model to evaluate novelty and inventiveness and determines the possibility of patent establishment. The evaluation result is generated as a detailed report and notified to the user.

[0311] Specific example:

[0312] The user uploads the specification of a new invention to the system. This invention includes a new type of battery technology. After the server analyzes the specification and investigates similar existing patents, it determines that this invention has novelty and a high possibility of being established as a patent, and notifies the user of the evaluation result.

[0313] 2. Patent Clustering

[0314] Subject: Server

[0315] The server collects patent data related to a specific technology field. Next, it uses an AI algorithm to extract key features (keywords and section summaries) from each patent. Based on the extracted features, the server clusters the patent data using a clustering algorithm. The clustering results are presented to the user visually, making it easier to manage the patent portfolio.

[0316] Specific example:

[0317] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them to the user.

[0318] 3. Confirmation of patent infringement

[0319] Subject: Server

[0320] The server receives product specifications uploaded by the user. It analyzes the product specifications using an NLP engine to extract key functions and features. The server then compares the product specifications with a patent database to identify similar patents and potentially infringing patents. An AI model is used to assess the patent infringement risk, and the assessment results, including assigned risk scores, are generated as a report and notified to the user.

[0321] Specific example:

[0322] A company developing a new smartphone uses a server to analyze its product specifications to check if it infringes on other companies' patents. The server then compares the specifications with existing patents and determines if certain features pose a risk of infringing on others' patents. The evaluation results are then communicated to the user to encourage appropriate action.

[0323] 4. Real-time patent monitoring

[0324] Subject: Server

[0325] Users can configure the server with the technology fields and keywords they wish to monitor. Based on these settings, the server monitors the patent database in real time, collecting and analyzing newly published patent information. The server identifies important patents against the configured criteria and notifies the user. This ensures that users always have access to the latest patent information.

[0326] Specific example:

[0327] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[0328] 5. Combination of emotional engines

[0329] In addition to handling patent-related tasks, this system is equipped with an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's input and actions during operation to recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.).

[0330] Subject: Server

[0331] The server collects data on how users interact with the system (e.g., input speed, frequency of actions, screen time, etc.) and analyzes it using an emotion engine. Based on the analysis results, it adjusts appropriate notification methods and user interfaces. For example, if a user is experiencing stress, notifications can be changed to gentler and simpler ones.

[0332] Specific example:

[0333] When a user is reviewing the results of a patent infringement risk assessment, the emotion engine determines that the user's stress level is high. The server detects this state and adjusts the notifications and suggestions provided to the user. This makes it possible to improve the user experience.

[0334] This system not only streamlines the processes of patent application, patent infringement verification, patent clustering, and real-time patent monitoring, but also enables flexible responses that take into account the user's emotional state.

[0335] The following describes the processing flow.

[0336] 1. Automated patent determination

[0337] Step 1:

[0338] The server receives patent specifications uploaded by the user to the system.

[0339] Step 2:

[0340] The server analyzes the text of the patent specification using an NLP engine and extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[0341] Step 3:

[0342] The server accesses the patent database and searches for similar existing patents based on the content of the patent specifications.

[0343] Step 4:

[0344] The server uses an AI model to evaluate novelty and inventiveness. This includes comparing the extracted features with similar patents.

[0345] Step 5:

[0346] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report of the results.

[0347] Step 6:

[0348] The server notifies the user of the generated evaluation results report. The user can view the report via a dashboard or email.

[0349] 2. Patent Clustering

[0350] Step 1:

[0351] The server collects patent data related to a specific technology field from a database.

[0352] Step 2:

[0353] The server uses an AI algorithm to extract key features (keywords and section summaries) from each patent.

[0354] Step 3:

[0355] The server clusters the patent data using a clustering algorithm (e.g., k-means clustering) based on the extracted features.

[0356] Step 4:

[0357] The server automatically assigns appropriate labels to each cluster. These labels include those based on specific technologies and functions.

[0358] Step 5:

[0359] The server visually displays the clustering results on a dashboard, allowing users to easily manage their patent portfolios.

[0360] 3. Confirmation of patent infringement

[0361] Step 1:

[0362] The server receives product specifications uploaded by the user to the system.

[0363] Step 2:

[0364] The server analyzes the product specifications using an NLP engine and extracts key functions and features.

[0365] Step 3:

[0366] Based on the analysis results, the server compares the product specifications with the patent database to identify similar patents and patents that may infringe.

[0367] Step 4:

[0368] The server uses an AI model to assess the risk of patent infringement and assigns a risk score to each patent.

[0369] Step 5:

[0370] The server generates a patent infringement risk report based on the evaluation results.

[0371] Step 6:

[0372] The server notifies the user of the generated report, enabling the user to review the risks and take appropriate action.

[0373] 4. Real-time patent monitoring

[0374] Step 1:

[0375] The user configures the server with the technical fields and keywords to be monitored.

[0376] Step 2:

[0377] The server monitors the patent database in real time based on configured conditions and collects newly published patent information.

[0378] Step 3:

[0379] The server analyzes the collected patent information, compares it against the set conditions, and identifies important patents.

[0380] Step 4:

[0381] The server will notify users of newly identified patents. Available notification methods include email and dashboard notifications.

[0382] Step 5:

[0383] The server continuously monitors the patent database, ensuring that users always have access to the latest patent information.

[0384] 5. Combination of emotional engines

[0385] Step 1:

[0386] The server collects data on how users interact with the system (e.g., input speed, frequency of operations, screen time, etc.).

[0387] Step 2:

[0388] The server analyzes the collected data using an emotion engine to recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.).

[0389] Step 3:

[0390] The server adjusts notification methods and user interfaces based on the analysis results. For example, if a user is experiencing stress, it changes notifications to gentler and simpler ones.

[0391] Specific example:

[0392] When a user is reviewing the results of a patent infringement risk assessment, the emotion engine determines that the user's stress level is high. The server detects this state and adjusts the notifications and suggestions provided to the user. This makes it possible to improve the user experience.

[0393] These steps enable the system to efficiently manage patent-related tasks, reduce the risk of patent infringement, and quickly access patent information, as well as provide flexible responses that take into account the user's emotional state.

[0394] (Example 2)

[0395] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0396] Conventional patent-related business systems lacked the flexibility to consider the user's emotional state when analyzing patent specifications or assessing patent infringement risks. Furthermore, patent clustering and real-time monitoring were burdensome for users, making efficient management difficult. Therefore, there was a need for a sophisticated system that could simultaneously improve user experience and streamline operations.

[0397] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for analyzing the patent specification with a natural language processing engine and extracting key sections, means for searching for similar patents from an existing patent database, and means for using an AI model to evaluate novelty and inventiveness. This enables the user to efficiently evaluate the novelty and inventiveness of patents, identify infringement risks, manage patents through clustering, and monitor patents in real time. In addition, the user experience is simultaneously improved by a function that uses an emotion analysis engine to determine the user's emotional state and adjust the notification content.

[0398] A "patent specification" is a document submitted for patent application, and includes a detailed description of the invention and drawings.

[0399] A "natural language processing engine" is an algorithm or software that analyzes text data to understand its meaning and extract information.

[0400] A "patent database" is a database that stores information related to patents, and includes data on published patent documents and patent applications.

[0401] An "AI model" is a model used to perform data analysis and prediction using artificial intelligence, and is based on machine learning algorithms.

[0402] An "emotion analysis engine" is an algorithm or software that analyzes a user's emotional state from their input and operation data.

[0403] "Novelty" refers to the characteristic of an invention that is new and not present in existing technology or patent information.

[0404] "Inventive step" refers to the characteristic of an invention that possesses technical features that are a step ahead of existing technology.

[0405] Clustering is the process of dividing data into multiple clusters based on similarity, and is a method for grouping similar data.

[0406] "Labeling" is the process of assigning tags or names to clusters to make them easier to identify.

[0407] "Real-time monitoring" means constantly monitoring the latest information and responding immediately if any changes occur.

[0408] This invention is an advanced intellectual property management system for efficiently performing patent-related tasks, and is a system that can recognize and respond flexibly to the emotional state of the user. This system mainly consists of three main elements: a server, a terminal, and a user.

[0409] Patent automated determination

[0410] The server receives patent specifications uploaded by the user to the system. Next, it analyzes the received patent specifications using a natural language processing (NLP) engine (e.g., OpenAI®'s GPT-3®) to extract key sections (such as the invention title, background, purpose, and detailed description). The server accesses a patent database (e.g., Google Patents) to search for similar existing patents based on the content of the patent specifications. The server uses an AI model (e.g., a Transformer-Based Model) to evaluate novelty and inventiveness and determine the likelihood of patent granting. The evaluation results are generated as a detailed report and notified to the user.

[0411] Specific example:

[0412] When a user uploads a patent specification document for a new battery technology to the system, the server analyzes the specification document and searches for similar existing patents. Based on the results, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[0413] Examples of prompts for a generative AI model:

[0414] "Analyze user-uploaded patent specifications, extract key sections, and evaluate the novelty of the patents by comparing them to existing patent databases."

[0415] Patent clustering

[0416] The server collects patent data related to a specific technology field. Next, it extracts key features (keywords and section summaries) from each patent using an AI algorithm (e.g., BERT). Based on the extracted features, the server clusters the patent data using a clustering algorithm (e.g., K-Means clustering). The clustering results are presented to the user visually.

[0417] Specific example:

[0418] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them through a user interface.

[0419] Examples of prompts for a generative AI model:

[0420] "Collect patent data related to a specific technology field, extract key features, and cluster them. Visualize the clustering results and provide them to the user."

[0421] Patent infringement confirmation

[0422] The server receives product specifications uploaded by the user. It analyzes the product specifications using an NLP engine (e.g., RoBERTa) to extract key functions and features. The server then compares the product specifications with a patent database to identify similar patents and potentially infringing patents. An AI model (e.g., SVM) is used to assess the patent infringement risk, and the assessment results, including assigned risk scores, are generated as a report and notified to the user.

[0423] Specific example:

[0424] A company developing a new smartphone uploads its product specifications to a server to check if it infringes on any other company's patents. The server analyzes the product specifications and compares them with existing patents. If it determines that certain features pose a risk of infringing on another company's patents, it notifies the user of the evaluation result.

[0425] Examples of prompts for a generative AI model:

[0426] "Analyze the product specifications uploaded by the user, compare them with the existing patent database, and assess the patent infringement risk. Output the assessment results as a report."

[0427] Real-time patent monitoring

[0428] Users can configure the server with the technology fields and keywords they wish to monitor. Based on these settings, the server monitors the patent database in real time, collecting and analyzing newly published patent information. The server identifies important patents against the configured criteria and notifies the user. This ensures that users always have access to the latest patent information.

[0429] Specific example:

[0430] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information and immediately notifies the user when a relevant patent is published.

[0431] Examples of prompts for a generative AI model:

[0432] "Monitor the patent database in real time based on keywords set by the user, collect and analyze newly published patent information, and notify the user."

[0433] Combination of emotional engines

[0434] The server collects data in real time as the user interacts with the system (e.g., input speed, frequency of operations, screen dwell time, etc.). The server uses an emotion analysis engine (e.g., Microsoft® Azure® Emotional Intelligence API) to analyze the collected data and recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.). Based on the analysis results, it adjusts notification methods and the user interface to improve the user experience.

[0435] Specific example:

[0436] When a user reviews their patent infringement risk assessment, if the emotion engine determines the user's stress level is high, the server modifies the notification to be simpler and more user-friendly before providing it to the user. This improves the user experience and facilitates appropriate action.

[0437] This invention streamlines patent-related operations and enables flexible responses that take into account the user's emotional state, thereby improving the user experience.

[0438] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0439] Patent automated determination processing steps

[0440] Step 1:

[0441] Subject: User

[0442] Users upload patent specifications for their inventions to the system. The uploaded patent specifications are sent to the server.

[0443] Input: Patent specification file

[0444] Output: Patent specifications uploaded to the server

[0445] Step 2:

[0446] Subject: Server

[0447] The server analyzes the received patent specification using a natural language processing (NLP) engine (e.g., OpenAI's GPT-3). It extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[0448] Input: Uploaded patent specification

[0449] Output: Extracted main sections

[0450] Step 3:

[0451] Subject: Server

[0452] The server accesses a patent database (e.g., Google Patents) and searches for similar existing patents based on the extracted sections.

[0453] Input: Extracted main sections

[0454] Output: List of similar existing patents

[0455] Step 4:

[0456] Subject: Server

[0457] The server uses an AI model (e.g., a Transformer-Based Model) to evaluate the novelty and inventiveness of the patent specification.

[0458] Input: List of similar existing patents

[0459] Output: Evaluation results for novelty and inventive step

[0460] Step 5:

[0461] Subject: Server

[0462] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report. The generated report is then notified to the user.

[0463] Input: Evaluation results for novelty and inventive step

[0464] Output: Evaluation results report notified to the user.

[0465] Patent clustering processing steps

[0466] Step 1:

[0467] Subject: Server

[0468] The server collects patent data related to a specific technical field from patent databases (e.g., USPTO).

[0469] Input: Search query for patent database

[0470] Output: Set of patent data

[0471] Step 2:

[0472] Subject: Server

[0473] The server extracts key features (keywords and section summaries) from the collected patent data using an AI algorithm (e.g., BERT).

[0474] Input: Set of patent data

[0475] Output: List of extracted key features

[0476] Step 3:

[0477] Subject: Server

[0478] The server clusters the patent data using a clustering algorithm (e.g., K-Means clustering) based on the extracted features.

[0479] Input: List of key features extracted

[0480] Output: Patent data separated by cluster

[0481] Step 4:

[0482] Subject: Server

[0483] The server labels the clusters and visualizes the clustering results for the user.

[0484] Input: Patent data separated by cluster

[0485] Output: Visualized clustering results

[0486] Patent infringement confirmation processing steps

[0487] Step 1:

[0488] Subject: User

[0489] Users upload specifications for products under development to the system. The uploaded product specifications are sent to the server.

[0490] Input: Product specification file

[0491] Output: Product specifications uploaded to the server

[0492] Step 2:

[0493] Subject: Server

[0494] The server analyzes the received product specifications using a natural language processing (NLP) engine to extract key functions and features.

[0495] Input: Uploaded product specifications

[0496] Output: List of extracted key features or characteristics

[0497] Step 3:

[0498] Subject: Server

[0499] The server compares the product specifications with the patent database to identify similar patents and potentially infringing patents.

[0500] Input: List of extracted key features or characteristics

[0501] Output: List of similar patents

[0502] Step 4:

[0503] Subject: Server

[0504] The server uses an AI model (e.g., SVM) to assess the risk of patent infringement and assign a risk score.

[0505] Input: List of similar patents

[0506] Output: Patent infringement risk score and evaluation results

[0507] Step 5:

[0508] Subject: Server

[0509] The server generates a report of the assessment results regarding the risk of infringement and notifies the user.

[0510] Input: Patent infringement risk score and evaluation results

[0511] Output: Evaluation results report notified to the user.

[0512] Processing steps for real-time patent monitoring

[0513] Step 1:

[0514] Subject: User

[0515] Users configure the server with the technology fields and keywords they want to monitor. This information is stored on the server.

[0516] Input: Technical fields and keywords to be monitored

[0517] Output: Monitoring settings stored on the server

[0518] Step 2:

[0519] Subject: Server

[0520] The server monitors the patent database in real time based on saved monitoring settings.

[0521] Input: Monitoring settings stored on the server

[0522] Output: Real-time patent data update information

[0523] Step 3:

[0524] Subject: Server

[0525] The server collects newly published patent information from the patent database and analyzes it using an analysis engine.

[0526] Input: Real-time patent data update information

[0527] Output: Analyzed new patent information

[0528] Step 4:

[0529] Subject: Server

[0530] The server identifies important patents by comparing the analysis results with the configured conditions.

[0531] Input: Analyzed new patent information

[0532] Output: Identified important patent information

[0533] Step 5:

[0534] Subject: Server

[0535] The server notifies users of any important patent information it has identified. This notification is sent via email or through system alerts.

[0536] Input: Identified important patent information

[0537] Output: Patent information alerts sent to the user.

[0538] Processing steps for combining emotion engines

[0539] Step 1:

[0540] Subject: Server

[0541] The server collects data in real time as users interact with the system (e.g., input speed, frequency of operations, screen dwell time, etc.).

[0542] Input: User operation data

[0543] Output: Collected operational data

[0544] Step 2:

[0545] Subject: Server

[0546] The server uses an emotion analysis engine (e.g., Microsoft Azure Emotional Intelligence API) to analyze the collected data and recognize the user's emotional state.

[0547] Input: Collected operational data

[0548] Output: Analysis results of the user's emotional state

[0549] Step 3:

[0550] Subject: Server

[0551] The server adjusts notification methods and user interfaces based on the analysis results. For example, if a user is experiencing stress, it will change notifications to gentler and simpler ones.

[0552] Input: Analysis results of the user's emotional state

[0553] Output: Adjusted notification content and user interface

[0554] Step 4:

[0555] Subject: Server

[0556] The server provides users with configured notifications and user interfaces, and gathers feedback. This feedback is then used to make further improvements.

[0557] Input: Adjusted notification content and user interface

[0558] Output: User feedback and its analysis results

[0559] (Application Example 2)

[0560] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0561] Traditional patent-related work involves individually performing tasks such as evaluating the novelty and inventiveness of patents, verifying patent infringement, and real-time monitoring, which is inefficient. Furthermore, it has been difficult to provide feedback that takes into account the user's emotional state, resulting in insufficient improvement of the user experience. In addition, the rapid technological advancements in the field of autonomous vehicles present challenges in managing patent information.

[0562] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for evaluating novelty and inventiveness, means for determining whether or not to grant a patent based on the evaluation results, means for notifying the user of the evaluation results, means for monitoring the patent database in real time based on keywords, means for notifying the user of the monitoring results, means for analyzing the user's emotions, and means for adjusting the notification content based on the emotion analysis. This makes it possible to streamline patent-related work and realize flexible feedback that responds to the user's emotional state. Furthermore, it makes it easier to manage technical information in the field of autonomous vehicles and to always obtain the latest patent information.

[0563] A "patent specification" is a document that describes the details of an invention in order to obtain a patent.

[0564] A "main section" refers to the particularly important part of a patent specification, specifically including the title of the invention, background, purpose, and detailed description.

[0565] A "patent database" refers to a database that stores existing patent information, including published patents, classification information, inventor information, and so on.

[0566] "Novelty" is a criterion used to evaluate whether an invention is different from existing technology.

[0567] "Inventive step" is a criterion used to evaluate how much an invention has advanced compared to existing technology.

[0568] "User" refers to an individual or organization that uses the system to perform patent-related tasks.

[0569] A "keyword" refers to a specific word or phrase used when searching a patent database.

[0570] "Real-time monitoring" is a process that continuously monitors a database based on specific conditions and retrieves the latest information immediately.

[0571] "Emotion analysis" is a technology that analyzes a user's emotional state at a given time based on their input and operation data.

[0572] "Adjusting notification content" refers to changing the information provided and how it is displayed according to the user's emotional state.

[0573] "Patent approval denied" is the result of determining whether or not a particular patent can be obtained.

[0574] Clustering is a technique for grouping multiple data sets based on specific criteria.

[0575] "Labeling" refers to assigning appropriate labels to clustered data.

[0576] "Visualization" refers to displaying data in a visual format, such as graphs or charts.

[0577] A "product specification" is a document that describes the functions, features, and performance of a particular product.

[0578] "Matching" refers to the process of verifying matching items across different datasets.

[0579] "Patent infringement risk" is a criterion for assessing the likelihood that a particular product or technology infringes on existing patents.

[0580] "Report generation" is the process of compiling the results of data analysis into a document.

[0581] This invention provides a system that efficiently handles patent-related tasks and enables flexible responses that take into account the user's emotional state. In particular, it supports the real-time acquisition and management of patent information in the field of autonomous vehicle technology.

[0582] System Configuration

[0583] The server receives patent specifications and uses an NLP (Natural Language Processing) engine to extract key sections. Examples of NLP engines used include spaCy. Based on the extracted sections, the server searches existing patent databases for similar patents. Google Patents API is one example of a patent database that can be used.

[0584] Next, the server uses an AI model to evaluate the novelty and inventiveness of the patent, and based on the evaluation results, it decides whether or not to grant the patent. This evaluation result is generated as a detailed report and notified to the user. The notification to the user is done in a simple text format or using data visualization tools.

[0585] Furthermore, the server provides a function to monitor the patent database in real time based on keywords set by the user. This allows for the immediate acquisition of newly published patent information and notification to the user when relevant patents are published.

[0586] Examples of applications of the emotion engine

[0587] An emotion engine, which analyzes a user's emotional state, collects and analyzes data such as the user's input speed, frequency of operations, and screen dwell time. EmotionAPI is a typical example of an emotion engine used. Based on the results of this emotion analysis, notification content and the user interface are adjusted. For example, if the emotion engine determines that the user is in a high-stress state, the notification content is simplified and presented in a format that is easy for the user to understand.

[0588] Specific example

[0589] For example, if a user uploads a patent specification for a new autonomous vehicle technology, the server analyzes the specification and searches for similar existing patents. An AI model is used to evaluate novelty and inventiveness, and a decision is made regarding patent approval. Simultaneously, the system monitors the patent database in real time based on keywords set by the user and notifies the user when new relevant patents are published. Furthermore, the user's emotional state is analyzed during this process, and if a high stress level is detected, the notification content is simplified.

[0590] Example of a prompt

[0591] Analyze the specifications of new autonomous vehicle technologies, monitor patent databases in real time, and retrieve relevant patent information. Additionally, develop an application that adjusts notification content based on the user's stress level and sends notifications at appropriate times.

[0592] In this way, by implementing the present invention, it is possible to simultaneously improve the efficiency of patent-related work and enhance the user experience.

[0593] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0594] Step 1:

[0595] The server receives a patent specification from the user. The input is the patent specification, and the output is the text data of the patent specification. This text data is then passed to the next processing step.

[0596] Step 2:

[0597] The server analyzes the received patent specification using an NLP engine. The input is the text data of the patent specification, and the output is extracted data of the main sections of the patent specification (such as the title of the invention, background, purpose, and detailed description).

[0598] Step 3:

[0599] The server uses the extracted key section data to search for similar patents in existing patent databases. The input is the data from the key sections of the patent specification, and the output is a list of similar patents.

[0600] Step 4:

[0601] The server uses an AI model to evaluate the novelty and inventiveness of a patent specification. The input is data from the main sections of the patent specification and a list of similar patents, and the output is the evaluation result of novelty and inventiveness (patent likelihood assessment).

[0602] Step 5:

[0603] The server determines whether a patent is granted based on the evaluation results. The input is the evaluation results for novelty and inventive step, and the output is the decision on whether or not to grant the patent.

[0604] Step 6:

[0605] The server notifies the user of the decision regarding permission or denial. The input is the decision regarding permission or denial, and the output is the notification message to the user. The notification message is displayed on the user's interface.

[0606] Step 7:

[0607] The server monitors the patent database in real time based on keywords set by the user. The input is the keywords set by the user, and the output is a list of newly published patent information.

[0608] Step 8:

[0609] The server monitors the system in real time and notifies the user if any new relevant patents are published. The input is a list of newly published patent information, and the output is a notification message to the user.

[0610] Step 9:

[0611] The server collects user interaction data (input speed, frequency of actions, screen time, etc.) and analyzes it using an emotion engine. The input is interaction data, and the output is the user's emotional state (stress level, etc.).

[0612] Step 10:

[0613] The server adjusts notification content based on sentiment analysis. The input is the user's emotional state, and the output is the adjusted notification message. Efforts such as simplifying the notification message are implemented.

[0614] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0615] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0616] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0617] [Second Embodiment]

[0618] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0619] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0620] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0621] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0622] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0623] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0624] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0625] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0626] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0627] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0628] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0629] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0630] This invention is an advanced AI system for efficiently performing patent-related tasks. The system operates as follows:

[0631] 1. Automated patent determination

[0632] Subject: Server

[0633] The server receives the patent specification uploaded by the user. The received specification is analyzed by an NLP (Natural Language Processing) engine, and key sections (such as the invention title, background, purpose, and detailed description) are extracted. The server searches the patent database for similar existing patents and evaluates novelty and inventiveness. Based on the evaluation results, it determines the likelihood of patent granting and notifies the user of the result in a detailed report.

[0634] Specific example:

[0635] A user uploads a specification document for a new invention to the system. This invention includes a new type of battery technology. The server analyzes the specification document and searches for similar existing patents. As a result, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[0636] 2. Patent Clustering

[0637] Subject: Server

[0638] The server collects patent data related to a specific technical field. Next, it extracts key features (keywords and section summaries) from each patent and classifies them using a clustering algorithm. This groups patents based on specific technical fields or application areas. The clustering results are presented to the user visually.

[0639] Specific example:

[0640] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it more efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them to the user.

[0641] 3. Confirmation of patent infringement

[0642] Subject: Server

[0643] The server receives product specifications uploaded by the user and analyzes them. Based on the analysis results, it compares them with a patent database and assesses the potential patent infringement risk. The assessment results are generated in report format and notified to the user.

[0644] Specific example:

[0645] If a company developing a new smartphone wants to check whether its product infringes on another company's patents, the server analyzes the product specifications and compares them with existing patents. Based on the results, it determines if certain features pose a risk of infringing another company's patents. This result is then notified to the user, prompting them to take appropriate action.

[0646] 4. Real-time patent monitoring

[0647] Subject: Server

[0648] Users can configure the server with the technology fields and keywords they wish to monitor. Based on these settings, the server monitors the public patent database in real time. It evaluates newly published patent information, identifies patents important to the user, and notifies them. This process is continuous, ensuring users always stay up-to-date.

[0649] Specific example:

[0650] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[0651] This system streamlines the processes of patent application, patent infringement verification, patent clustering, and real-time patent monitoring, significantly improving a company's patent management operations.

[0652] The following describes the processing flow.

[0653] 1. Automated patent determination

[0654] Step 1:

[0655] The server receives patent specifications uploaded by the user to the system.

[0656] Step 2:

[0657] The server uses an NLP engine to perform text analysis on the received patent specification and extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[0658] Step 3:

[0659] The server accesses the patent database and searches for similar existing patents based on the content of the patent specifications.

[0660] Step 4:

[0661] The server uses an AI model to evaluate the novelty and inventiveness of the patent specification. This includes comparing the extracted features with similar patents.

[0662] Step 5:

[0663] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report of the results.

[0664] Step 6:

[0665] The server notifies the user of the generated evaluation results report. The user can view the report via a dashboard or email.

[0666] 2. Patent Clustering

[0667] Step 1:

[0668] The server collects patent data related to a specific technology field from a database.

[0669] Step 2:

[0670] The server uses an AI algorithm to extract key features (keywords and section summaries) from each patent.

[0671] Step 3:

[0672] The server clusters the patent data using a clustering algorithm (e.g., k-means clustering) based on the extracted features.

[0673] Step 4:

[0674] The server automatically assigns appropriate labels to each cluster. These labels include those based on specific technologies and functions.

[0675] Step 5:

[0676] The server visually displays clustering results on a dashboard, allowing users to easily understand and manage their patent portfolio.

[0677] 3. Confirmation of patent infringement

[0678] Step 1:

[0679] The server receives product specifications uploaded by the user to the system.

[0680] Step 2:

[0681] The server uses an NLP engine to perform text analysis on the received product specifications and extracts key functions and features.

[0682] Step 3:

[0683] The server compares the product specifications with the patent database to identify similar patents and potentially infringing patents.

[0684] Step 4:

[0685] The server uses an AI model to assess the risk of patent infringement and assigns a risk score to each patent.

[0686] Step 5:

[0687] The server generates a patent infringement risk report based on the evaluation results.

[0688] Step 6:

[0689] The server notifies the user of the generated report, enabling the user to review the risks and take appropriate action.

[0690] 4. Real-time patent monitoring

[0691] Step 1:

[0692] The user configures the server with the technical fields and keywords to be monitored.

[0693] Step 2:

[0694] The server monitors the patent database in real time based on configured conditions and collects newly published patent information.

[0695] Step 3:

[0696] The server analyzes the collected patent information, compares it against the set conditions, and identifies important patents.

[0697] Step 4:

[0698] The server will notify users of newly identified patents. Available notification methods include email and dashboard notifications.

[0699] Step 5:

[0700] The server continuously monitors the patent database, ensuring that users always have access to the latest patent information.

[0701] By following the steps outlined above, this system efficiently manages patent-related tasks, reducing the risk of patent infringement and enabling rapid access to patent information.

[0702] (Example 1)

[0703] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0704] Traditional patent-related work, including evaluating patent applications, analyzing patent clusters, and confirming patent infringement risks, was manual, time-consuming, labor-intensive, and lacked accuracy and consistency. Furthermore, evaluating novelty and inventiveness required specialized knowledge, making it difficult for companies to manage patents efficiently. In addition, real-time monitoring of patent information and access to the latest updates were insufficient.

[0705] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0706] In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for analyzing the specifications using a natural language processing engine, means for evaluating novelty and inventiveness using a machine learning model, and means for notifying the user of the generated report via a user interface or email. This enables the automation and streamlining of the patent application process.

[0707] The server also includes means for collecting patent data, means for extracting the features of the patents, means for clustering the patents, means for labeling the clusters, means for visualizing the clustering results, means for classifying the patents using a clustering algorithm, and means for providing the visualized clustering results through a user interface. This facilitates the organization and analysis of patent portfolios, enabling the formulation of patent strategies and appropriate decision-making.

[0708] Furthermore, the server includes means for receiving product specifications, means for analyzing the product specifications, means for comparing the product specifications with a patent database, means for evaluating patent infringement risk using a generated AI model, means for generating the evaluation results as a report, means for notifying the user of the evaluation results, and means for evaluating patent grantability based on customizable evaluation criteria. This enables rapid evaluation of patent infringement risk and appropriate response during product development.

[0709] A "patent specification" is a document submitted when filing a patent application, and it describes the details of the invention.

[0710] A "patent database" is a database that stores existing patent information and is used for searching and matching patents.

[0711] A "natural language processing engine" is a software module that provides technology for processing and analyzing human language using computers.

[0712] A "machine learning model" is an algorithm that learns from past data and uses that knowledge to make predictions and classifications.

[0713] A "user interface" is an interface through which a user interacts with a computer system, and includes graphical displays and methods of operation.

[0714] "Patent clustering" is a method of grouping and classifying related patents.

[0715] A "clustering algorithm" is an algorithm that automatically classifies data into meaningful groups (clusters).

[0716] A "generative AI model" is a model used by artificial intelligence (AI) to automatically generate and analyze data, and it includes pre-trained algorithms specialized for a specific purpose.

[0717] "Evaluation criteria" refer to the standards and indicators used when evaluating patents or products.

[0718] A "report" is a document that details the results of analysis and evaluation.

[0719] "Patent infringement risk" refers to the possibility that a product or technology infringes on existing patent rights.

[0720] This invention is an AI system for efficiently performing patent-related tasks. This system provides functions for automatic patent identification, patent clustering, patent infringement verification, and real-time patent monitoring.

[0721] Patent automated determination

[0722] In the automated patent evaluation function, the user uploads a patent specification, and the server receives it. The server uses an NLP (Natural Language Processing) engine to analyze the patent specification and extract key sections (such as the invention title, background, purpose, and detailed description). Next, the server compares it with an existing patent database to evaluate the novelty and inventiveness of the patent. The evaluation results are notified to the user as an automatically generated report. This allows the user to check the likelihood of patent granting with high accuracy.

[0723] Specific example:

[0724] When a user uploads a patent specification for a new battery technology to the system, the server analyzes the specification, compares it with existing patents, and evaluates the novelty and inventiveness of the invention. The evaluation results are then notified to the user in the form of a report.

[0725] Example of a prompt:

[0726] "Please evaluate this patent specification from the perspectives of novelty and inventive step."

[0727] "Please analyze the patent specifications for the new battery technology and create a report evaluating its patentability."

[0728] Patent clustering

[0729] In the patent clustering function, the server collects patent data related to a specific technical field. The server extracts the characteristics of each patent (keywords and section summaries) and classifies the patents using a clustering algorithm. The clustering results are visualized, allowing users to efficiently manage their patent portfolios.

[0730] Specific example:

[0731] A company with a large amount of patent data related to electric vehicle technology uses the system to organize its patents into clusters. The server uses a clustering algorithm to create clusters such as battery technology, motor control, and charging infrastructure, and provides them in a visualized format on the user interface.

[0732] Example of a prompt:

[0733] "Please cluster patents related to electric vehicle technology."

[0734] "Classify patents in a specific technical field and visualize the clustered results."

[0735] Patent infringement confirmation

[0736] In the patent infringement verification function, the user uploads the specifications of the product under development to the server. The server analyzes the product specifications, compares them with an existing patent database, and assesses the potential patent infringement risk. The assessment results are generated as a report and notified to the user.

[0737] Specific example:

[0738] Companies developing new smartphones use the system to check for patent infringement risks. The server analyzes the product specifications, compares them with existing patents, assesses the infringement risk, and notifies the user of the results in a report.

[0739] Example of a prompt:

[0740] "Please check if these product specifications infringe on any other company's patent rights."

[0741] "Please analyze the specifications of the new smartphone and create a report assessing the patent infringement risk."

[0742] Real-time patent monitoring

[0743] In the real-time patent monitoring function, users configure the server with the technical fields and keywords to be monitored. The server monitors the patent database in real time based on the configured conditions and evaluates newly published patent information. When relevant patents are published, the user is immediately notified.

[0744] Specific example:

[0745] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[0746] Example of a prompt:

[0747] "Monitor new patents related to artificial intelligence technology in real time."

[0748] "Please collect newly published patent information based on specific keywords and notify us of any relevant information."

[0749] This system automates the patent application process, enabling efficient evaluation of patent novelty and inventiveness, patent clustering, patent infringement risk assessment, and real-time patent monitoring. This allows companies to significantly improve their patent management operations.

[0750] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0751] Step 1:

[0752] The user uploads the patent specification.

[0753] Input: The user selects a patent specification file (PDF or Word document) and clicks the upload button.

[0754] Specific actions:

[0755] The user selects a file and presses the upload button.

[0756] The server receives the uploaded files and temporarily stores them in the database.

[0757] Output: The patent specifications are saved on the server.

[0758] Step 2:

[0759] The server passes the patent specification to the NLP engine, which analyzes it and extracts the key sections.

[0760] Input: Saved patent specification file.

[0761] Specific actions:

[0762] The server calls a natural language processing engine (such as SpaCy or BERT) to analyze the document.

[0763] Using NLU (Natural Language Understanding) technology, sections such as "Title of Invention," "Background," "Purpose," and "Detailed Description" are extracted from the patent specification.

[0764] Output: The extracted section information is generated as structured data in JSON format.

[0765] Step 3:

[0766] The server searches existing patent databases based on the information in the patent specifications to find similar patents.

[0767] Input: Section information in JSON format.

[0768] Specific actions:

[0769] The server issues SQL or NoSQL queries to a patent database (e.g., the USPTO database).

[0770] Patents with high similarity are listed in a ranked format.

[0771] The search results are scored and sorted so that the most similar patents appear at the top.

[0772] Output: List and score of similar patents.

[0773] Step 4:

[0774] The server evaluates novelty and inventiveness.

[0775] Input: List and score of similar patents.

[0776] Specific actions:

[0777] The server uses machine learning models (e.g., random forests or support vector machines) to calculate evaluation scores for novelty and inventiveness.

[0778] The evaluation will take into account patent classification, cited documents, technical field, etc.

[0779] Output: Evaluation results for novelty and inventive step.

[0780] Step 5:

[0781] The server determines the likelihood of patent approval based on the evaluation results.

[0782] Input: Evaluation results for novelty and inventive step.

[0783] Specific actions:

[0784] The evaluation score is analyzed, and the likelihood of patent approval is assessed on a three-point scale: "high," "medium," and "low."

[0785] The results after the assessment will be compiled into a report.

[0786] Output: A detailed report on the likelihood of patent approval.

[0787] Step 6:

[0788] The server notifies the user of the evaluation results via the user interface or email.

[0789] Input: A report on the likelihood of patent approval.

[0790] Specific actions:

[0791] Generates reports in HTML and PDF formats.

[0792] The report will be sent to the user's email address or available for download on the dashboard.

[0793] Output: Reports accessible to the user.

[0794] Step 7:

[0795] The user sends a request to the server to cluster additional patent data.

[0796] Input: Conditions and keywords for the patent data you want to cluster.

[0797] Specific actions:

[0798] The user specifies the patent's technical field and keywords and submits a clustering request.

[0799] The server collects patent data based on specified conditions.

[0800] Output: Patent dataset for clustering.

[0801] Step 8:

[0802] The server uses a patent clustering algorithm to classify and visualize patents.

[0803] Input: Patent dataset for clustering.

[0804] Specific actions:

[0805] The server extracts features from the patent data and applies clustering algorithms (e.g., K-means, hierarchical clustering).

[0806] Assign an appropriate label to each classified cluster.

[0807] Visualize the clustering results and display them on the user interface.

[0808] Output: Visualized patent clustering results that can be viewed by the user.

[0809] Step 9:

[0810] The user uploads product specifications and initiates a process to assess patent infringement risk.

[0811] Input: Product specifications provided by the user.

[0812] Specific actions:

[0813] The user uploads the product specifications.

[0814] The server analyzes the uploaded product specifications.

[0815] Output: Analyzed product specification data.

[0816] Step 10:

[0817] The server analyzes the product specifications and compares them with the patent database.

[0818] Input: Analyzed product specification data.

[0819] Specific actions:

[0820] The server compares product specification data with the patent database.

[0821] The patent infringement risk is assessed based on the matching results.

[0822] Output: Patent infringement risk assessment results.

[0823] Step 11:

[0824] The server generates a report of the patent infringement risk assessment results and notifies the user.

[0825] Input: Patent infringement risk assessment results.

[0826] Specific actions:

[0827] Generate reports and provide them in HTML or PDF format.

[0828] The report will be sent to the user's email address or displayed on the dashboard.

[0829] Output: Report on the results of the patent infringement risk assessment.

[0830] In this way, the system efficiently supports patent-related tasks and provides advanced functions that enable users to manage patents quickly and accurately.

[0831] (Application Example 1)

[0832] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0833] Patent-related work requires a wide range of tasks, including evaluating novelty and inventiveness, confirming patent infringement risk, and clustering and visualizing patents, making it difficult to perform each process efficiently. Furthermore, risk assessment to avoid infringing on competitors' patents during new product development is crucial, but this also consumes significant time and resources. To address these challenges, a system that can streamline patent-related work in a centralized manner is needed.

[0834] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0835] In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for receiving product specifications and evaluating the risk of patent infringement based on those specifications, means for determining whether or not to grant a patent based on the evaluation results, means for notifying the user of the evaluation results, and means for evaluating the risk from the content of similar patents and providing it to the user visually. This enables centralized processing of patent-related tasks, allows for rapid evaluation of novelty and inventiveness, automatic determination of patent infringement risk, and visual management of patents using clustering, thereby significantly improving the efficiency of a company's product development process and patent management.

[0836] A "patent specification" is a document submitted when filing a patent application, and it includes the title of the invention, background, purpose, detailed description, and other information that describes the technical content of the invention.

[0837] "Analysis" is the process of analyzing input data and extracting specific information or patterns.

[0838] "Key sections" refer to important components within a patent specification, such as sections containing the invention's title, background, and detailed description.

[0839] A "patent database" is a data store where existing patent information is accumulated, and it is used for searching and matching patents.

[0840] "Similar patents" refer to existing patents that have similar technical content to the entered patent specifications or product specifications.

[0841] "Novelty" refers to the characteristic that indicates whether an invention differs from existing technology and is being publicly disclosed for the first time.

[0842] "Inventive step" is a characteristic that indicates the extent to which an invention has advanced compared to existing technology, in other words, whether or not there is a technological advancement.

[0843] "Evaluation results" refer to conclusions or opinions obtained based on an assessment of novelty, inventiveness, or patent infringement risk.

[0844] A "product specification" is a document that contains detailed descriptions of a product's design, functions, performance, and other aspects.

[0845] "Patent infringement risk" refers to the assessment of the likelihood that a particular product or technology infringes on a patent held by another company.

[0846] Clustering is a method of grouping data based on specific criteria to consolidate similar patents.

[0847] "Labeling" refers to assigning appropriate tags or names to each cluster.

[0848] "Visualization" refers to methods of displaying analysis results and evaluation results in the form of graphs, charts, and other diagrams to make them easier to understand.

[0849] "User" refers to a person who uses the system to perform patent-related tasks.

[0850] This invention provides an advanced AI system for efficiently performing patent-related tasks. The system integrates multiple functions, including receiving patent specifications, receiving product specifications, searching patent databases, risk assessment, clustering, and notifications.

[0851] server

[0852] The server receives patent specifications and product specifications from users, analyzes them, and extracts key sections. It uses a natural language processing (NLP) engine to perform document analysis and automatically extract each section of the patent specification (such as the invention title, background, and detailed description). It also queries patent databases to search for similar patents and evaluates novelty and inventive step.

[0853] The server further analyzes the product specifications and assesses the patent infringement risk based on those specifications. This risk assessment is achieved by cross-referencing with existing patent information. The assessment results are communicated to the user in a detailed report format.

[0854] A clustering algorithm is used to classify the collected patent data and group patents based on specific technical fields or application areas. The clustering results are provided to the user using a visualization tool.

[0855] terminal

[0856] The terminal functions as an interface for users to input patent specifications and product specifications and receive evaluation results and reports from the server. Users can access the system using devices such as smartphones, smart glasses, and head-mounted displays.

[0857] User

[0858] Users can upload patent specifications or product specifications for new inventions to the system and receive evaluation results. Based on these evaluation results, users can decide whether or not to file a patent application or assess the development risks of new products.

[0859] The server uses the following specific hardware and software to perform the processing:

[0860] Hardware: Servers and cloud infrastructure (e.g., AWS, Google Cloud)

[0861] Software: Python, requests library, BeautifulSoup, regular expressions, scikit-learn library, natural language processing (NLP) engine

[0862] Specific example

[0863] Here are some specific examples:

[0864] Development of a new smartphone

[0865] This process involves users uploading specifications for their new smartphones to a system, which then evaluates the patent infringement risk. The system analyzes the specifications, searches for similar patents, and identifies specific patents as high-risk. This evaluation is then provided to the user in a detailed report, allowing them to visually confirm which parts were deemed risky.

[0866] Example of a prompt

[0867] "Please upload the specifications for your new product. We will use them to assess the patent infringement risk. Please clearly state any keywords or phrases you will be using."

[0868] "To use the patent infringement risk assessment system, you upload your product specifications, and the system searches the patent database based on the section information extracted using the NLP engine to generate a risk assessment report. It is necessary to specify certain keywords or phrases."

[0869] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0870] Step 1:

[0871] The user uploads the specifications (text file) for a new product to the system using a terminal. Once the user uploads the specifications, the terminal sends the data to the server. The input data is the specifications (text format) for the new product. The output data is the product specifications sent to the server.

[0872] Step 2:

[0873] The server analyzes the received product specifications. Specifically, it uses a natural language processing (NLP) engine to extract the main sections of the product specifications (e.g., technical specifications, feature descriptions, etc.). The input data is the uploaded product specifications. The output data is the extracted section information (in text format).

[0874] Step 3:

[0875] The server searches the patent database based on the extracted section information. Here, it uses the patent database API to query for similar patents. The input data is the extracted section information. The output data is similar patent information obtained from the patent database.

[0876] Step 4:

[0877] The server uses the acquired similar patent information to assess patent infringement risk. A specific algorithm is used to calculate the likelihood that a new product infringes existing patents. The input data is similar patent information. The output data is the patent infringement risk assessment result.

[0878] Step 5:

[0879] The server generates a detailed report based on the risk assessment results. This report includes detailed information about the patent infringement risk and its rationale. The report is generated in a visually easy-to-understand format. The input data is the patent infringement risk assessment results. The output data is a detailed risk assessment report (PDF or HTML format).

[0880] Step 6:

[0881] The server notifies the user of the generated risk assessment report. Specifically, it sends the report to the terminal, allowing the user to view it through the interface. The input data is the detailed risk assessment report. The output data is the report information notified to the user.

[0882] Step 7:

[0883] The user uses a terminal to view a risk assessment report sent from the server. This allows the user to check the patent infringement risk of the new product and take necessary actions. The input data is the notified risk assessment report. The output data is the report information viewed by the user.

[0884] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0885] This invention combines an advanced AI system for efficiently performing patent-related tasks with an emotion engine that recognizes user emotions. The system operates as follows:

[0886] 1. Automated patent determination

[0887] Subject: Server

[0888] The server receives patent specifications uploaded by the user to the system. Next, it analyzes the received patent specifications using an NLP engine and extracts key sections (such as the invention title, background, purpose, and detailed description). The server accesses a patent database and searches for similar existing patents based on the content of the patent specifications. The server uses an AI model to evaluate novelty and inventiveness and determine the likelihood of patent granting. The evaluation results are generated as a detailed report and notified to the user.

[0889] Specific example:

[0890] A user uploads a specification document for a new invention to the system. This invention includes a new type of battery technology. The server analyzes the specification document and searches for similar existing patents. As a result, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[0891] 2. Patent Clustering

[0892] Subject: Server

[0893] The server collects patent data related to a specific technology field. Next, it uses an AI algorithm to extract key features (keywords and section summaries) from each patent. Based on the extracted features, the server clusters the patent data using a clustering algorithm. The clustering results are presented to the user visually, making it easier to manage the patent portfolio.

[0894] Specific example:

[0895] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them to the user.

[0896] 3. Confirmation of patent infringement

[0897] Subject: Server

[0898] The server receives product specifications uploaded by the user. It analyzes the product specifications using an NLP engine to extract key functions and features. The server then compares the product specifications with a patent database to identify similar patents and potentially infringing patents. An AI model is used to assess the patent infringement risk, and the assessment results, including assigned risk scores, are generated as a report and notified to the user.

[0899] Specific example:

[0900] A company developing a new smartphone uses a server to analyze its product specifications to check if it infringes on other companies' patents. The server then compares the specifications with existing patents and determines if certain features pose a risk of infringing on others' patents. The evaluation results are then communicated to the user to encourage appropriate action.

[0901] 4. Real-time patent monitoring

[0902] Subject: Server

[0903] Users can configure the server with the technology fields and keywords they wish to monitor. Based on these settings, the server monitors the patent database in real time, collecting and analyzing newly published patent information. The server identifies important patents against the configured criteria and notifies the user. This ensures that users always have access to the latest patent information.

[0904] Specific example:

[0905] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[0906] 5. Combination of emotional engines

[0907] In addition to handling patent-related tasks, this system is equipped with an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's input and actions during operation to recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.).

[0908] Subject: Server

[0909] The server collects data on how users interact with the system (e.g., input speed, frequency of actions, screen time, etc.) and analyzes it using an emotion engine. Based on the analysis results, it adjusts appropriate notification methods and user interfaces. For example, if a user is experiencing stress, notifications can be changed to gentler and simpler ones.

[0910] Specific example:

[0911] When a user is reviewing the results of a patent infringement risk assessment, the emotion engine determines that the user's stress level is high. The server detects this state and adjusts the notifications and suggestions provided to the user. This makes it possible to improve the user experience.

[0912] This system not only streamlines the processes of patent application, patent infringement verification, patent clustering, and real-time patent monitoring, but also enables flexible responses that take into account the user's emotional state.

[0913] The following describes the processing flow.

[0914] 1. Automated patent determination

[0915] Step 1:

[0916] The server receives patent specifications uploaded by the user to the system.

[0917] Step 2:

[0918] The server analyzes the text of the patent specification using an NLP engine and extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[0919] Step 3:

[0920] The server accesses the patent database and searches for similar existing patents based on the content of the patent specifications.

[0921] Step 4:

[0922] The server uses an AI model to evaluate novelty and inventiveness. This includes comparing the extracted features with similar patents.

[0923] Step 5:

[0924] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report of the results.

[0925] Step 6:

[0926] The server notifies the user of the generated evaluation results report. The user can view the report via a dashboard or email.

[0927] 2. Patent Clustering

[0928] Step 1:

[0929] The server collects patent data related to a specific technology field from a database.

[0930] Step 2:

[0931] The server uses an AI algorithm to extract key features (keywords and section summaries) from each patent.

[0932] Step 3:

[0933] The server clusters the patent data using a clustering algorithm (e.g., k-means clustering) based on the extracted features.

[0934] Step 4:

[0935] The server automatically assigns appropriate labels to each cluster. These labels include those based on specific technologies and functions.

[0936] Step 5:

[0937] The server visually displays the clustering results on a dashboard, allowing users to easily manage their patent portfolios.

[0938] 3. Confirmation of patent infringement

[0939] Step 1:

[0940] The server receives product specifications uploaded by the user to the system.

[0941] Step 2:

[0942] The server analyzes the product specifications using an NLP engine and extracts key functions and features.

[0943] Step 3:

[0944] Based on the analysis results, the server compares the product specifications with the patent database to identify similar patents and patents that may infringe.

[0945] Step 4:

[0946] The server uses an AI model to assess the risk of patent infringement and assigns a risk score to each patent.

[0947] Step 5:

[0948] The server generates a patent infringement risk report based on the evaluation results.

[0949] Step 6:

[0950] The server notifies the user of the generated report, enabling the user to review the risks and take appropriate action.

[0951] 4. Real-time patent monitoring

[0952] Step 1:

[0953] The user configures the server with the technical fields and keywords to be monitored.

[0954] Step 2:

[0955] The server monitors the patent database in real time based on configured conditions and collects newly published patent information.

[0956] Step 3:

[0957] The server analyzes the collected patent information, compares it against the set conditions, and identifies important patents.

[0958] Step 4:

[0959] The server will notify users of newly identified patents. Available notification methods include email and dashboard notifications.

[0960] Step 5:

[0961] The server continuously monitors the patent database, ensuring that users always have access to the latest patent information.

[0962] 5. Combination of emotional engines

[0963] Step 1:

[0964] The server collects data on how users interact with the system (e.g., input speed, frequency of operations, screen time, etc.).

[0965] Step 2:

[0966] The server analyzes the collected data using an emotion engine to recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.).

[0967] Step 3:

[0968] The server adjusts notification methods and user interfaces based on the analysis results. For example, if a user is experiencing stress, it changes notifications to gentler and simpler ones.

[0969] Specific example:

[0970] When a user is reviewing the results of a patent infringement risk assessment, the emotion engine determines that the user's stress level is high. The server detects this state and adjusts the notifications and suggestions provided to the user. This makes it possible to improve the user experience.

[0971] These steps enable the system to efficiently manage patent-related tasks, reduce the risk of patent infringement, and quickly access patent information, as well as provide flexible responses that take into account the user's emotional state.

[0972] (Example 2)

[0973] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0974] Conventional patent-related business systems lacked the flexibility to consider the user's emotional state when analyzing patent specifications or assessing patent infringement risks. Furthermore, patent clustering and real-time monitoring were burdensome for users, making efficient management difficult. Therefore, there was a need for a sophisticated system that could simultaneously improve user experience and streamline operations.

[0975] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for analyzing the patent specification with a natural language processing engine and extracting key sections, means for searching for similar patents from an existing patent database, and means for using an AI model to evaluate novelty and inventiveness. This enables the user to efficiently evaluate the novelty and inventiveness of patents, identify infringement risks, manage patents through clustering, and monitor patents in real time. In addition, the user experience is simultaneously improved by a function that uses an emotion analysis engine to determine the user's emotional state and adjust the notification content.

[0976] A "patent specification" is a document submitted for patent application, and includes a detailed description of the invention and drawings.

[0977] A "natural language processing engine" is an algorithm or software that analyzes text data to understand its meaning and extract information.

[0978] A "patent database" is a database that stores information related to patents, and includes data on published patent documents and patent applications.

[0979] An "AI model" is a model used to perform data analysis and prediction using artificial intelligence, and is based on machine learning algorithms.

[0980] An "emotion analysis engine" is an algorithm or software that analyzes a user's emotional state from their input and operation data.

[0981] "Novelty" refers to the characteristic of an invention that is new and not present in existing technology or patent information.

[0982] "Inventive step" refers to the characteristic of an invention that possesses technical features that are a step ahead of existing technology.

[0983] Clustering is the process of dividing data into multiple clusters based on similarity, and is a method for grouping similar data.

[0984] "Labeling" is the process of assigning tags or names to clusters to make them easier to identify.

[0985] "Real-time monitoring" means constantly monitoring the latest information and responding immediately if any changes occur.

[0986] This invention is an advanced intellectual property management system for efficiently performing patent-related tasks, and is a system that can recognize and respond flexibly to the emotional state of the user. This system mainly consists of three main elements: a server, a terminal, and a user.

[0987] Patent automated determination

[0988] The server receives patent specifications uploaded by the user to the system. Next, it analyzes the received patent specifications using a natural language processing (NLP) engine (e.g., OpenAI's GPT-3) to extract key sections (such as the invention title, background, purpose, and detailed description). The server accesses a patent database (e.g., Google Patents) to search for similar existing patents based on the content of the patent specifications. The server uses an AI model (e.g., a Transformer-Based Model) to evaluate novelty and inventiveness and determine the likelihood of patent granting. The evaluation results are generated as a detailed report and notified to the user.

[0989] Specific example:

[0990] When a user uploads a patent specification document for a new battery technology to the system, the server analyzes the specification document and searches for similar existing patents. Based on the results, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[0991] Examples of prompts for a generative AI model:

[0992] "Analyze user-uploaded patent specifications, extract key sections, and evaluate the novelty of the patents by comparing them to existing patent databases."

[0993] Patent clustering

[0994] The server collects patent data related to a specific technology field. Next, it extracts key features (keywords and section summaries) from each patent using an AI algorithm (e.g., BERT). Based on the extracted features, the server clusters the patent data using a clustering algorithm (e.g., K-Means clustering). The clustering results are presented to the user visually.

[0995] Specific example:

[0996] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them through a user interface.

[0997] Examples of prompts for a generative AI model:

[0998] "Collect patent data related to a specific technology field, extract key features, and cluster them. Visualize the clustering results and provide them to the user."

[0999] Patent infringement confirmation

[1000] The server receives product specifications uploaded by the user. It analyzes the product specifications using an NLP engine (e.g., RoBERTa) to extract key functions and features. The server then compares the product specifications with a patent database to identify similar patents and potentially infringing patents. An AI model (e.g., SVM) is used to assess the patent infringement risk, and the assessment results, including assigned risk scores, are generated as a report and notified to the user.

[1001] Specific example:

[1002] A company developing a new smartphone uploads its product specifications to a server to check if it infringes on any other company's patents. The server analyzes the product specifications and compares them with existing patents. If it determines that certain features pose a risk of infringing on another company's patents, it notifies the user of the evaluation result.

[1003] Examples of prompts for a generative AI model:

[1004] "Analyze the product specifications uploaded by the user, compare them with the existing patent database, and assess the patent infringement risk. Output the assessment results as a report."

[1005] Real-time patent monitoring

[1006] Users can configure the server with the technology fields and keywords they want to monitor. Based on these settings, the server monitors the patent database in real time, collecting and analyzing newly published patent information. The server identifies important patents against the configured criteria and notifies the user. This ensures that users always have access to the latest patent information.

[1007] Specific example:

[1008] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information and immediately notifies the user when a relevant patent is published.

[1009] Examples of prompts for a generative AI model:

[1010] "Monitor the patent database in real time based on keywords set by the user, collect and analyze newly published patent information, and notify the user."

[1011] Combination of emotional engines

[1012] The server collects data in real time as the user interacts with the system (e.g., input speed, frequency of operations, screen dwell time, etc.). The server uses an emotion analysis engine (e.g., Microsoft Azure Emotional Intelligence API) to analyze the collected data and recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.). Based on the analysis results, the server adjusts notification methods and the user interface to improve the user experience.

[1013] Specific example:

[1014] When a user reviews their patent infringement risk assessment, if the emotion engine determines the user's stress level is high, the server modifies the notification to be simpler and more user-friendly before providing it to the user. This improves the user experience and facilitates appropriate action.

[1015] This invention streamlines patent-related operations and enables flexible responses that take into account the user's emotional state, thereby improving the user experience.

[1016] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1017] Patent automated determination processing steps

[1018] Step 1:

[1019] Subject: User

[1020] Users upload patent specifications for their inventions to the system. The uploaded patent specifications are sent to the server.

[1021] Input: Patent specification file

[1022] Output: Patent specifications uploaded to the server

[1023] Step 2:

[1024] Subject: Server

[1025] The server analyzes the received patent specification using a natural language processing (NLP) engine (e.g., OpenAI's GPT-3). It extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[1026] Input: Uploaded patent specification

[1027] Output: Extracted main sections

[1028] Step 3:

[1029] Subject: Server

[1030] The server accesses a patent database (e.g., Google Patents) and searches for similar existing patents based on the extracted sections.

[1031] Input: Extracted main sections

[1032] Output: List of similar existing patents

[1033] Step 4:

[1034] Subject: Server

[1035] The server uses an AI model (e.g., a Transformer-Based Model) to evaluate the novelty and inventiveness of the patent specification.

[1036] Input: List of similar existing patents

[1037] Output: Evaluation results for novelty and inventive step

[1038] Step 5:

[1039] Subject: Server

[1040] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report. The generated report is then notified to the user.

[1041] Input: Evaluation results for novelty and inventive step

[1042] Output: Evaluation results report notified to the user.

[1043] Patent clustering processing steps

[1044] Step 1:

[1045] Subject: Server

[1046] The server collects patent data related to a specific technical field from patent databases (e.g., USPTO).

[1047] Input: Search query for patent database

[1048] Output: Set of patent data

[1049] Step 2:

[1050] Subject: Server

[1051] The server extracts key features (keywords and section summaries) from the collected patent data using an AI algorithm (e.g., BERT).

[1052] Input: Set of patent data

[1053] Output: List of extracted key features

[1054] Step 3:

[1055] Subject: Server

[1056] The server clusters the patent data using a clustering algorithm (e.g., K-Means clustering) based on the extracted features.

[1057] Input: List of key features extracted

[1058] Output: Patent data separated by cluster

[1059] Step 4:

[1060] Subject: Server

[1061] The server labels the clusters and visualizes the clustering results for the user.

[1062] Input: Patent data separated by cluster

[1063] Output: Visualized clustering results

[1064] Patent infringement confirmation processing steps

[1065] Step 1:

[1066] Subject: User

[1067] Users upload specifications for products under development to the system. The uploaded product specifications are sent to the server.

[1068] Input: Product specification file

[1069] Output: Product specifications uploaded to the server

[1070] Step 2:

[1071] Subject: Server

[1072] The server analyzes the received product specifications using a natural language processing (NLP) engine to extract key functions and features.

[1073] Input: Uploaded product specifications

[1074] Output: List of extracted key features or characteristics

[1075] Step 3:

[1076] Subject: Server

[1077] The server compares the product specifications with the patent database to identify similar patents and potentially infringing patents.

[1078] Input: List of extracted key features or characteristics

[1079] Output: List of similar patents

[1080] Step 4:

[1081] Subject: Server

[1082] The server uses an AI model (e.g., SVM) to assess the risk of patent infringement and assign a risk score.

[1083] Input: List of similar patents

[1084] Output: Patent infringement risk score and evaluation results

[1085] Step 5:

[1086] Subject: Server

[1087] The server generates a report of the assessment results regarding the risk of infringement and notifies the user.

[1088] Input: Patent infringement risk score and evaluation results

[1089] Output: Evaluation results report notified to the user.

[1090] Processing steps for real-time patent monitoring

[1091] Step 1:

[1092] Subject: User

[1093] Users configure the server with the technology fields and keywords they want to monitor. This information is stored on the server.

[1094] Input: Technical fields and keywords to be monitored

[1095] Output: Monitoring settings stored on the server

[1096] Step 2:

[1097] Subject: Server

[1098] The server monitors the patent database in real time based on saved monitoring settings.

[1099] Input: Monitoring settings stored on the server

[1100] Output: Real-time patent data update information

[1101] Step 3:

[1102] Subject: Server

[1103] The server collects newly published patent information from the patent database and analyzes it using an analysis engine.

[1104] Input: Real-time patent data update information

[1105] Output: Analyzed new patent information

[1106] Step 4:

[1107] Subject: Server

[1108] The server identifies important patents by comparing the analysis results with the configured conditions.

[1109] Input: Analyzed new patent information

[1110] Output: Identified important patent information

[1111] Step 5:

[1112] Subject: Server

[1113] The server notifies users of any important patent information it has identified. This notification is sent via email or through system alerts.

[1114] Input: Identified important patent information

[1115] Output: Patent information alerts sent to the user.

[1116] Processing steps for combining emotion engines

[1117] Step 1:

[1118] Subject: Server

[1119] The server collects data in real time as users interact with the system (e.g., input speed, frequency of operations, screen dwell time, etc.).

[1120] Input: User operation data

[1121] Output: Collected operational data

[1122] Step 2:

[1123] Subject: Server

[1124] The server uses an emotion analysis engine (e.g., Microsoft Azure Emotional Intelligence API) to analyze the collected data and recognize the user's emotional state.

[1125] Input: Collected operational data

[1126] Output: Analysis results of the user's emotional state

[1127] Step 3:

[1128] Subject: Server

[1129] The server adjusts notification methods and user interfaces based on the analysis results. For example, if a user is experiencing stress, it will change notifications to gentler and simpler ones.

[1130] Input: Analysis results of the user's emotional state

[1131] Output: Adjusted notification content and user interface

[1132] Step 4:

[1133] Subject: Server

[1134] The server provides users with configured notifications and user interfaces, and receives feedback. This feedback is then used to make further improvements.

[1135] Input: Adjusted notification content and user interface

[1136] Output: User feedback and its analysis results

[1137] (Application Example 2)

[1138] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[1139] Traditional patent-related work involves individually performing tasks such as evaluating the novelty and inventiveness of patents, verifying patent infringement, and real-time monitoring, which is inefficient. Furthermore, it has been difficult to provide feedback that takes into account the user's emotional state, resulting in insufficient improvement of the user experience. In addition, the rapid technological advancements in the field of autonomous vehicles present challenges in managing patent information.

[1140] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for evaluating novelty and inventiveness, means for determining whether or not to grant a patent based on the evaluation results, means for notifying the user of the evaluation results, means for monitoring the patent database in real time based on keywords, means for notifying the user of the monitoring results, means for analyzing the user's emotions, and means for adjusting the notification content based on the emotion analysis. This makes it possible to streamline patent-related work and realize flexible feedback that responds to the user's emotional state. Furthermore, it makes it easier to manage technical information in the field of autonomous vehicles and to always obtain the latest patent information.

[1141] A "patent specification" is a document that describes the details of an invention in order to obtain a patent.

[1142] A "main section" refers to the particularly important part of a patent specification, specifically including the title of the invention, background, purpose, and detailed description.

[1143] A "patent database" refers to a database that stores existing patent information, including published patents, classification information, inventor information, and so on.

[1144] "Novelty" is a criterion used to evaluate whether an invention is different from existing technology.

[1145] "Inventive step" is a criterion used to evaluate how much an invention has advanced compared to existing technology.

[1146] "User" refers to an individual or organization that uses the system to perform patent-related tasks.

[1147] A "keyword" refers to a specific word or phrase used when searching a patent database.

[1148] "Real-time monitoring" is a process that continuously monitors a database based on specific conditions and retrieves the latest information immediately.

[1149] "Emotion analysis" is a technology that analyzes a user's emotional state at a given time based on their input and operation data.

[1150] "Adjusting notification content" refers to changing the information provided and how it is displayed according to the user's emotional state.

[1151] "Patent approval denied" is the result of determining whether or not a particular patent can be obtained.

[1152] Clustering is a technique for grouping multiple data sets based on specific criteria.

[1153] "Labeling" refers to assigning appropriate labels to clustered data.

[1154] "Visualization" refers to displaying data in a visual format, such as graphs or charts.

[1155] A "product specification" is a document that describes the functions, features, and performance of a particular product.

[1156] "Matching" refers to the process of verifying matching items across different datasets.

[1157] "Patent infringement risk" is a criterion for assessing the likelihood that a particular product or technology infringes on existing patents.

[1158] "Report generation" is the process of compiling the results of data analysis into a document.

[1159] This invention provides a system that efficiently handles patent-related tasks and enables flexible responses that take into account the user's emotional state. In particular, it supports the real-time acquisition and management of patent information in the field of autonomous vehicle technology.

[1160] System Configuration

[1161] The server receives patent specifications and uses an NLP (Natural Language Processing) engine to extract key sections. Examples of NLP engines used include spaCy. Based on the extracted sections, the server searches existing patent databases for similar patents. Google Patents API is one example of a patent database that can be used.

[1162] Next, the server uses an AI model to evaluate the novelty and inventiveness of the patent, and based on the evaluation results, it decides whether or not to grant the patent. This evaluation result is generated as a detailed report and notified to the user. The notification to the user is done in a simple text format or using data visualization tools.

[1163] Furthermore, the server provides a function to monitor the patent database in real time based on keywords set by the user. This allows for the immediate acquisition of newly published patent information and notification to the user when relevant patents are published.

[1164] Examples of applications of the emotion engine

[1165] An emotion engine, which analyzes a user's emotional state, collects and analyzes data such as the user's input speed, frequency of operations, and screen dwell time. EmotionAPI is a typical example of an emotion engine used. Based on the results of this emotion analysis, notification content and the user interface are adjusted. For example, if the emotion engine determines that the user is in a high-stress state, the notification content is simplified and presented in a format that is easy for the user to understand.

[1166] Specific example

[1167] For example, if a user uploads a patent specification for a new autonomous vehicle technology, the server analyzes the specification and searches for similar existing patents. An AI model is used to evaluate novelty and inventiveness, and a decision is made regarding patent approval. Simultaneously, the system monitors the patent database in real time based on keywords set by the user and notifies the user when new relevant patents are published. Furthermore, the user's emotional state is analyzed during this process, and if a high stress level is detected, the notification content is simplified.

[1168] Example of a prompt

[1169] Analyze the specifications of new autonomous vehicle technologies, monitor patent databases in real time, and retrieve relevant patent information. Additionally, develop an application that adjusts notification content based on the user's stress level and sends notifications at appropriate times.

[1170] In this way, by implementing the present invention, it is possible to simultaneously improve the efficiency of patent-related work and enhance the user experience.

[1171] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1172] Step 1:

[1173] The server receives a patent specification from the user. The input is the patent specification, and the output is the text data of the patent specification. This text data is then passed to the next processing step.

[1174] Step 2:

[1175] The server analyzes the received patent specification using an NLP engine. The input is the text data of the patent specification, and the output is extracted data of the main sections of the patent specification (such as the title of the invention, background, purpose, and detailed description).

[1176] Step 3:

[1177] The server uses the extracted key section data to search for similar patents in existing patent databases. The input is the data from the key sections of the patent specification, and the output is a list of similar patents.

[1178] Step 4:

[1179] The server uses an AI model to evaluate the novelty and inventiveness of a patent specification. The input is data from the main sections of the patent specification and a list of similar patents, and the output is the evaluation result of novelty and inventiveness (patent likelihood assessment).

[1180] Step 5:

[1181] The server determines whether a patent is granted based on the evaluation results. The input is the evaluation results for novelty and inventive step, and the output is the decision on whether or not to grant the patent.

[1182] Step 6:

[1183] The server notifies the user of the decision regarding permission or denial. The input is the decision regarding permission or denial, and the output is the notification message to the user. The notification message is displayed on the user's interface.

[1184] Step 7:

[1185] The server monitors the patent database in real time based on keywords set by the user. The input is the keywords set by the user, and the output is a list of newly published patent information.

[1186] Step 8:

[1187] The server monitors the system in real time and notifies the user if any new relevant patents are published. The input is a list of newly published patent information, and the output is a notification message to the user.

[1188] Step 9:

[1189] The server collects user interaction data (input speed, frequency of actions, screen time, etc.) and analyzes it using an emotion engine. The input is interaction data, and the output is the user's emotional state (stress level, etc.).

[1190] Step 10:

[1191] The server adjusts notification content based on sentiment analysis. The input is the user's emotional state, and the output is the adjusted notification message. Efforts such as simplifying the notification message are implemented.

[1192] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1193] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1194] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[1195] [Third Embodiment]

[1196] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1197] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1198] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1199] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[1200] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1201] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1202] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1203] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1204] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1205] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1206] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1207] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[1208] This invention is an advanced AI system for efficiently performing patent-related tasks. The system operates as follows:

[1209] 1. Automated patent determination

[1210] Subject: Server

[1211] The server receives the patent specification uploaded by the user. The received specification is analyzed by an NLP (Natural Language Processing) engine, and key sections (such as the invention title, background, purpose, and detailed description) are extracted. The server searches the patent database for similar existing patents and evaluates novelty and inventiveness. Based on the evaluation results, it determines the likelihood of patent granting and notifies the user of the result in a detailed report.

[1212] Specific example:

[1213] A user uploads a specification document for a new invention to the system. This invention includes a new type of battery technology. The server analyzes the specification document and searches for similar existing patents. As a result, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[1214] 2. Patent Clustering

[1215] Subject: Server

[1216] The server collects patent data related to a specific technical field. Next, it extracts key features (keywords and section summaries) from each patent and classifies them using a clustering algorithm. This groups patents based on specific technical fields or application areas. The clustering results are presented to the user visually.

[1217] Specific example:

[1218] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it more efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them to the user.

[1219] 3. Confirmation of patent infringement

[1220] Subject: Server

[1221] The server receives product specifications uploaded by the user and analyzes them. Based on the analysis results, it compares them with a patent database and assesses the potential patent infringement risk. The assessment results are generated in report format and notified to the user.

[1222] Specific example:

[1223] If a company developing a new smartphone wants to check whether its product infringes on another company's patents, the server analyzes the product specifications and compares them with existing patents. Based on the results, it determines if certain features pose a risk of infringing another company's patents. This result is then notified to the user, prompting them to take appropriate action.

[1224] 4. Real-time patent monitoring

[1225] Subject: Server

[1226] Users can configure the server with the technology fields and keywords they wish to monitor. Based on these settings, the server monitors the public patent database in real time. It evaluates newly published patent information, identifies patents important to the user, and notifies them. This process is continuous, ensuring users always stay up-to-date.

[1227] Specific example:

[1228] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[1229] This system streamlines the processes of patent application, patent infringement verification, patent clustering, and real-time patent monitoring, significantly improving a company's patent management operations.

[1230] The following describes the processing flow.

[1231] 1. Automated patent determination

[1232] Step 1:

[1233] The server receives patent specifications uploaded by the user to the system.

[1234] Step 2:

[1235] The server uses an NLP engine to perform text analysis on the received patent specification and extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[1236] Step 3:

[1237] The server accesses the patent database and searches for similar existing patents based on the content of the patent specifications.

[1238] Step 4:

[1239] The server uses an AI model to evaluate the novelty and inventiveness of the patent specification. This includes comparing the extracted features with similar patents.

[1240] Step 5:

[1241] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report of the results.

[1242] Step 6:

[1243] The server notifies the user of the generated evaluation results report. The user can view the report via a dashboard or email.

[1244] 2. Patent Clustering

[1245] Step 1:

[1246] The server collects patent data related to a specific technology field from a database.

[1247] Step 2:

[1248] The server uses an AI algorithm to extract key features (keywords and section summaries) from each patent.

[1249] Step 3:

[1250] The server clusters the patent data using a clustering algorithm (e.g., k-means clustering) based on the extracted features.

[1251] Step 4:

[1252] The server automatically assigns appropriate labels to each cluster. These labels include those based on specific technologies and functions.

[1253] Step 5:

[1254] The server visually displays clustering results on a dashboard, allowing users to easily understand and manage their patent portfolio.

[1255] 3. Confirmation of patent infringement

[1256] Step 1:

[1257] The server receives product specifications uploaded by the user to the system.

[1258] Step 2:

[1259] The server uses an NLP engine to perform text analysis on the received product specifications and extracts key functions and features.

[1260] Step 3:

[1261] The server compares the product specifications with the patent database to identify similar patents and potentially infringing patents.

[1262] Step 4:

[1263] The server uses an AI model to assess the risk of patent infringement and assigns a risk score to each patent.

[1264] Step 5:

[1265] The server generates a patent infringement risk report based on the evaluation results.

[1266] Step 6:

[1267] The server notifies the user of the generated report, enabling the user to review the risks and take appropriate action.

[1268] 4. Real-time patent monitoring

[1269] Step 1:

[1270] The user configures the server with the technical fields and keywords to be monitored.

[1271] Step 2:

[1272] The server monitors the patent database in real time based on configured conditions and collects newly published patent information.

[1273] Step 3:

[1274] The server analyzes the collected patent information, compares it against the set conditions, and identifies important patents.

[1275] Step 4:

[1276] The server will notify users of newly identified patents. Available notification methods include email and dashboard notifications.

[1277] Step 5:

[1278] The server continuously monitors the patent database, ensuring that users always have access to the latest patent information.

[1279] By following the steps outlined above, this system efficiently manages patent-related tasks, reducing the risk of patent infringement and enabling rapid access to patent information.

[1280] (Example 1)

[1281] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1282] Traditional patent-related work, including evaluating patent applications, analyzing patent clusters, and confirming patent infringement risks, was manual, time-consuming, labor-intensive, and lacked accuracy and consistency. Furthermore, evaluating novelty and inventiveness required specialized knowledge, making it difficult for companies to manage patents efficiently. In addition, real-time monitoring of patent information and access to the latest updates were insufficient.

[1283] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1284] In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for analyzing the specifications using a natural language processing engine, means for evaluating novelty and inventiveness using a machine learning model, and means for notifying the user of the generated report via a user interface or email. This enables the automation and streamlining of the patent application process.

[1285] The server also includes means for collecting patent data, means for extracting the features of the patents, means for clustering the patents, means for labeling the clusters, means for visualizing the clustering results, means for classifying the patents using a clustering algorithm, and means for providing the visualized clustering results through a user interface. This facilitates the organization and analysis of patent portfolios, enabling the formulation of patent strategies and appropriate decision-making.

[1286] Furthermore, the server includes means for receiving product specifications, means for analyzing the product specifications, means for comparing the product specifications with a patent database, means for evaluating patent infringement risk using a generated AI model, means for generating the evaluation results as a report, means for notifying the user of the evaluation results, and means for evaluating patent grantability based on customizable evaluation criteria. This enables rapid evaluation of patent infringement risk and appropriate response during product development.

[1287] A "patent specification" is a document submitted when filing a patent application, and it describes the details of the invention.

[1288] A "patent database" is a database that stores existing patent information and is used for searching and matching patents.

[1289] A "natural language processing engine" is a software module that provides technology for processing and analyzing human language using computers.

[1290] A "machine learning model" is an algorithm that learns from past data and uses that knowledge to make predictions and classifications.

[1291] A "user interface" is an interface through which a user interacts with a computer system, and includes graphical displays and methods of operation.

[1292] "Patent clustering" is a method of grouping and classifying related patents.

[1293] A "clustering algorithm" is an algorithm that automatically classifies data into meaningful groups (clusters).

[1294] A "generative AI model" is a model used by artificial intelligence (AI) to automatically generate and analyze data, and it includes pre-trained algorithms specialized for a specific purpose.

[1295] "Evaluation criteria" refer to the standards and indicators used when evaluating patents or products.

[1296] A "report" is a document that details the results of analysis and evaluation.

[1297] "Patent infringement risk" refers to the possibility that a product or technology infringes on existing patent rights.

[1298] This invention is an AI system for efficiently performing patent-related tasks. This system provides functions for automatic patent identification, patent clustering, patent infringement verification, and real-time patent monitoring.

[1299] Patent automated determination

[1300] In the automated patent evaluation function, the user uploads a patent specification, and the server receives it. The server uses an NLP (Natural Language Processing) engine to analyze the patent specification and extract key sections (such as the invention title, background, purpose, and detailed description). Next, the server compares it with an existing patent database to evaluate the novelty and inventiveness of the patent. The evaluation results are notified to the user as an automatically generated report. This allows the user to check the likelihood of patent granting with high accuracy.

[1301] Specific example:

[1302] When a user uploads a patent specification for a new battery technology to the system, the server analyzes the specification, compares it with existing patents, and evaluates the novelty and inventiveness of the invention. The evaluation results are then notified to the user in the form of a report.

[1303] Example of a prompt:

[1304] "Please evaluate this patent specification from the perspectives of novelty and inventive step."

[1305] "Please analyze the patent specifications for the new battery technology and create a report evaluating its patentability."

[1306] Patent clustering

[1307] In the patent clustering function, the server collects patent data related to a specific technical field. The server extracts the characteristics of each patent (keywords and section summaries) and classifies the patents using a clustering algorithm. The clustering results are visualized, allowing users to efficiently manage their patent portfolios.

[1308] Specific example:

[1309] A company with a large amount of patent data related to electric vehicle technology uses the system to organize its patents into clusters. The server uses a clustering algorithm to create clusters such as battery technology, motor control, and charging infrastructure, and provides them in a visualized format on the user interface.

[1310] Example of a prompt:

[1311] "Please cluster patents related to electric vehicle technology."

[1312] "Classify patents in a specific technical field and visualize the clustered results."

[1313] Patent infringement confirmation

[1314] In the patent infringement verification function, the user uploads the specifications of the product under development to the server. The server analyzes the product specifications, compares them with an existing patent database, and assesses the potential patent infringement risk. The assessment results are generated as a report and notified to the user.

[1315] Specific example:

[1316] Companies developing new smartphones use the system to check for patent infringement risks. The server analyzes the product specifications, compares them with existing patents, assesses the infringement risk, and notifies the user of the results in a report.

[1317] Example of a prompt:

[1318] "Please check if these product specifications infringe on any other company's patent rights."

[1319] "Please analyze the specifications of the new smartphone and create a report assessing the patent infringement risk."

[1320] Real-time patent monitoring

[1321] In the real-time patent monitoring function, users configure the server with the technical fields and keywords to be monitored. The server monitors the patent database in real time based on the configured conditions and evaluates newly published patent information. When relevant patents are published, the user is immediately notified.

[1322] Specific example:

[1323] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[1324] Example of a prompt:

[1325] "Monitor new patents related to artificial intelligence technology in real time."

[1326] "Please collect newly published patent information based on specific keywords and notify us of any relevant information."

[1327] This system automates the patent application process, enabling efficient evaluation of patent novelty and inventiveness, patent clustering, patent infringement risk assessment, and real-time patent monitoring. This allows companies to significantly improve their patent management operations.

[1328] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1329] Step 1:

[1330] The user uploads the patent specification.

[1331] Input: The user selects a patent specification file (PDF or Word document) and clicks the upload button.

[1332] Specific actions:

[1333] The user selects a file and presses the upload button.

[1334] The server receives the uploaded files and temporarily stores them in the database.

[1335] Output: The patent specifications are saved on the server.

[1336] Step 2:

[1337] The server passes the patent specification to the NLP engine, which analyzes it and extracts the key sections.

[1338] Input: Saved patent specification file.

[1339] Specific actions:

[1340] The server calls a natural language processing engine (such as SpaCy or BERT) to analyze the document.

[1341] Using NLU (Natural Language Understanding) technology, sections such as "Title of Invention," "Background," "Purpose," and "Detailed Description" are extracted from the patent specification.

[1342] Output: The extracted section information is generated as structured data in JSON format.

[1343] Step 3:

[1344] The server searches existing patent databases based on the information in the patent specifications to find similar patents.

[1345] Input: Section information in JSON format.

[1346] Specific actions:

[1347] The server issues SQL or NoSQL queries to a patent database (e.g., the USPTO database).

[1348] Patents with high similarity are listed in a ranked format.

[1349] The search results are scored and sorted so that the most similar patents appear at the top.

[1350] Output: List and score of similar patents.

[1351] Step 4:

[1352] The server evaluates novelty and inventiveness.

[1353] Input: List and score of similar patents.

[1354] Specific actions:

[1355] The server uses machine learning models (e.g., random forests or support vector machines) to calculate evaluation scores for novelty and inventiveness.

[1356] The evaluation will take into account patent classification, cited documents, technical field, etc.

[1357] Output: Evaluation results for novelty and inventive step.

[1358] Step 5:

[1359] The server determines the likelihood of patent approval based on the evaluation results.

[1360] Input: Evaluation results for novelty and inventive step.

[1361] Specific actions:

[1362] The evaluation score is analyzed, and the likelihood of patent approval is assessed on a three-point scale: "high," "medium," and "low."

[1363] The results after the assessment will be compiled into a report.

[1364] Output: A detailed report on the likelihood of patent approval.

[1365] Step 6:

[1366] The server notifies the user of the evaluation results via the user interface or email.

[1367] Input: A report on the likelihood of patent approval.

[1368] Specific actions:

[1369] Generates reports in HTML and PDF formats.

[1370] The report will be sent to the user's email address or available for download on the dashboard.

[1371] Output: Reports accessible to the user.

[1372] Step 7:

[1373] The user sends a request to the server to cluster additional patent data.

[1374] Input: Conditions and keywords for the patent data you want to cluster.

[1375] Specific actions:

[1376] The user specifies the patent's technical field and keywords and submits a clustering request.

[1377] The server collects patent data based on specified conditions.

[1378] Output: Patent dataset for clustering.

[1379] Step 8:

[1380] The server uses a patent clustering algorithm to classify and visualize patents.

[1381] Input: Patent dataset for clustering.

[1382] Specific actions:

[1383] The server extracts features from the patent data and applies clustering algorithms (e.g., K-means, hierarchical clustering).

[1384] Assign an appropriate label to each classified cluster.

[1385] Visualize the clustering results and display them on the user interface.

[1386] Output: Visualized patent clustering results that can be viewed by the user.

[1387] Step 9:

[1388] The user uploads product specifications and initiates a process to assess patent infringement risk.

[1389] Input: Product specifications provided by the user.

[1390] Specific actions:

[1391] The user uploads the product specifications.

[1392] The server analyzes the uploaded product specifications.

[1393] Output: Analyzed product specification data.

[1394] Step 10:

[1395] The server analyzes the product specifications and compares them with the patent database.

[1396] Input: Analyzed product specification data.

[1397] Specific actions:

[1398] The server compares product specification data with the patent database.

[1399] The patent infringement risk is assessed based on the matching results.

[1400] Output: Patent infringement risk assessment results.

[1401] Step 11:

[1402] The server generates a report of the patent infringement risk assessment results and notifies the user.

[1403] Input: Patent infringement risk assessment results.

[1404] Specific actions:

[1405] Generate reports and provide them in HTML or PDF format.

[1406] The report will be sent to the user's email address or displayed on the dashboard.

[1407] Output: Report on the results of the patent infringement risk assessment.

[1408] In this way, the system efficiently supports patent-related tasks and provides advanced functions that enable users to manage patents quickly and accurately.

[1409] (Application Example 1)

[1410] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1411] Patent-related work requires a wide range of tasks, including evaluating novelty and inventiveness, confirming patent infringement risk, and clustering and visualizing patents, making it difficult to perform each process efficiently. Furthermore, risk assessment to avoid infringing on competitors' patents during new product development is crucial, but this also consumes significant time and resources. To address these challenges, a system that can streamline patent-related work in a centralized manner is needed.

[1412] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1413] In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for receiving product specifications and evaluating the risk of patent infringement based on those specifications, means for determining whether or not to grant a patent based on the evaluation results, means for notifying the user of the evaluation results, and means for evaluating the risk from the content of similar patents and providing it to the user visually. This enables centralized processing of patent-related tasks, allows for rapid evaluation of novelty and inventiveness, automatic determination of patent infringement risk, and visual management of patents using clustering, thereby significantly improving the efficiency of a company's product development process and patent management.

[1414] A "patent specification" is a document submitted when filing a patent application, and it includes the title of the invention, background, purpose, detailed description, and other information that describes the technical content of the invention.

[1415] "Analysis" is the process of analyzing input data and extracting specific information or patterns.

[1416] "Key sections" refer to important components within a patent specification, such as sections containing the invention's title, background, and detailed description.

[1417] A "patent database" is a data store where existing patent information is accumulated, and it is used for searching and matching patents.

[1418] "Similar patents" refer to existing patents that have similar technical content to the entered patent specifications or product specifications.

[1419] "Novelty" refers to the characteristic that indicates whether an invention differs from existing technology and is being publicly disclosed for the first time.

[1420] "Inventive step" is a characteristic that indicates the extent to which an invention has advanced compared to existing technology, in other words, whether or not there is a technological advancement.

[1421] "Evaluation results" refer to conclusions or opinions obtained based on an assessment of novelty, inventiveness, or patent infringement risk.

[1422] A "product specification" is a document that contains detailed descriptions of a product's design, functions, performance, and other aspects.

[1423] "Patent infringement risk" refers to the assessment of the likelihood that a particular product or technology infringes on a patent held by another company.

[1424] Clustering is a method of grouping data based on specific criteria to consolidate similar patents.

[1425] "Labeling" refers to assigning appropriate tags or names to each cluster.

[1426] "Visualization" refers to methods of displaying analysis results and evaluation results in the form of graphs, charts, and other diagrams to make them easier to understand.

[1427] "User" refers to a person who uses the system to perform patent-related tasks.

[1428] This invention provides an advanced AI system for efficiently performing patent-related tasks. The system integrates multiple functions, including receiving patent specifications, receiving product specifications, searching patent databases, risk assessment, clustering, and notifications.

[1429] server

[1430] The server receives patent specifications and product specifications from users, analyzes them, and extracts key sections. It uses a natural language processing (NLP) engine to perform document analysis and automatically extract each section of the patent specification (such as the invention title, background, and detailed description). It also queries patent databases to search for similar patents and evaluates novelty and inventive step.

[1431] The server further analyzes the product specifications and assesses the patent infringement risk based on those specifications. This risk assessment is achieved by cross-referencing with existing patent information. The assessment results are communicated to the user in a detailed report format.

[1432] A clustering algorithm is used to classify the collected patent data and group patents based on specific technical fields or application areas. The clustering results are provided to the user using a visualization tool.

[1433] terminal

[1434] The terminal functions as an interface for users to input patent specifications and product specifications and receive evaluation results and reports from the server. Users can access the system using devices such as smartphones, smart glasses, and head-mounted displays.

[1435] User

[1436] Users can upload patent specifications or product specifications for new inventions to the system and receive evaluation results. Based on these evaluation results, users can decide whether or not to file a patent application or assess the development risks of new products.

[1437] The server uses the following specific hardware and software to perform the processing:

[1438] Hardware: Servers and cloud infrastructure (e.g., AWS, Google Cloud)

[1439] Software: Python, requests library, BeautifulSoup, regular expressions, scikit-learn library, natural language processing (NLP) engine

[1440] Specific example

[1441] Here are some specific examples:

[1442] Development of a new smartphone

[1443] This process involves users uploading specifications for their new smartphones to a system, which then evaluates the patent infringement risk. The system analyzes the specifications, searches for similar patents, and identifies specific patents as high-risk. This evaluation is then provided to the user in a detailed report, allowing them to visually confirm which parts were deemed risky.

[1444] Example of a prompt

[1445] "Please upload the specifications for your new product. We will use them to assess the patent infringement risk. Please clearly state any keywords or phrases you will be using."

[1446] "To use the patent infringement risk assessment system, you upload your product specifications, and the system searches the patent database based on the section information extracted using the NLP engine to generate a risk assessment report. It is necessary to specify certain keywords or phrases."

[1447] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1448] Step 1:

[1449] The user uploads the specifications (text file) for a new product to the system using a terminal. Once the user uploads the specifications, the terminal sends the data to the server. The input data is the specifications (text format) for the new product. The output data is the product specifications sent to the server.

[1450] Step 2:

[1451] The server analyzes the received product specifications. Specifically, it uses a natural language processing (NLP) engine to extract the main sections of the product specifications (e.g., technical specifications, feature descriptions, etc.). The input data is the uploaded product specifications. The output data is the extracted section information (in text format).

[1452] Step 3:

[1453] The server searches the patent database based on the extracted section information. Here, it uses the patent database API to query for similar patents. The input data is the extracted section information. The output data is similar patent information obtained from the patent database.

[1454] Step 4:

[1455] The server uses the acquired similar patent information to assess patent infringement risk. A specific algorithm is used to calculate the likelihood that a new product infringes existing patents. The input data is similar patent information. The output data is the patent infringement risk assessment result.

[1456] Step 5:

[1457] The server generates a detailed report based on the risk assessment results. This report includes detailed information about the patent infringement risk and its rationale. The report is generated in a visually easy-to-understand format. The input data is the patent infringement risk assessment results. The output data is a detailed risk assessment report (PDF or HTML format).

[1458] Step 6:

[1459] The server notifies the user of the generated risk assessment report. Specifically, it sends the report to the terminal, allowing the user to view it through the interface. The input data is the detailed risk assessment report. The output data is the report information notified to the user.

[1460] Step 7:

[1461] The user uses a terminal to view a risk assessment report sent from the server. This allows the user to check the patent infringement risk of the new product and take necessary actions. The input data is the notified risk assessment report. The output data is the report information viewed by the user.

[1462] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1463] This invention combines an advanced AI system for efficiently performing patent-related tasks with an emotion engine that recognizes user emotions. The system operates as follows:

[1464] 1. Automated patent determination

[1465] Subject: Server

[1466] The server receives patent specifications uploaded by the user to the system. Next, it analyzes the received patent specifications using an NLP engine and extracts key sections (such as the invention title, background, purpose, and detailed description). The server accesses a patent database and searches for similar existing patents based on the content of the patent specifications. The server uses an AI model to evaluate novelty and inventiveness and determine the likelihood of patent granting. The evaluation results are generated as a detailed report and notified to the user.

[1467] Specific example:

[1468] A user uploads a specification document for a new invention to the system. This invention includes a new type of battery technology. The server analyzes the specification document and searches for similar existing patents. As a result, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[1469] 2. Patent Clustering

[1470] Subject: Server

[1471] The server collects patent data related to a specific technology field. Next, it uses an AI algorithm to extract key features (keywords and section summaries) from each patent. Based on the extracted features, the server clusters the patent data using a clustering algorithm. The clustering results are presented to the user visually, making it easier to manage the patent portfolio.

[1472] Specific example:

[1473] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them to the user.

[1474] 3. Confirmation of patent infringement

[1475] Subject: Server

[1476] The server receives product specifications uploaded by the user. It analyzes the product specifications using an NLP engine to extract key functions and features. The server then compares the product specifications with a patent database to identify similar patents and potentially infringing patents. An AI model is used to assess the patent infringement risk, and the assessment results, including assigned risk scores, are generated as a report and notified to the user.

[1477] Specific example:

[1478] A company developing a new smartphone uses a server to analyze its product specifications to check if it infringes on other companies' patents. The server then compares the specifications with existing patents and determines if certain features pose a risk of infringing on others' patents. The evaluation results are then communicated to the user to encourage appropriate action.

[1479] 4. Real-time patent monitoring

[1480] Subject: Server

[1481] Users can configure the server with the technology fields and keywords they wish to monitor. Based on these settings, the server monitors the patent database in real time, collecting and analyzing newly published patent information. The server identifies important patents against the configured criteria and notifies the user. This ensures that users always have access to the latest patent information.

[1482] Specific example:

[1483] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[1484] 5. Combination of emotional engines

[1485] In addition to handling patent-related tasks, this system is equipped with an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's input and actions during operation to recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.).

[1486] Subject: Server

[1487] The server collects data on how users interact with the system (e.g., input speed, frequency of actions, screen time, etc.) and analyzes it using an emotion engine. Based on the analysis results, it adjusts appropriate notification methods and user interfaces. For example, if a user is experiencing stress, notifications can be changed to gentler and simpler ones.

[1488] Specific example:

[1489] When a user is reviewing the results of a patent infringement risk assessment, the emotion engine determines that the user's stress level is high. The server detects this state and adjusts the notifications and suggestions provided to the user. This makes it possible to improve the user experience.

[1490] This system not only streamlines the processes of patent application, patent infringement verification, patent clustering, and real-time patent monitoring, but also enables flexible responses that take into account the user's emotional state.

[1491] The following describes the processing flow.

[1492] 1. Automated patent determination

[1493] Step 1:

[1494] The server receives patent specifications uploaded by the user to the system.

[1495] Step 2:

[1496] The server analyzes the text of the patent specification using an NLP engine and extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[1497] Step 3:

[1498] The server accesses the patent database and searches for similar existing patents based on the content of the patent specifications.

[1499] Step 4:

[1500] The server uses an AI model to evaluate novelty and inventiveness. This includes comparing the extracted features with similar patents.

[1501] Step 5:

[1502] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report of the results.

[1503] Step 6:

[1504] The server notifies the user of the generated evaluation results report. The user can view the report via a dashboard or email.

[1505] 2. Patent Clustering

[1506] Step 1:

[1507] The server collects patent data related to a specific technology field from a database.

[1508] Step 2:

[1509] The server uses an AI algorithm to extract key features (keywords and section summaries) from each patent.

[1510] Step 3:

[1511] The server clusters the patent data using a clustering algorithm (e.g., k-means clustering) based on the extracted features.

[1512] Step 4:

[1513] The server automatically assigns appropriate labels to each cluster. These labels include those based on specific technologies and functions.

[1514] Step 5:

[1515] The server visually displays the clustering results on a dashboard, allowing users to easily manage their patent portfolios.

[1516] 3. Confirmation of patent infringement

[1517] Step 1:

[1518] The server receives product specifications uploaded by the user to the system.

[1519] Step 2:

[1520] The server analyzes the product specifications using an NLP engine and extracts key functions and features.

[1521] Step 3:

[1522] Based on the analysis results, the server compares the product specifications with the patent database to identify similar patents and patents that may infringe.

[1523] Step 4:

[1524] The server uses an AI model to assess the risk of patent infringement and assigns a risk score to each patent.

[1525] Step 5:

[1526] The server generates a patent infringement risk report based on the evaluation results.

[1527] Step 6:

[1528] The server notifies the user of the generated report, enabling the user to review the risks and take appropriate action.

[1529] 4. Real-time patent monitoring

[1530] Step 1:

[1531] The user configures the server with the technical fields and keywords to be monitored.

[1532] Step 2:

[1533] The server monitors the patent database in real time based on configured conditions and collects newly published patent information.

[1534] Step 3:

[1535] The server analyzes the collected patent information, compares it against the set conditions, and identifies important patents.

[1536] Step 4:

[1537] The server will notify users of newly identified patents. Available notification methods include email and dashboard notifications.

[1538] Step 5:

[1539] The server continuously monitors the patent database, ensuring that users always have access to the latest patent information.

[1540] 5. Combination of emotional engines

[1541] Step 1:

[1542] The server collects data on how users interact with the system (e.g., input speed, frequency of operations, screen time, etc.).

[1543] Step 2:

[1544] The server analyzes the collected data using an emotion engine to recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.).

[1545] Step 3:

[1546] The server adjusts notification methods and user interfaces based on the analysis results. For example, if a user is experiencing stress, it changes notifications to gentler and simpler ones.

[1547] Specific example:

[1548] When a user is reviewing the results of a patent infringement risk assessment, the emotion engine determines that the user's stress level is high. The server detects this state and adjusts the notifications and suggestions provided to the user. This makes it possible to improve the user experience.

[1549] These steps enable the system to efficiently manage patent-related tasks, reduce the risk of patent infringement, and quickly access patent information, as well as provide flexible responses that take into account the user's emotional state.

[1550] (Example 2)

[1551] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1552] Conventional patent-related business systems lacked the flexibility to consider the user's emotional state when analyzing patent specifications or assessing patent infringement risks. Furthermore, patent clustering and real-time monitoring were burdensome for users, making efficient management difficult. Therefore, there was a need for a sophisticated system that could simultaneously improve user experience and streamline operations.

[1553] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for analyzing the patent specification with a natural language processing engine and extracting key sections, means for searching for similar patents from an existing patent database, and means for using an AI model to evaluate novelty and inventiveness. This enables the user to efficiently evaluate the novelty and inventiveness of patents, identify infringement risks, manage patents through clustering, and monitor patents in real time. In addition, the user experience is simultaneously improved by a function that uses an emotion analysis engine to determine the user's emotional state and adjust the notification content.

[1554] A "patent specification" is a document submitted for patent application, and includes a detailed description of the invention and drawings.

[1555] A "natural language processing engine" is an algorithm or software that analyzes text data to understand its meaning and extract information.

[1556] A "patent database" is a database that stores information related to patents, and includes data on published patent documents and patent applications.

[1557] An "AI model" is a model used to perform data analysis and prediction using artificial intelligence, and is based on machine learning algorithms.

[1558] An "emotion analysis engine" is an algorithm or software that analyzes a user's emotional state from their input and operation data.

[1559] "Novelty" refers to the characteristic of an invention that is new and not present in existing technology or patent information.

[1560] "Inventive step" refers to the characteristic of an invention that possesses technical features that are a step ahead of existing technology.

[1561] Clustering is the process of dividing data into multiple clusters based on similarity, and is a method for grouping similar data.

[1562] "Labeling" is the process of assigning tags or names to clusters to make them easier to identify.

[1563] "Real-time monitoring" means constantly monitoring the latest information and responding immediately if any changes occur.

[1564] This invention is an advanced intellectual property management system for efficiently performing patent-related tasks, and is a system that can recognize and respond flexibly to the emotional state of the user. This system mainly consists of three main elements: a server, a terminal, and a user.

[1565] Patent automated determination

[1566] The server receives patent specifications uploaded by the user to the system. Next, it analyzes the received patent specifications using a natural language processing (NLP) engine (e.g., OpenAI's GPT-3) to extract key sections (such as the invention title, background, purpose, and detailed description). The server accesses a patent database (e.g., Google Patents) to search for similar existing patents based on the content of the patent specifications. The server uses an AI model (e.g., a Transformer-Based Model) to evaluate novelty and inventiveness and determine the likelihood of patent granting. The evaluation results are generated as a detailed report and notified to the user.

[1567] Specific example:

[1568] When a user uploads a patent specification document for a new battery technology to the system, the server analyzes the specification document and searches for similar existing patents. Based on the results, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[1569] Examples of prompts for a generative AI model:

[1570] "Analyze user-uploaded patent specifications, extract key sections, and evaluate the novelty of the patents by comparing them to existing patent databases."

[1571] Patent clustering

[1572] The server collects patent data related to a specific technology field. Next, it extracts key features (keywords and section summaries) from each patent using an AI algorithm (e.g., BERT). Based on the extracted features, the server clusters the patent data using a clustering algorithm (e.g., K-Means clustering). The clustering results are presented to the user visually.

[1573] Specific example:

[1574] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them through a user interface.

[1575] Examples of prompts for a generative AI model:

[1576] "Collect patent data related to a specific technology field, extract key features, and cluster them. Visualize the clustering results and provide them to the user."

[1577] Patent infringement confirmation

[1578] The server receives product specifications uploaded by the user. It analyzes the product specifications using an NLP engine (e.g., RoBERTa) to extract key functions and features. The server then compares the product specifications with a patent database to identify similar patents and potentially infringing patents. An AI model (e.g., SVM) is used to assess the patent infringement risk, and the assessment results, including assigned risk scores, are generated as a report and notified to the user.

[1579] Specific example:

[1580] A company developing a new smartphone uploads its product specifications to a server to check if it infringes on any other company's patents. The server analyzes the product specifications and compares them with existing patents. If it determines that certain features pose a risk of infringing on another company's patents, it notifies the user of the evaluation result.

[1581] Examples of prompts for a generative AI model:

[1582] "Analyze the product specifications uploaded by the user, compare them with the existing patent database, and assess the patent infringement risk. Output the assessment results as a report."

[1583] Real-time patent monitoring

[1584] Users can configure the server with the technology fields and keywords they want to monitor. Based on these settings, the server monitors the patent database in real time, collecting and analyzing newly published patent information. The server identifies important patents against the configured criteria and notifies the user. This ensures that users always have access to the latest patent information.

[1585] Specific example:

[1586] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information and immediately notifies the user when a relevant patent is published.

[1587] Examples of prompts for a generative AI model:

[1588] "Monitor the patent database in real time based on keywords set by the user, collect and analyze newly published patent information, and notify the user."

[1589] Combination of emotional engines

[1590] The server collects data in real time as the user interacts with the system (e.g., input speed, frequency of operations, screen dwell time, etc.). The server uses an emotion analysis engine (e.g., Microsoft Azure Emotional Intelligence API) to analyze the collected data and recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.). Based on the analysis results, the server adjusts notification methods and the user interface to improve the user experience.

[1591] Specific example:

[1592] When a user reviews their patent infringement risk assessment, if the emotion engine determines the user's stress level is high, the server modifies the notification to be simpler and more user-friendly before providing it to the user. This improves the user experience and facilitates appropriate action.

[1593] This invention streamlines patent-related operations and enables flexible responses that take into account the user's emotional state, thereby improving the user experience.

[1594] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1595] Patent automated determination processing steps

[1596] Step 1:

[1597] Subject: User

[1598] Users upload patent specifications for their inventions to the system. The uploaded patent specifications are sent to the server.

[1599] Input: Patent specification file

[1600] Output: Patent specifications uploaded to the server

[1601] Step 2:

[1602] Subject: Server

[1603] The server analyzes the received patent specification using a natural language processing (NLP) engine (e.g., OpenAI's GPT-3). It extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[1604] Input: Uploaded patent specification

[1605] Output: Extracted main sections

[1606] Step 3:

[1607] Subject: Server

[1608] The server accesses a patent database (e.g., Google Patents) and searches for similar existing patents based on the extracted sections.

[1609] Input: Extracted main sections

[1610] Output: List of similar existing patents

[1611] Step 4:

[1612] Subject: Server

[1613] The server uses an AI model (e.g., a Transformer-Based Model) to evaluate the novelty and inventiveness of the patent specification.

[1614] Input: List of similar existing patents

[1615] Output: Evaluation results for novelty and inventive step

[1616] Step 5:

[1617] Subject: Server

[1618] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report. The generated report is then notified to the user.

[1619] Input: Evaluation results for novelty and inventive step

[1620] Output: Evaluation results report notified to the user.

[1621] Patent clustering processing steps

[1622] Step 1:

[1623] Subject: Server

[1624] The server collects patent data related to a specific technical field from patent databases (e.g., USPTO).

[1625] Input: Search query for patent database

[1626] Output: Set of patent data

[1627] Step 2:

[1628] Subject: Server

[1629] The server extracts key features (keywords and section summaries) from the collected patent data using an AI algorithm (e.g., BERT).

[1630] Input: Set of patent data

[1631] Output: List of extracted key features

[1632] Step 3:

[1633] Subject: Server

[1634] The server clusters the patent data using a clustering algorithm (e.g., K-Means clustering) based on the extracted features.

[1635] Input: List of key features extracted

[1636] Output: Patent data separated by cluster

[1637] Step 4:

[1638] Subject: Server

[1639] The server labels the clusters and visualizes the clustering results for the user.

[1640] Input: Patent data separated by cluster

[1641] Output: Visualized clustering results

[1642] Patent infringement confirmation processing steps

[1643] Step 1:

[1644] Subject: User

[1645] Users upload specifications for products under development to the system. The uploaded product specifications are sent to the server.

[1646] Input: Product specification file

[1647] Output: Product specifications uploaded to the server

[1648] Step 2:

[1649] Subject: Server

[1650] The server analyzes the received product specifications using a natural language processing (NLP) engine to extract key functions and features.

[1651] Input: Uploaded product specifications

[1652] Output: List of extracted key features or characteristics

[1653] Step 3:

[1654] Subject: Server

[1655] The server compares the product specifications with the patent database to identify similar patents and potentially infringing patents.

[1656] Input: List of extracted key features or characteristics

[1657] Output: List of similar patents

[1658] Step 4:

[1659] Subject: Server

[1660] The server uses an AI model (e.g., SVM) to assess the risk of patent infringement and assign a risk score.

[1661] Input: List of similar patents

[1662] Output: Patent infringement risk score and evaluation results

[1663] Step 5:

[1664] Subject: Server

[1665] The server generates a report of the assessment results regarding the risk of infringement and notifies the user.

[1666] Input: Patent infringement risk score and evaluation results

[1667] Output: Evaluation results report notified to the user.

[1668] Processing steps for real-time patent monitoring

[1669] Step 1:

[1670] Subject: User

[1671] Users configure the server with the technology fields and keywords they want to monitor. This information is stored on the server.

[1672] Input: Technical fields and keywords to be monitored

[1673] Output: Monitoring settings stored on the server

[1674] Step 2:

[1675] Subject: Server

[1676] The server monitors the patent database in real time based on saved monitoring settings.

[1677] Input: Monitoring settings stored on the server

[1678] Output: Real-time patent data update information

[1679] Step 3:

[1680] Subject: Server

[1681] The server collects newly published patent information from the patent database and analyzes it using an analysis engine.

[1682] Input: Real-time patent data update information

[1683] Output: Analyzed new patent information

[1684] Step 4:

[1685] Subject: Server

[1686] The server identifies important patents by comparing the analysis results with the configured conditions.

[1687] Input: Analyzed new patent information

[1688] Output: Identified important patent information

[1689] Step 5:

[1690] Subject: Server

[1691] The server notifies users of any important patent information it has identified. This notification is sent via email or through system alerts.

[1692] Input: Identified important patent information

[1693] Output: Patent information alerts sent to the user.

[1694] Processing steps for combining emotion engines

[1695] Step 1:

[1696] Subject: Server

[1697] The server collects data in real time as users interact with the system (e.g., input speed, frequency of operations, screen dwell time, etc.).

[1698] Input: User operation data

[1699] Output: Collected operational data

[1700] Step 2:

[1701] Subject: Server

[1702] The server uses an emotion analysis engine (e.g., Microsoft Azure Emotional Intelligence API) to analyze the collected data and recognize the user's emotional state.

[1703] Input: Collected operational data

[1704] Output: Analysis results of the user's emotional state

[1705] Step 3:

[1706] Subject: Server

[1707] The server adjusts notification methods and user interfaces based on the analysis results. For example, if a user is experiencing stress, it will change notifications to gentler and simpler ones.

[1708] Input: Analysis results of the user's emotional state

[1709] Output: Adjusted notification content and user interface

[1710] Step 4:

[1711] Subject: Server

[1712] The server provides users with configured notifications and user interfaces, and receives feedback. This feedback is then used to make further improvements.

[1713] Input: Adjusted notification content and user interface

[1714] Output: User feedback and its analysis results

[1715] (Application Example 2)

[1716] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1717] Traditional patent-related work involves individually performing tasks such as evaluating the novelty and inventiveness of patents, verifying patent infringement, and real-time monitoring, which is inefficient. Furthermore, it has been difficult to provide feedback that takes into account the user's emotional state, resulting in insufficient improvement of the user experience. In addition, the rapid technological advancements in the field of autonomous vehicles present challenges in managing patent information.

[1718] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for evaluating novelty and inventiveness, means for determining whether or not to grant a patent based on the evaluation results, means for notifying the user of the evaluation results, means for monitoring the patent database in real time based on keywords, means for notifying the user of the monitoring results, means for analyzing the user's emotions, and means for adjusting the notification content based on the emotion analysis. This makes it possible to streamline patent-related work and realize flexible feedback that responds to the user's emotional state. Furthermore, it makes it easier to manage technical information in the field of autonomous vehicles and to always obtain the latest patent information.

[1719] A "patent specification" is a document that describes the details of an invention in order to obtain a patent.

[1720] A "main section" refers to the particularly important part of a patent specification, specifically including the title of the invention, background, purpose, and detailed description.

[1721] A "patent database" refers to a database that stores existing patent information, including published patents, classification information, inventor information, and so on.

[1722] "Novelty" is a criterion used to evaluate whether an invention is different from existing technology.

[1723] "Inventive step" is a criterion used to evaluate how much an invention has advanced compared to existing technology.

[1724] "User" refers to an individual or organization that uses the system to perform patent-related tasks.

[1725] A "keyword" refers to a specific word or phrase used when searching a patent database.

[1726] "Real-time monitoring" is a process that continuously monitors a database based on specific conditions and retrieves the latest information immediately.

[1727] "Emotion analysis" is a technology that analyzes a user's emotional state at a given time based on their input and operation data.

[1728] "Adjusting notification content" refers to changing the information provided and how it is displayed according to the user's emotional state.

[1729] "Patent approval denied" is the result of determining whether or not a particular patent can be obtained.

[1730] Clustering is a technique for grouping multiple data sets based on specific criteria.

[1731] "Labeling" refers to assigning appropriate labels to clustered data.

[1732] "Visualization" refers to displaying data in a visual format, such as graphs or charts.

[1733] A "product specification" is a document that describes the functions, features, and performance of a particular product.

[1734] "Matching" refers to the process of verifying matching items across different datasets.

[1735] "Patent infringement risk" is a criterion for assessing the likelihood that a particular product or technology infringes on existing patents.

[1736] "Report generation" is the process of compiling the results of data analysis into a document.

[1737] This invention provides a system that efficiently handles patent-related tasks and enables flexible responses that take into account the user's emotional state. In particular, it supports the real-time acquisition and management of patent information in the field of autonomous vehicle technology.

[1738] System Configuration

[1739] The server receives patent specifications and uses an NLP (Natural Language Processing) engine to extract key sections. Examples of NLP engines used include spaCy. Based on the extracted sections, the server searches existing patent databases for similar patents. Google Patents API is one example of a patent database that can be used.

[1740] Next, the server uses an AI model to evaluate the novelty and inventiveness of the patent, and based on the evaluation results, it decides whether or not to grant the patent. This evaluation result is generated as a detailed report and notified to the user. The notification to the user is done in a simple text format or using data visualization tools.

[1741] Furthermore, the server provides a function to monitor the patent database in real time based on keywords set by the user. This allows for the immediate acquisition of newly published patent information and notification to the user when relevant patents are published.

[1742] Examples of applications of the emotion engine

[1743] An emotion engine, which analyzes a user's emotional state, collects and analyzes data such as the user's input speed, frequency of operations, and screen dwell time. EmotionAPI is a typical example of an emotion engine used. Based on the results of this emotion analysis, notification content and the user interface are adjusted. For example, if the emotion engine determines that the user is in a high-stress state, the notification content is simplified and presented in a format that is easy for the user to understand.

[1744] Specific example

[1745] For example, if a user uploads a patent specification for a new autonomous vehicle technology, the server analyzes the specification and searches for similar existing patents. An AI model is used to evaluate novelty and inventiveness, and a decision is made regarding patent approval. Simultaneously, the system monitors the patent database in real time based on keywords set by the user and notifies the user when new relevant patents are published. Furthermore, the user's emotional state is analyzed during this process, and if a high stress level is detected, the notification content is simplified.

[1746] Example of a prompt

[1747] Analyze the specifications of new autonomous vehicle technologies, monitor patent databases in real time, and retrieve relevant patent information. Additionally, develop an application that adjusts notification content based on the user's stress level and sends notifications at appropriate times.

[1748] In this way, by implementing the present invention, it is possible to simultaneously improve the efficiency of patent-related work and enhance the user experience.

[1749] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1750] Step 1:

[1751] The server receives a patent specification from the user. The input is the patent specification, and the output is the text data of the patent specification. This text data is then passed to the next processing step.

[1752] Step 2:

[1753] The server analyzes the received patent specification using an NLP engine. The input is the text data of the patent specification, and the output is extracted data of the main sections of the patent specification (such as the title of the invention, background, purpose, and detailed description).

[1754] Step 3:

[1755] The server uses the extracted key section data to search for similar patents in existing patent databases. The input is the data from the key sections of the patent specification, and the output is a list of similar patents.

[1756] Step 4:

[1757] The server uses an AI model to evaluate the novelty and inventiveness of a patent specification. The input is data from the main sections of the patent specification and a list of similar patents, and the output is the evaluation result of novelty and inventiveness (patent likelihood assessment).

[1758] Step 5:

[1759] The server determines whether a patent is granted based on the evaluation results. The input is the evaluation results for novelty and inventive step, and the output is the decision on whether or not to grant the patent.

[1760] Step 6:

[1761] The server notifies the user of the decision regarding permission or denial. The input is the decision regarding permission or denial, and the output is the notification message to the user. The notification message is displayed on the user's interface.

[1762] Step 7:

[1763] The server monitors the patent database in real time based on keywords set by the user. The input is the keywords set by the user, and the output is a list of newly published patent information.

[1764] Step 8:

[1765] The server monitors the system in real time and notifies the user if any new relevant patents are published. The input is a list of newly published patent information, and the output is a notification message to the user.

[1766] Step 9:

[1767] The server collects user interaction data (input speed, frequency of actions, screen time, etc.) and analyzes it using an emotion engine. The input is interaction data, and the output is the user's emotional state (stress level, etc.).

[1768] Step 10:

[1769] The server adjusts notification content based on sentiment analysis. The input is the user's emotional state, and the output is the adjusted notification message. Efforts such as simplifying the notification message are implemented.

[1770] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1771] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1772] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1773] [Fourth Embodiment]

[1774] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1775] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1776] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1777] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1778] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1779] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1780] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1781] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1782] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1783] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1784] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1785] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1786] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1787] This invention is an advanced AI system for efficiently performing patent-related tasks. The system operates as follows:

[1788] 1. Automated patent determination

[1789] Subject: Server

[1790] The server receives the patent specification uploaded by the user. The received specification is analyzed by an NLP (Natural Language Processing) engine, and key sections (such as the invention title, background, purpose, and detailed description) are extracted. The server searches the patent database for similar existing patents and evaluates novelty and inventiveness. Based on the evaluation results, it determines the likelihood of patent granting and notifies the user of the result in a detailed report.

[1791] Specific example:

[1792] A user uploads a specification document for a new invention to the system. This invention includes a new type of battery technology. The server analyzes the specification document and searches for similar existing patents. As a result, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[1793] 2. Patent Clustering

[1794] Subject: Server

[1795] The server collects patent data related to a specific technical field. Next, it extracts key features (keywords and section summaries) from each patent and classifies them using a clustering algorithm. This groups patents based on specific technical fields or application areas. The clustering results are presented to the user visually.

[1796] Specific example:

[1797] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it more efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them to the user.

[1798] 3. Confirmation of patent infringement

[1799] Subject: Server

[1800] The server receives product specifications uploaded by the user and analyzes them. Based on the analysis results, it compares them with a patent database and assesses the potential patent infringement risk. The assessment results are generated in report format and notified to the user.

[1801] Specific example:

[1802] If a company developing a new smartphone wants to check whether its product infringes on another company's patents, the server analyzes the product specifications and compares them with existing patents. Based on the results, it determines if certain features pose a risk of infringing another company's patents. This result is then notified to the user, prompting them to take appropriate action.

[1803] 4. Real-time patent monitoring

[1804] Subject: Server

[1805] Users can configure the server with the technology fields and keywords they wish to monitor. Based on these settings, the server monitors the public patent database in real time. It evaluates newly published patent information, identifies patents important to the user, and notifies them. This process is continuous, ensuring users always stay up-to-date.

[1806] Specific example:

[1807] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[1808] This system streamlines the processes of patent application, patent infringement verification, patent clustering, and real-time patent monitoring, significantly improving a company's patent management operations.

[1809] The following describes the processing flow.

[1810] 1. Automated patent determination

[1811] Step 1:

[1812] The server receives patent specifications uploaded by the user to the system.

[1813] Step 2:

[1814] The server uses an NLP engine to perform text analysis on the received patent specification and extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[1815] Step 3:

[1816] The server accesses the patent database and searches for similar existing patents based on the content of the patent specifications.

[1817] Step 4:

[1818] The server uses an AI model to evaluate the novelty and inventiveness of the patent specification. This includes comparing the extracted features with similar patents.

[1819] Step 5:

[1820] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report of the results.

[1821] Step 6:

[1822] The server notifies the user of the generated evaluation results report. The user can view the report via a dashboard or email.

[1823] 2. Patent Clustering

[1824] Step 1:

[1825] The server collects patent data related to a specific technology field from a database.

[1826] Step 2:

[1827] The server uses an AI algorithm to extract key features (keywords and section summaries) from each patent.

[1828] Step 3:

[1829] The server clusters the patent data using a clustering algorithm (e.g., k-means clustering) based on the extracted features.

[1830] Step 4:

[1831] The server automatically assigns appropriate labels to each cluster. These labels include those based on specific technologies and functions.

[1832] Step 5:

[1833] The server visually displays clustering results on a dashboard, allowing users to easily understand and manage their patent portfolio.

[1834] 3. Confirmation of patent infringement

[1835] Step 1:

[1836] The server receives product specifications uploaded by the user to the system.

[1837] Step 2:

[1838] The server uses an NLP engine to perform text analysis on the received product specifications and extracts key functions and features.

[1839] Step 3:

[1840] The server compares the product specifications with the patent database to identify similar patents and potentially infringing patents.

[1841] Step 4:

[1842] The server uses an AI model to assess the risk of patent infringement and assigns a risk score to each patent.

[1843] Step 5:

[1844] The server generates a patent infringement risk report based on the evaluation results.

[1845] Step 6:

[1846] The server notifies the user of the generated report, enabling the user to review the risks and take appropriate action.

[1847] 4. Real-time patent monitoring

[1848] Step 1:

[1849] The user configures the server with the technical fields and keywords to be monitored.

[1850] Step 2:

[1851] The server monitors the patent database in real time based on configured conditions and collects newly published patent information.

[1852] Step 3:

[1853] The server analyzes the collected patent information, compares it against the set conditions, and identifies important patents.

[1854] Step 4:

[1855] The server will notify users of newly identified patents. Available notification methods include email and dashboard notifications.

[1856] Step 5:

[1857] The server continuously monitors the patent database, ensuring that users always have access to the latest patent information.

[1858] By following the steps outlined above, this system efficiently manages patent-related tasks, reducing the risk of patent infringement and enabling rapid access to patent information.

[1859] (Example 1)

[1860] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1861] Traditional patent-related work, including evaluating patent applications, analyzing patent clusters, and confirming patent infringement risks, was manual, time-consuming, labor-intensive, and lacked accuracy and consistency. Furthermore, evaluating novelty and inventiveness required specialized knowledge, making it difficult for companies to manage patents efficiently. In addition, real-time monitoring of patent information and access to the latest updates were insufficient.

[1862] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1863] In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for analyzing the specifications using a natural language processing engine, means for evaluating novelty and inventiveness using a machine learning model, and means for notifying the user of the generated report via a user interface or email. This enables the automation and streamlining of the patent application process.

[1864] The server also includes means for collecting patent data, means for extracting the features of the patents, means for clustering the patents, means for labeling the clusters, means for visualizing the clustering results, means for classifying the patents using a clustering algorithm, and means for providing the visualized clustering results through a user interface. This facilitates the organization and analysis of patent portfolios, enabling the formulation of patent strategies and appropriate decision-making.

[1865] Furthermore, the server includes means for receiving product specifications, means for analyzing the product specifications, means for comparing the product specifications with a patent database, means for evaluating patent infringement risk using a generated AI model, means for generating the evaluation results as a report, means for notifying the user of the evaluation results, and means for evaluating patent grantability based on customizable evaluation criteria. This enables rapid evaluation of patent infringement risk and appropriate response during product development.

[1866] A "patent specification" is a document submitted when filing a patent application, and it describes the details of the invention.

[1867] A "patent database" is a database that stores existing patent information and is used for searching and matching patents.

[1868] A "natural language processing engine" is a software module that provides technology for processing and analyzing human language using computers.

[1869] A "machine learning model" is an algorithm that learns from past data and uses that knowledge to make predictions and classifications.

[1870] A "user interface" is an interface through which a user interacts with a computer system, and includes graphical displays and methods of operation.

[1871] "Patent clustering" is a method of grouping and classifying related patents.

[1872] A "clustering algorithm" is an algorithm that automatically classifies data into meaningful groups (clusters).

[1873] A "generative AI model" is a model used by artificial intelligence (AI) to automatically generate and analyze data, and it includes pre-trained algorithms specialized for a specific purpose.

[1874] "Evaluation criteria" refer to the standards and indicators used when evaluating patents or products.

[1875] A "report" is a document that details the results of analysis and evaluation.

[1876] "Patent infringement risk" refers to the possibility that a product or technology infringes on existing patent rights.

[1877] This invention is an AI system for efficiently performing patent-related tasks. This system provides functions for automatic patent identification, patent clustering, patent infringement verification, and real-time patent monitoring.

[1878] Patent automated determination

[1879] In the automated patent evaluation function, the user uploads a patent specification, and the server receives it. The server uses an NLP (Natural Language Processing) engine to analyze the patent specification and extract key sections (such as the invention title, background, purpose, and detailed description). Next, the server compares it with an existing patent database to evaluate the novelty and inventiveness of the patent. The evaluation results are notified to the user as an automatically generated report. This allows the user to check the likelihood of patent granting with high accuracy.

[1880] Specific example:

[1881] When a user uploads a patent specification for a new battery technology to the system, the server analyzes the specification, compares it with existing patents, and evaluates the novelty and inventiveness of the invention. The evaluation results are then notified to the user in the form of a report.

[1882] Example of a prompt:

[1883] "Please evaluate this patent specification from the perspectives of novelty and inventive step."

[1884] "Please analyze the patent specifications for the new battery technology and create a report evaluating its patentability."

[1885] Patent clustering

[1886] In the patent clustering function, the server collects patent data related to a specific technical field. The server extracts the characteristics of each patent (keywords and section summaries) and classifies the patents using a clustering algorithm. The clustering results are visualized, allowing users to efficiently manage their patent portfolios.

[1887] Specific example:

[1888] A company with a large amount of patent data related to electric vehicle technology uses the system to organize its patents into clusters. The server uses a clustering algorithm to create clusters such as battery technology, motor control, and charging infrastructure, and provides them in a visualized format on the user interface.

[1889] Example of a prompt:

[1890] "Please cluster patents related to electric vehicle technology."

[1891] "Classify patents in a specific technical field and visualize the clustered results."

[1892] Patent infringement confirmation

[1893] In the patent infringement verification function, the user uploads the specifications of the product under development to the server. The server analyzes the product specifications, compares them with an existing patent database, and assesses the potential patent infringement risk. The assessment results are generated as a report and notified to the user.

[1894] Specific example:

[1895] Companies developing new smartphones use the system to check for patent infringement risks. The server analyzes the product specifications, compares them with existing patents, assesses the infringement risk, and notifies the user of the results in a report.

[1896] Example of a prompt:

[1897] "Please check if these product specifications infringe on any other company's patent rights."

[1898] "Please analyze the specifications of the new smartphone and create a report assessing the patent infringement risk."

[1899] Real-time patent monitoring

[1900] In the real-time patent monitoring function, users configure the server with the technical fields and keywords to be monitored. The server monitors the patent database in real time based on the configured conditions and evaluates newly published patent information. When relevant patents are published, the user is immediately notified.

[1901] Specific example:

[1902] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[1903] Example of a prompt:

[1904] "Monitor new patents related to artificial intelligence technology in real time."

[1905] "Please collect newly published patent information based on specific keywords and notify us of any relevant information."

[1906] This system automates the patent application process, enabling efficient evaluation of patent novelty and inventiveness, patent clustering, patent infringement risk assessment, and real-time patent monitoring. This allows companies to significantly improve their patent management operations.

[1907] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1908] Step 1:

[1909] The user uploads the patent specification.

[1910] Input: The user selects a patent specification file (PDF or Word document) and clicks the upload button.

[1911] Specific actions:

[1912] The user selects a file and presses the upload button.

[1913] The server receives the uploaded files and temporarily stores them in the database.

[1914] Output: The patent specifications are saved on the server.

[1915] Step 2:

[1916] The server passes the patent specification to the NLP engine, which analyzes it and extracts the key sections.

[1917] Input: Saved patent specification file.

[1918] Specific actions:

[1919] The server calls a natural language processing engine (such as SpaCy or BERT) to analyze the document.

[1920] Using NLU (Natural Language Understanding) technology, sections such as "Title of Invention," "Background," "Purpose," and "Detailed Description" are extracted from the patent specification.

[1921] Output: The extracted section information is generated as structured data in JSON format.

[1922] Step 3:

[1923] The server searches existing patent databases based on the information in the patent specifications to find similar patents.

[1924] Input: Section information in JSON format.

[1925] Specific actions:

[1926] The server issues SQL or NoSQL queries to a patent database (e.g., the USPTO database).

[1927] Patents with high similarity are listed in a ranked format.

[1928] The search results are scored and sorted so that the most similar patents appear at the top.

[1929] Output: List and score of similar patents.

[1930] Step 4:

[1931] The server evaluates novelty and inventiveness.

[1932] Input: List and score of similar patents.

[1933] Specific actions:

[1934] The server uses machine learning models (e.g., random forests or support vector machines) to calculate evaluation scores for novelty and inventiveness.

[1935] The evaluation will take into account patent classification, cited documents, technical field, etc.

[1936] Output: Evaluation results for novelty and inventive step.

[1937] Step 5:

[1938] The server determines the likelihood of patent approval based on the evaluation results.

[1939] Input: Evaluation results for novelty and inventive step.

[1940] Specific actions:

[1941] The evaluation score is analyzed, and the likelihood of patent approval is assessed on a three-point scale: "high," "medium," and "low."

[1942] The results after the assessment will be compiled into a report.

[1943] Output: A detailed report on the likelihood of patent approval.

[1944] Step 6:

[1945] The server notifies the user of the evaluation results via the user interface or email.

[1946] Input: A report on the likelihood of patent approval.

[1947] Specific actions:

[1948] Generates reports in HTML and PDF formats.

[1949] The report will be sent to the user's email address or available for download on the dashboard.

[1950] Output: Reports accessible to the user.

[1951] Step 7:

[1952] The user sends a request to the server to cluster additional patent data.

[1953] Input: Conditions and keywords for the patent data you want to cluster.

[1954] Specific actions:

[1955] The user specifies the patent's technical field and keywords and submits a clustering request.

[1956] The server collects patent data based on specified conditions.

[1957] Output: Patent dataset for clustering.

[1958] Step 8:

[1959] The server uses a patent clustering algorithm to classify and visualize patents.

[1960] Input: Patent dataset for clustering.

[1961] Specific actions:

[1962] The server extracts features from the patent data and applies clustering algorithms (e.g., K-means, hierarchical clustering).

[1963] Assign an appropriate label to each classified cluster.

[1964] Visualize the clustering results and display them on the user interface.

[1965] Output: Visualized patent clustering results that can be viewed by the user.

[1966] Step 9:

[1967] The user uploads product specifications and initiates a process to assess patent infringement risk.

[1968] Input: Product specifications provided by the user.

[1969] Specific actions:

[1970] The user uploads the product specifications.

[1971] The server analyzes the uploaded product specifications.

[1972] Output: Analyzed product specification data.

[1973] Step 10:

[1974] The server analyzes the product specifications and compares them with the patent database.

[1975] Input: Analyzed product specification data.

[1976] Specific actions:

[1977] The server compares product specification data with the patent database.

[1978] The patent infringement risk is assessed based on the matching results.

[1979] Output: Patent infringement risk assessment results.

[1980] Step 11:

[1981] The server generates a report of the patent infringement risk assessment results and notifies the user.

[1982] Input: Patent infringement risk assessment results.

[1983] Specific actions:

[1984] Generate reports and provide them in HTML or PDF format.

[1985] The report will be sent to the user's email address or displayed on the dashboard.

[1986] Output: Report on the results of the patent infringement risk assessment.

[1987] In this way, the system efficiently supports patent-related tasks and provides advanced functions that enable users to manage patents quickly and accurately.

[1988] (Application Example 1)

[1989] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1990] Patent-related work requires a wide range of tasks, including evaluating novelty and inventiveness, confirming patent infringement risk, and clustering and visualizing patents, making it difficult to perform each process efficiently. Furthermore, risk assessment to avoid infringing on competitors' patents during new product development is crucial, but this also consumes significant time and resources. To address these challenges, a system that can streamline patent-related work in a centralized manner is needed.

[1991] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1992] In this invention, the server includes means for receiving patent specifications, means for analyzing the patent specifications and extracting key sections, means for searching for similar patents from an existing patent database, means for receiving product specifications and evaluating the risk of patent infringement based on those specifications, means for determining whether or not to grant a patent based on the evaluation results, means for notifying the user of the evaluation results, and means for evaluating the risk from the content of similar patents and providing it to the user visually. This enables centralized processing of patent-related tasks, allows for rapid evaluation of novelty and inventiveness, automatic determination of patent infringement risk, and visual management of patents using clustering, thereby significantly improving the efficiency of a company's product development process and patent management.

[1993] A "patent specification" is a document submitted when filing a patent application, and it includes the title of the invention, background, purpose, detailed description, and other information that describes the technical content of the invention.

[1994] "Analysis" is the process of analyzing input data and extracting specific information or patterns.

[1995] "Key sections" refer to important components within a patent specification, such as sections containing the invention's title, background, and detailed description.

[1996] A "patent database" is a data store where existing patent information is accumulated, and it is used for searching and matching patents.

[1997] "Similar patents" refer to existing patents that have similar technical content to the entered patent specifications or product specifications.

[1998] "Novelty" refers to the characteristic that indicates whether an invention differs from existing technology and is being publicly disclosed for the first time.

[1999] "Inventive step" is a characteristic that indicates the extent to which an invention has advanced compared to existing technology, in other words, whether or not there is a technological advancement.

[2000] "Evaluation results" refer to conclusions or opinions obtained based on an assessment of novelty, inventiveness, or patent infringement risk.

[2001] A "product specification" is a document that contains detailed descriptions of a product's design, functions, performance, and other aspects.

[2002] "Patent infringement risk" refers to the assessment of the likelihood that a particular product or technology infringes on a patent held by another company.

[2003] Clustering is a method of grouping data based on specific criteria to consolidate similar patents.

[2004] "Labeling" refers to assigning appropriate tags or names to each cluster.

[2005] "Visualization" refers to methods of displaying analysis results and evaluation results in the form of graphs, charts, and other diagrams to make them easier to understand.

[2006] "User" refers to a person who uses the system to perform patent-related tasks.

[2007] This invention provides an advanced AI system for efficiently performing patent-related tasks. The system integrates multiple functions, including receiving patent specifications, receiving product specifications, searching patent databases, risk assessment, clustering, and notifications.

[2008] server

[2009] The server receives patent specifications and product specifications from users, analyzes them, and extracts key sections. It uses a natural language processing (NLP) engine to perform document analysis and automatically extract each section of the patent specification (such as the invention title, background, and detailed description). It also queries patent databases to search for similar patents and evaluates novelty and inventive step.

[2010] The server further analyzes the product specifications and assesses the patent infringement risk based on those specifications. This risk assessment is achieved by cross-referencing with existing patent information. The assessment results are communicated to the user in a detailed report format.

[2011] A clustering algorithm is used to classify the collected patent data and group patents based on specific technical fields or application areas. The clustering results are provided to the user using a visualization tool.

[2012] terminal

[2013] The terminal functions as an interface for users to input patent specifications and product specifications and receive evaluation results and reports from the server. Users can access the system using devices such as smartphones, smart glasses, and head-mounted displays.

[2014] User

[2015] Users can upload patent specifications or product specifications for new inventions to the system and receive evaluation results. Based on these evaluation results, users can decide whether or not to file a patent application or assess the development risks of new products.

[2016] The server uses the following specific hardware and software to perform the processing:

[2017] Hardware: Servers and cloud infrastructure (e.g., AWS, Google Cloud)

[2018] Software: Python, requests library, BeautifulSoup, regular expressions, scikit-learn library, natural language processing (NLP) engine

[2019] Specific example

[2020] Here are some specific examples:

[2021] Development of a new smartphone

[2022] This process involves users uploading specifications for their new smartphones to a system, which then evaluates the patent infringement risk. The system analyzes the specifications, searches for similar patents, and identifies specific patents as high-risk. This evaluation is then provided to the user in a detailed report, allowing them to visually confirm which parts were deemed risky.

[2023] Example of a prompt

[2024] "Please upload the specifications for your new product. We will use them to assess the patent infringement risk. Please clearly state any keywords or phrases you will be using."

[2025] "To use the patent infringement risk assessment system, you upload your product specifications, and the system searches the patent database based on the section information extracted using the NLP engine to generate a risk assessment report. It is necessary to specify certain keywords or phrases."

[2026] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[2027] Step 1:

[2028] The user uploads the specifications (text file) for a new product to the system using a terminal. Once the user uploads the specifications, the terminal sends the data to the server. The input data is the specifications (text format) for the new product. The output data is the product specifications sent to the server.

[2029] Step 2:

[2030] The server analyzes the received product specifications. Specifically, it uses a natural language processing (NLP) engine to extract the main sections of the product specifications (e.g., technical specifications, feature descriptions, etc.). The input data is the uploaded product specifications. The output data is the extracted section information (in text format).

[2031] Step 3:

[2032] The server searches the patent database based on the extracted section information. Here, it uses the patent database API to query for similar patents. The input data is the extracted section information. The output data is similar patent information obtained from the patent database.

[2033] Step 4:

[2034] The server uses the acquired similar patent information to assess patent infringement risk. A specific algorithm is used to calculate the likelihood that a new product infringes existing patents. The input data is similar patent information. The output data is the patent infringement risk assessment result.

[2035] Step 5:

[2036] The server generates a detailed report based on the risk assessment results. This report includes detailed information about the patent infringement risk and its rationale. The report is generated in a visually easy-to-understand format. The input data is the patent infringement risk assessment results. The output data is a detailed risk assessment report (PDF or HTML format).

[2037] Step 6:

[2038] The server notifies the user of the generated risk assessment report. Specifically, it sends the report to the terminal, allowing the user to view it through the interface. The input data is the detailed risk assessment report. The output data is the report information notified to the user.

[2039] Step 7:

[2040] The user uses a terminal to view a risk assessment report sent from the server. This allows the user to check the patent infringement risk of the new product and take necessary actions. The input data is the notified risk assessment report. The output data is the report information viewed by the user.

[2041] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[2042] This invention combines an advanced AI system for efficiently performing patent-related tasks with an emotion engine that recognizes user emotions. The system operates as follows:

[2043] 1. Automated patent determination

[2044] Subject: Server

[2045] The server receives patent specifications uploaded by the user to the system. Next, it analyzes the received patent specifications using an NLP engine and extracts key sections (such as the invention title, background, purpose, and detailed description). The server accesses a patent database and searches for similar existing patents based on the content of the patent specifications. The server uses an AI model to evaluate novelty and inventiveness and determine the likelihood of patent granting. The evaluation results are generated as a detailed report and notified to the user.

[2046] Specific example:

[2047] A user uploads a specification document for a new invention to the system. This invention includes a new type of battery technology. The server analyzes the specification document and searches for similar existing patents. As a result, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[2048] 2. Patent Clustering

[2049] Subject: Server

[2050] The server collects patent data related to a specific technology field. Next, it uses an AI algorithm to extract key features (keywords and section summaries) from each patent. Based on the extracted features, the server clusters the patent data using a clustering algorithm. The clustering results are presented to the user visually, making it easier to manage the patent portfolio.

[2051] Specific example:

[2052] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them to the user.

[2053] 3. Confirmation of patent infringement

[2054] Subject: Server

[2055] The server receives product specifications uploaded by the user. It analyzes the product specifications using an NLP engine to extract key functions and features. The server then compares the product specifications with a patent database to identify similar patents and potentially infringing patents. An AI model is used to assess the patent infringement risk, and the assessment results, including assigned risk scores, are generated as a report and notified to the user.

[2056] Specific example:

[2057] A company developing a new smartphone uses a server to analyze its product specifications to check if it infringes on other companies' patents. The server then compares the specifications with existing patents and determines if certain features pose a risk of infringing on others' patents. The evaluation results are then communicated to the user to encourage appropriate action.

[2058] 4. Real-time patent monitoring

[2059] Subject: Server

[2060] Users can configure the server with the technology fields and keywords they wish to monitor. Based on these settings, the server monitors the patent database in real time, collecting and analyzing newly published patent information. The server identifies important patents against the configured criteria and notifies the user. This ensures that users always have access to the latest patent information.

[2061] Specific example:

[2062] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information daily and immediately notifies users when relevant patents are published.

[2063] 5. Combination of emotional engines

[2064] In addition to handling patent-related tasks, this system is equipped with an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's input and actions during operation to recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.).

[2065] Subject: Server

[2066] The server collects data on how users interact with the system (e.g., input speed, frequency of actions, screen time, etc.) and analyzes it using an emotion engine. Based on the analysis results, it adjusts appropriate notification methods and user interfaces. For example, if a user is experiencing stress, notifications can be changed to gentler and simpler ones.

[2067] Specific example:

[2068] When a user is reviewing the results of a patent infringement risk assessment, the emotion engine determines that the user's stress level is high. The server detects this state and adjusts the notifications and suggestions provided to the user. This makes it possible to improve the user experience.

[2069] This system not only streamlines the processes of patent application, patent infringement verification, patent clustering, and real-time patent monitoring, but also enables flexible responses that take into account the user's emotional state.

[2070] The following describes the processing flow.

[2071] 1. Automated patent determination

[2072] Step 1:

[2073] The server receives patent specifications uploaded by the user to the system.

[2074] Step 2:

[2075] The server analyzes the text of the patent specification using an NLP engine and extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[2076] Step 3:

[2077] The server accesses the patent database and searches for similar existing patents based on the content of the patent specifications.

[2078] Step 4:

[2079] The server uses an AI model to evaluate novelty and inventiveness. This includes comparing the extracted features with similar patents.

[2080] Step 5:

[2081] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report of the results.

[2082] Step 6:

[2083] The server notifies the user of the generated evaluation results report. The user can view the report via a dashboard or email.

[2084] 2. Patent Clustering

[2085] Step 1:

[2086] The server collects patent data related to a specific technology field from a database.

[2087] Step 2:

[2088] The server uses an AI algorithm to extract key features (keywords and section summaries) from each patent.

[2089] Step 3:

[2090] The server clusters the patent data using a clustering algorithm (e.g., k-means clustering) based on the extracted features.

[2091] Step 4:

[2092] The server automatically assigns appropriate labels to each cluster. These labels include those based on specific technologies and functions.

[2093] Step 5:

[2094] The server visually displays the clustering results on a dashboard, allowing users to easily manage their patent portfolios.

[2095] 3. Confirmation of patent infringement

[2096] Step 1:

[2097] The server receives product specifications uploaded by the user to the system.

[2098] Step 2:

[2099] The server analyzes the product specifications using an NLP engine and extracts key functions and features.

[2100] Step 3:

[2101] Based on the analysis results, the server compares the product specifications with the patent database to identify similar patents and patents that may infringe.

[2102] Step 4:

[2103] The server uses an AI model to assess the risk of patent infringement and assigns a risk score to each patent.

[2104] Step 5:

[2105] The server generates a patent infringement risk report based on the evaluation results.

[2106] Step 6:

[2107] The server notifies the user of the generated report, enabling the user to review the risks and take appropriate action.

[2108] 4. Real-time patent monitoring

[2109] Step 1:

[2110] The user configures the server with the technical fields and keywords to be monitored.

[2111] Step 2:

[2112] The server monitors the patent database in real time based on configured conditions and collects newly published patent information.

[2113] Step 3:

[2114] The server analyzes the collected patent information, compares it against the set conditions, and identifies important patents.

[2115] Step 4:

[2116] The server will notify users of newly identified patents. Available notification methods include email and dashboard notifications.

[2117] Step 5:

[2118] The server continuously monitors the patent database, ensuring that users always have access to the latest patent information.

[2119] 5. Combination of emotional engines

[2120] Step 1:

[2121] The server collects data on how users interact with the system (e.g., input speed, frequency of operations, screen time, etc.).

[2122] Step 2:

[2123] The server analyzes the collected data using an emotion engine to recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.).

[2124] Step 3:

[2125] The server adjusts notification methods and user interfaces based on the analysis results. For example, if a user is experiencing stress, it changes notifications to gentler and simpler ones.

[2126] Specific example:

[2127] When a user is reviewing the results of a patent infringement risk assessment, the emotion engine determines that the user's stress level is high. The server detects this state and adjusts the notifications and suggestions provided to the user. This makes it possible to improve the user experience.

[2128] These steps enable the system to efficiently manage patent-related tasks, reduce the risk of patent infringement, and quickly access patent information, as well as provide flexible responses that take into account the user's emotional state.

[2129] (Example 2)

[2130] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2131] Conventional patent-related business systems lacked the flexibility to consider the user's emotional state when analyzing patent specifications or assessing patent infringement risks. Furthermore, patent clustering and real-time monitoring were burdensome for users, making efficient management difficult. Therefore, there was a need for a sophisticated system that could simultaneously improve user experience and streamline operations.

[2132] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for analyzing the patent specification with a natural language processing engine and extracting key sections, means for searching for similar patents from an existing patent database, and means for using an AI model to evaluate novelty and inventiveness. This enables the user to efficiently evaluate the novelty and inventiveness of patents, identify infringement risks, manage patents through clustering, and monitor patents in real time. In addition, the user experience is simultaneously improved by a function that uses an emotion analysis engine to determine the user's emotional state and adjust the notification content.

[2133] A "patent specification" is a document submitted for patent application, and includes a detailed description of the invention and drawings.

[2134] A "natural language processing engine" is an algorithm or software that analyzes text data to understand its meaning and extract information.

[2135] A "patent database" is a database that stores information related to patents, and includes data on published patent documents and patent applications.

[2136] An "AI model" is a model used to perform data analysis and prediction using artificial intelligence, and is based on machine learning algorithms.

[2137] An "emotion analysis engine" is an algorithm or software that analyzes a user's emotional state from their input and operation data.

[2138] "Novelty" refers to the characteristic of an invention that is new and not present in existing technology or patent information.

[2139] "Inventive step" refers to the characteristic of an invention that possesses technical features that are a step ahead of existing technology.

[2140] Clustering is the process of dividing data into multiple clusters based on similarity, and is a method for grouping similar data.

[2141] "Labeling" is the process of assigning tags or names to clusters to make them easier to identify.

[2142] "Real-time monitoring" means constantly monitoring the latest information and responding immediately if any changes occur.

[2143] This invention is an advanced intellectual property management system for efficiently performing patent-related tasks, and is a system that can recognize and respond flexibly to the emotional state of the user. This system mainly consists of three main elements: a server, a terminal, and a user.

[2144] Patent automated determination

[2145] The server receives patent specifications uploaded by the user to the system. Next, it analyzes the received patent specifications using a natural language processing (NLP) engine (e.g., OpenAI's GPT-3) to extract key sections (such as the invention title, background, purpose, and detailed description). The server accesses a patent database (e.g., Google Patents) to search for similar existing patents based on the content of the patent specifications. The server uses an AI model (e.g., a Transformer-Based Model) to evaluate novelty and inventiveness and determine the likelihood of patent granting. The evaluation results are generated as a detailed report and notified to the user.

[2146] Specific example:

[2147] When a user uploads a patent specification document for a new battery technology to the system, the server analyzes the specification document and searches for similar existing patents. Based on the results, it determines that the invention is novel and has a high probability of being patentable, and notifies the user of the evaluation result.

[2148] Examples of prompts for a generative AI model:

[2149] "Analyze user-uploaded patent specifications, extract key sections, and evaluate the novelty of the patents by comparing them to existing patent databases."

[2150] Patent clustering

[2151] The server collects patent data related to a specific technology field. Next, it extracts key features (keywords and section summaries) from each patent using an AI algorithm (e.g., BERT). Based on the extracted features, the server clusters the patent data using a clustering algorithm (e.g., K-Means clustering). The clustering results are presented to the user visually.

[2152] Specific example:

[2153] A company with numerous patents related to electric vehicle technology wants to organize its patent portfolio into clusters to manage it efficiently. The server creates clusters for battery technology, motor control, charging infrastructure, etc., and provides them through a user interface.

[2154] Examples of prompts for a generative AI model:

[2155] "Collect patent data related to a specific technology field, extract key features, and cluster them. Visualize the clustering results and provide them to the user."

[2156] Patent infringement confirmation

[2157] The server receives product specifications uploaded by the user. It analyzes the product specifications using an NLP engine (e.g., RoBERTa) to extract key functions and features. The server then compares the product specifications with a patent database to identify similar patents and potentially infringing patents. An AI model (e.g., SVM) is used to assess the patent infringement risk, and the assessment results, including assigned risk scores, are generated as a report and notified to the user.

[2158] Specific example:

[2159] A company developing a new smartphone uploads its product specifications to a server to check if it infringes on any other company's patents. The server analyzes the product specifications and compares them with existing patents. If it determines that certain features pose a risk of infringing on another company's patents, it notifies the user of the evaluation result.

[2160] Examples of prompts for a generative AI model:

[2161] "Analyze the product specifications uploaded by the user, compare them with the existing patent database, and assess the patent infringement risk. Output the assessment results as a report."

[2162] Real-time patent monitoring

[2163] Users can configure the server with the technology fields and keywords they want to monitor. Based on these settings, the server monitors the patent database in real time, collecting and analyzing newly published patent information. The server identifies important patents against the configured criteria and notifies the user. This ensures that users always have access to the latest patent information.

[2164] Specific example:

[2165] Companies that want to stay informed about patents related to artificial intelligence technology can set specific keywords and register them as targets for monitoring. The server collects and analyzes newly published patent information and immediately notifies the user when a relevant patent is published.

[2166] Examples of prompts for a generative AI model:

[2167] "Monitor the patent database in real time based on keywords set by the user, collect and analyze newly published patent information, and notify the user."

[2168] Combination of emotional engines

[2169] The server collects data in real time as the user interacts with the system (e.g., input speed, frequency of operations, screen dwell time, etc.). The server uses an emotion analysis engine (e.g., Microsoft Azure Emotional Intelligence API) to analyze the collected data and recognize the user's emotional state (e.g., excitement, anger, anxiety, etc.). Based on the analysis results, the server adjusts notification methods and the user interface to improve the user experience.

[2170] Specific example:

[2171] When a user reviews their patent infringement risk assessment, if the emotion engine determines the user's stress level is high, the server modifies the notification to be simpler and more user-friendly before providing it to the user. This improves the user experience and facilitates appropriate action.

[2172] This invention streamlines patent-related operations and enables flexible responses that take into account the user's emotional state, thereby improving the user experience.

[2173] The flow of the specific processing in Example 2 will be explained using Figure 13.

[2174] Patent automated determination processing steps

[2175] Step 1:

[2176] Subject: User

[2177] Users upload patent specifications for their inventions to the system. The uploaded patent specifications are sent to the server.

[2178] Input: Patent specification file

[2179] Output: Patent specifications uploaded to the server

[2180] Step 2:

[2181] Subject: Server

[2182] The server analyzes the received patent specification using a natural language processing (NLP) engine (e.g., OpenAI's GPT-3). It extracts key sections (such as the title of the invention, background, purpose, and detailed description).

[2183] Input: Uploaded patent specification

[2184] Output: Extracted main sections

[2185] Step 3:

[2186] Subject: Server

[2187] The server accesses a patent database (e.g., Google Patents) and searches for similar existing patents based on the extracted sections.

[2188] Input: Extracted main sections

[2189] Output: List of similar existing patents

[2190] Step 4:

[2191] Subject: Server

[2192] The server uses an AI model (e.g., a Transformer-Based Model) to evaluate the novelty and inventiveness of the patent specification.

[2193] Input: List of similar existing patents

[2194] Output: Evaluation results for novelty and inventive step

[2195] Step 5:

[2196] Subject: Server

[2197] The server determines the likelihood of patent granting based on the evaluation results and generates a detailed report. The generated report is then notified to the user.

[2198] Input: Evaluation results for novelty and inventive step

[2199] Output: Evaluation results report notified to the user.

[2200] Patent clustering processing steps

[2201] Step 1:

[2202] Subject: Server

[2203] The server collects patent data related to a specific technical field from patent databases (e.g., USPTO).

[2204] Input: Search query for patent database

[2205] Output: Set of patent data

[2206] Step 2:

[2207] Subject: Server

[2208] The server extracts key features (keywords and section summaries) from the collected patent data using an AI algorithm (e.g., BERT).

[2209] Input: Set of patent data

[2210] Output: List of extracted key features

[2211] Step 3:

[2212] Subject: Server

[2213] The server clusters the patent data using a clustering algorithm (e.g., K-Means clustering) based on the extracted features.

[2214] Input: List of key features extracted

[2215] Output: Patent data separated by cluster

[2216] Step 4:

[2217] Subject: Server

[2218] The server labels the clusters and visualizes the clustering results for the user.

[2219] Input: Patent data separated by cluster

[2220] Output: Visualized clustering results

[2221] Patent infringement confirmation processing steps

[2222] Step 1:

[2223] Subject: User

[2224] Users upload specifications for products under development to the system. The uploaded product specifications are sent to the server.

[2225] Input: Product specification file

[2226] Output: Product specifications uploaded to the server

[2227] Step 2:

[2228] Subject: Server

[2229] The server analyzes the received product specifications using a natural language processing (NLP) engine to extract key functions and features.

[2230] Input: Uploaded product specifications

[2231] Output: List of extracted key features or characteristics

[2232] Step 3:

[2233] Subject: Server

[2234] The server compares the product specifications with the patent database to identify similar patents and poten...

Claims

1. Means for receiving patent specifications, A means for analyzing the aforementioned patent specification and extracting key sections, A means of searching for similar patents from existing patent databases, Means for evaluating novelty and inventiveness, A means of determining whether or not to grant special permission based on the evaluation results, A means for notifying the user of the evaluation results, A system that includes this.

2. Means for collecting patent data, Means for extracting the features of the aforementioned patent, A means for clustering patents, A method for labeling clusters, A means of visualizing clustering results, The system according to claim 1, further comprising:

3. A means of receiving product specifications, A means for analyzing the aforementioned product specifications, A means for comparing a patent database with product specifications, Methods for assessing patent infringement risk, A means of generating evaluation results as a report, A means for notifying the user of the evaluation results, The system according to claim 1, further comprising:

4. A means of accepting settings for the target of monitoring, A means of monitoring the patent database in real time, A means of evaluating and notifying newly published patents, Means for continuously monitoring the patent database, The system according to claim 1, further comprising:

Images