Prior Art Search Acceleration

Method for providing patent document analysis results through a neural network model that has trained patent data and patent determination data from a patent office or court. The method involves receiving target patent information and displaying a user interface with visualizations like colored cores distinguishing key components and similarity assessment results. The interface also shows mappings of similar documents' core components onto the target patent. This visualization helps users analyze similarities and differences between patents using the neural network analysis.

Source

Efficiently finding similar documents to a given reference document in a large document repository using natural language processing techniques. The method involves analyzing both the text and metadata of documents to identify similarities. It involves generating vectors representing words and n-grams from the documents, applying weighting to emphasize important words, and then comparing vector similarities between the reference document and other documents. This allows identifying documents with similar content even in different languages or domains.

Source

Identifying relevant documents to a query using machine learning models that achieves high accuracy while maintaining efficiency. The system generates relevance scores for documents using only retrieved token vectors of candidate documents rather than all token vectors of the documents. This approach reduces computational complexity and memory requirements compared to evaluating token-level interactions. The system employs a query encoder neural network and a document encoder neural network that are trained jointly or separately, with the query encoder and document encoder architectures being similar or distinct. The system leverages SIMD intrinsics to speed up computing similarity measures.

Source

Novelty checking method for inventions that uses a combination of natural language processing (NLP) and heuristic algorithms to accurately assess novelty by narrowing and ranking prior art searches. The method involves identifying an invention's details, defining a search space of prior art, refining it based on NLP comparison and heuristics, rating matches, and using that to assess the invention's novelty.

Source

A system and method for extracting accurate data from large corpora using vector search and labeling. The system involves a server and a company knowledge base containing vectorized patent data. The server receives keywords, vectors them, searches the knowledge base, labels the found results, vectors the labels, and stores them to improve future searches. This iterative labeling and vectorization step trains the knowledge base to better match similar data.

Source

Filtering search results to improve efficiency and accuracy in document review processes like legal discovery. The method involves initially showing small search term results containing primary search terms to the user to quickly filter out irrelevant ones. This helps avoid reviewing entire documents. The user indicates relevance or removal for each result. The method also extracts secondary search terms from names/contact info, emails, and external sources. This allows finding relevant documents without reviewing every one.

Source

Efficient and accurate search through large document corpora using a sparse matrix-based scoring algorithm that allows for fast, relevant and size-aware search. The algorithm scores documents for relevance to a query by multiplying a sparse document matrix (containing term contribution values), a query vector (containing query term values), and an optional word coverage factor vector (containing document coverage values). The documents are then sorted by scores and the best ones returned in response to the query. This allows finding most relevant documents for a query quickly and accurately, as well as providing sizing information on how many documents match a query term.

Source

In recent years, AI-related technology referred to as RAG (Retrieval Augmented Generation) (Lewis, 2020) gained a lot of attention. the RAG-approach, custom sources information are used seed knowledge obtained from LLM (Large Language Model), thus forming an approach which solves issue adapting cope with external information. Using RAGscenario, various processing use cases can be implemented, such AI-based document management, AI-enhanced web search, online service support, etc. This paper outlines main components RAG-workflow chunking and embedding input documents, well similarity -based user query processing. The is illustrated via Python implementation validate procedure simple example multi-topic document. Experimental results discussed showing feasibility this approach, illustrating need for further research enhancements, by RAPTOR concept (Sarthi, 2024).

Source

Large language model (LLM) tools are transforming the way evidence is retrieved by converting natural prompts into quick, synthesized outputs. These platforms significantly reduce time required for literature searches, making them more accessible to users unfamiliar with formal search strategies. A close evaluation of four prominent platformsUndermind.ai, Scite.ai, Consensus.app, and OpenEvidencehighlights both notable advantages ongoing limitations. Undermind Consensus utilize extensive Semantic Scholar database over 200 million records, Scite enhances results Smart Citations that indicate supportive or opposing references, OpenEvidence applies a medically-focused LLM trained on licensed sources, including complete NEJM archive. Despite their benefits, key limitations persist: opaque algorithms, inconsistent responses identical queries, paywalls sign-up barriers, incomplete recall may compromise systematic reviews. To support critical appraisal, we outline essential information-retrieval metricsincluding recall, precision, F1-score, mean average specificityand prov... Read More

Source

Patent management system with tools to help analyze, organize, and search patent portfolios. The system allows quick and relevant patent analysis using features like claim mapping, concept organization, similarity indexing, and expanded search results. It provides tools to efficiently manage and analyze patent portfolios by leveraging automated techniques like mining, mapping, and indexing to extract insights from patent claims and texts. The system also enables expanded search results, concept organization, and claim mapping to help with quick and relevant patent analysis.

Source

Efficiently searching large databases by sharding the index using multiple dimensions to reduce search times and costs. The sharding is done on three dimensions: record type, year, and name. Instead of randomly distributing records across shards, they are organized by surname within each shard. This allows searches for specific names to only query a subset of shards, instead of all of them like in random sharding. The optimization is based on a greedy algorithm to balance shard allocation between dimensions.

Source

The exponential growth of academic literature has rendered traditional keyword-based search engines increasingly inadequate for scholars seeking contextually relevant research. This study presents the design and implementation an intelligent semantic engine tailored journals, integrating state-of-the-art Artificial Intelligence (AI) Natural Language Processing (NLP) techniques. proposed system leverages sentence transformer models (all-mpnet-base-v2) embeddings, enabling vector-based similarity searches, alongside spaCy tokenization entity recognition to enhance syntactic understanding. An ontology-based matching mechanism further aligns user queries with domain-specific research topics, while fuzzy regular expressions improve error tolerance numeric filtering (e.g., CiteScore, Impact Factor). architecture combines these NLP layers Elasticsearch's hybrid capabilities process rank peer-reviewed journal metadata sourced from Scopus DOAJ. A modular FastAPI-based backend ensures scalability responsiveness, a lightweight frontend interface facilitates interactive input. contributes novel ... Read More

Source

Artificial intelligence (AI) system for analyzing patent documents to provide insights into technology development trends, competitive intelligence, and patent analysis. The system uses AI techniques like semantic analysis and relationship analysis to automatically identify the most critical patents and documents in a collection, determine what they reveal, and provide summaries and visualizations. It consolidates metadata like citations, litigation, and expiration dates to provide statistics like citation indices and influence factors. The AI also generates alerts, flags, and indicators based on patent quality, relevance, and expiration. The system aims to automate tasks like competitive analysis, prior art search, patentability assessment, and freedom to operate analysis using AI instead of manual review.

Source

AI-assisted legal research system that enables more effective and efficient legal research by using AI techniques like large language models (LLMs) to generate summaries and analyze legal documents based on user inputs. The system receives natural language search queries instead of keywords and synthesizes responses using LLMs. It also ranks and quantifies relevance of documents to the query. This provides more accurate and conversational search results compared to traditional keyword-based search engines.

Source

Cognitively identifying potential issues with claims using AI to analyze claim characteristics and compare them to similar claims to determine if an alert is warranted. The AI model identifies similarities and differences between a new claim and past claims to flag potential issues. If the new claim has similar characteristics to ones with issues, it generates an alert notification to the user.

Source

A platform for analyzing intellectual property portfolios of entities by identifying similarities between portfolios and clustering IP assets based on technical aspects. The platform allows users to seed searches based on technical fields, competitor portfolios, etc. to find IP assets similar to their own. It then clusters the assets at varying levels of granularity and generates visual representations of the clusters. This helps users efficiently analyze and compare portfolios, identify gaps and saturation, and assess exposure.

Source

Searching and analyzing extremely large numbers of documents efficiently, with tools for complex query building, visualization, and publishing results. The system uses a search engine to index all fields of the documents instead of a database for faster searching. The system organizes documents into collections based on format, and allows querying across collections. The interface allows building complex queries, viewing results, flagging, and directly accessing documents. It also generates visualizations and lets sharing/publishing results. This scales for millions of documents by using a search engine instead of a database.

Source

Automated analysis of patent claims to help evaluate relative breadth, identify corresponding products, and find related patents. The system analyzes claims using techniques like stemming and normalization to extract features like unique word counts. It generates claim profiles based on these features and compares them to determine relative breadth. It also searches for products and related patents based on identified elements in the claims.

Source

Similar patent document retrieval is an essential task that reduces the scope of claimants searches, and numerous studies have attempted to provide automated search services. Recently, Retrieval-Augmented Generation (RAG) based on generative language models has emerged as excellent method for accessing utilizing knowledge environments. RAG-based services offer enhanced ranking performance AI by providing similar queries. However, achieving optimal similarity-based in remains a challenging task, methods similarity do not adequately address characteristics documents. Unlike general retrieval, documents must take into account prior art relationships. To this issue, we propose PAI-NET, deep neural network computing similarities incorporating expert We demonstrate our proposed outperforms current state-of-the-art classification tasks through semantic distance evaluation USPD KPRIS datasets. PAI-NET presents candidates, demonstrating superior improvement 15% over methods.

Source

Efficiently comparing and viewing differences in large sets of documents to speed up document review processes. The method involves identifying shared text with variations across multiple documents, sorting documents into groups based on those variations, and generating a single "shared difference" document that highlights areas of variation between sections of shared text. This allows quickly comparing and navigating through many documents at once instead of reading them one by one.

Source

Large language model (LLM)-based information retrieval for large datasets using indices to improve query response times and scalability. The method involves creating an index of the input text files containing vector embeddings of the text in voronoi cells. When a query comes in, the query embedding is compared to the vectors in the voronoi cell with the closest match to generate a response without needing to search the full text. This allows using a subset of embeddings for querying instead of the entire text. The indices can also be merged and partitioned to further reduce the search space.

Source

Searching across multiple ontologic knowledge graphs using an overlay graph to improve efficiency and accuracy. The method involves generating overlay graphs that map entities and relations from multiple source graphs. When a search request comes in, an appropriate overlay graph is selected based on the entity and relation. The search is then executed on the overlay graph, which translates the request into queries for the source graphs. Results are received from the sources and used to respond to the search.

Source

Dynamic indexing and searching technique to improve search efficiency and reduce search times. The search space is split into a primary subset and a secondary subset. When a query is received, only the primary subset is initially searched. If the results are insufficient, a supplemental search is performed on the secondary subset using the same query. This asynchronous secondary search occurs after the primary search and presents additional results. Items from the secondary search can also be moved to the primary subset for future queries. This reduces search times by prioritizing the most relevant subset for initial searches.

Source

Method for searching for relevant patent documents using machine learning and AI techniques to improve the effectiveness and efficiency of patent searching. The method involves parsing claim elements, selecting keywords, identifying synonyms, forming search strings, ranking documents, determining citations, and mapping elements. This allows targeted searching of databases using user-defined elements instead of keywords, improving recall and reducing false positives compared to keyword-based searching.

Source

Patent document retrieval method that uses metadata and semantic conversion to improve efficiency and accuracy compared to keyword search. The method involves selecting a retrieval method: 1) similarity search using metadata fields like invention title, patent type, etc., 2) keyword search in metadata, or 3) semantic search using a summary field. For method 1, similarity to metadata fields filters doc IDs. For method 2, keywords filter doc IDs from metadata. For method 3, semantic similarity to summary field filters top 5 doc IDs.

Source

Patent management system that provides tools to help in decision making at each stage of the patenting process. The system allows quick claim scope determination, prior art analysis, portfolio management, and other tasks. It uses mapping, mining, and analytics techniques to organize, search, and visualize patent concepts. The system also provides features like expanded search results, claim similarity indexing, and highlighting key terms across claims. It aims to enable efficient patent research, comparison, and analysis.

Source

Assisting patent retrieval using user selection instructions to iteratively refine and expand search criteria. The method involves selecting a patent document, extracting classification data and keywords, generating new search conditions based on the selected patent's classification and keywords, then searching for more documents matching those conditions. This iterative refinement allows refining and expanding the search beyond the initial search criteria.

Source

A hybrid AI system for patent claims analysis that combines machine learning and user input to accurately search and analyze patent documents. The system allows users to initiate searches based on reference documents, narrow the search focus, and select sections. It uses ML to transform the document and claims into embeddings representing their meaning. Element-by-element searches find matches based on these embeddings. Confidence scores rank the results. The system provides feedback to improve future searches.

Source

Semi-automatic patent claims analysis that improves the speed and accuracy of a computer or server when performing a patent search, particularly for infringement analysis. The analysis involves using preconditioning rules to determine relevant portions of a reference document, a query is submitted to find a set of patents that are similar to the reference document, a set of matching patents from the query are received and claim construction rules are used to determine how a meaning of claim elements of the patents will be ascertained in further analysis, Patent infringement rules and an artificial intelligence (AI) system are used to search the relevant portions of the reference document with the claim constructed claim elements of the patents returned from the search, the set of patents are ranked based on assigned confidence scores indicating a degree to which respective claim elements match the relevant portions of the reference document, and the results are presented to a client device.

Source

A method for assisting users in searching patents. The method involves enabling users to select and search within patent documents based on key criteria like classification codes and related tags. The system retrieves patent documents matching the initial search criteria. Users can then select one or more patents of interest and extract additional criteria like classification codes from them. The system generates a refined search using the extracted criteria to find more relevant patents. This iterative process allows users to navigate and refine patent searches using criteria specific to patents.

Source

Patent search system that helps users find relevant patents by assisting with search criteria selection. The system receives an initial search criteria, then when a user selects a patent, it extracts keywords from that patent. The system generates a refined search with those keywords. This iterative process of selecting patents and extracting keywords allows focusing the search.

Source

Generating search strings from seed documents or collection of seed documents. The search strings are generated based on the content of the seed documents, the number of claims analyzed, the number of associated documents, and the number of search strings generated from the search strings.

Source

A system to automatically create patent application templates based on prior art references to reduce the time and expertise needed to draft patent applications. The system parses prior art documents to identify sections related to the invention. It then generates a template patent application by removing claims and sections from the prior art that are not relevant to the invention. The template can be edited by the drafter to insert the novel features of the invention.

Source

A platform for conducting patent searches and analyzing patent value using a combination of AI, machine learning, and blockchain technology. The platform leverages natural language processing to improve patent search relevance by analyzing similarities between claim limitations and prior art specifications. It also uses AI algorithms to rank prior art based on relevance to specific claim elements. The platform further provides tools for highlighting, weighting, and customizing search parameters.

Source

Searching, filtering, and analyzing large numbers of patent-related documents to find documents of interest to a user. It provides user interface tools for filtering and analyzing patent-related documents to find documents of interest. It also allows searching for specific patent drawings.

Source

Data-driven method to efficiently screen a large number of patent documents for relevancy to a target subject matter using a triage approach. The method involves mapping key concepts from claims of the identified patent pool to quickly screen and rule out irrelevant patents. The concepts marked as "definitely not in target subject matter" are used to eliminate patents without manually reviewing each one.

Source

A method and device for searching similar patents using a claim tree and word sets to improve accuracy and efficiency compared to manual searching. The method involves generating a claim tree diagram from the patent claims, with each node representing a claim. Multiple word sets are generated from the patent text. Search formulas are created for each word set and used to search the patent database. The search results are mapped back to the claim tree nodes. Similarity between each patent and the original patent is calculated using a weighted model. The patent with highest similarity per node is retained. This set of similar patents is returned. The claim tree and word set approach helps find key retrieval information and improves patent search accuracy.

Source

Patent search system that improves efficiency by using data analysis to identify key patents in a technology field. The system has a patent database, data processing unit, and screening management unit. The processing unit analyzes patent data to find important metrics like citation counts and inventor affiliations. The screening management unit uses these metrics to automatically identify key patents in a search result, instead of manually reviewing each patent. This reduces time and improves efficiency compared to traditional patent searching.

Source

Computer systems and methods for searching, filtering, and analyzing large numbers of patent-related documents to find documents of interest to a user. The system utilizes a comprehensive patent-based parts index derived from the US patent corpus to identify and interrelate part expressions (i.e., part names) in patents. This allows users to search for drawings containing specific parts, search for patents with similar drawings to a selected drawing, and determine non-literal support for claim terms in patent specifications using conceptually related variants.

Source

Optimizing patent literature searches by taking a user's search formula and expanding it to a higher-quality search by adding relevant search terms and operators. The system receives a user's search formula, classifies it based on operators, adds additional relevant search terms to each group, combines the groups with operators to create a final optimized search formula, and provides that to the user. This expands and improves the user's initial search.

Source

Semi-automatic patent infringement analysis using hybrid artificial intelligence. The system narrows the focus of patent search results by identifying relevant portions of a reference document. It then uses artificial intelligence to compare claim elements of patents to those relevant sections. This provides more accurate infringement analysis compared to using machine learning alone. The AI determines word meanings, searches element by element, and outputs matches with confidence scores.

Source

Patent search system that improves efficiency and reduces user effort compared to traditional methods. The system uses stem extraction, data vectorization, and user feedback to enhance patent retrieval. It creates a stem index to extract stem words from patents, counts stem frequencies, and combines them into data vectors. Searching involves matching user input against vector statistics, displaying results, and allowing user evaluation to bias future searches. The system also removes common words to reduce stem overlap.

Source

A patent search method that enables more efficient and comprehensive patent searches by leveraging citations and citation relationships between patent documents. The method involves searching for a patent inventor's theses, identifying the theses cited by others, and then searching for patents related to those cited theses. This additional step using citation connections allows finding patent documents that might be missed by keyword searches alone. It also provides insights into technology transfer, infringement risks, competitive technologies, and avoidance designs.

Source

Retrieval method and device for more accurate and efficient searching of patents using keywords and classification numbers. The method involves obtaining a query platform, extracting classification numbers from the platform, retrieving patent documents using both the extracted classification numbers and keywords. This allows searching patents based on both specific technical areas and keywords, improving accuracy compared to just keywords.

Source

Method and device for denoising a patent search database to improve accuracy and completeness of patent search results. The method involves using patent classification numbers to denoise a search database. It involves obtaining a patent document, getting its classification number, checking if it matches another document's classification, denoising if not, then repeating with keywords. This iterative process involving keywords and classification numbers improves search results accuracy compared to just keywords.

Source

Method and device to improve accuracy of patent searching by denoising and verifying search results based on competing product information. It involves removing patents with similar applicants to prevent interference. Keywords are used to validate denoised patents to check for relevant content.

Source

Continuous iterative patent search method to find related patents beyond just keyword matches. It starts with a keyword to get a patent database, then finds a second keyword from that database with a relationship to the first. If the relationship satisfies a condition, it gets a second database using the second keyword. Finally, it combines the first and second databases to expand the search results. This iterative process finds patents with similar research directions to the initial keyword.

Source

A patent retrieval method using keywords and multi-platform classification numbers to improve accuracy compared to traditional patent searching. The method involves obtaining competitive product information, retrieving patents for those products, extracting classification numbers and keywords from the patents, and using them to search for more relevant patents. This leverages existing patent data to determine better classification numbers and keywords for more accurate patent retrieval.

Source

Patent search method that improves accuracy and comprehensiveness of patent searches by integrating historical data from patent databases. The method involves obtaining a target patent, determining a keyword from it, getting the search term for that keyword from the database, calculating impact factors for the keyword and term, combining them, checking if above a threshold, and using the combined keyword to find the target patent. This leverages correlation between keywords and terms to find more relevant patents.

Source