Landscape Mapping Innovations for IP

Enterprise innovation service method and system using standardized knowledge graphs to improve efficiency and effectiveness of innovation services for enterprises. The method involves extracting, integrating, and visualizing enterprise innovation data from sources like patents, research papers, and standards to create a standardized knowledge graph. This graph represents the innovation landscape and trends of the enterprise's industry. It provides accurate, targeted, and intelligent innovation services by leveraging the graph's structured connections and visualizations.

Source

Generating an intellectual property (IP) data information map system to visualize and analyze IP data in a more engaging and useful way. The system collects IP data, classifies it into patent, trademark, and copyright categories, filters out useless information, and generates interactive maps that show IP activity by region, industry, applicant, etc. This allows real-time updates, provides context and insights, and improves efficiency compared to static tabular data.

Source

General information interaction method for a technology transfer office that enables efficient retrieval, viewing, evaluation, analysis, and recommendation of patents to support technology transfer. The method involves forming a data processing instruction based on user input, querying the required patent information, processing it to obtain specific data, and then interacting with the user using the processed results. The processing steps include feature extraction, filtering, and analysis techniques like time sequence analysis, semantic clustering, and fuzzy matching. It leverages techniques like patent network modeling, bidirectional value evaluation, and character string filtering to provide enhanced patent analysis capabilities.

Source

Patent evaluation method and system using natural language processing and complex network algorithms to objectively evaluate patent quality by comparing patents with technologies in the global same industry. The method involves collecting patent documents, generating technical points and clusters from the documents using NLP, and comparing patent clusters to assess the depth and breadth of technology, predict patent lifespan, and judge R&D strength. It leverages NLP to aggregate patents into clusters based on their technical attributes and then compares cluster similarity to evaluate patent quality.

Source

Automatically analyzing competitor patent portfolios in a customized way for a given user by classifying the patents based on the user's specific technology areas. This allows extracting patterns in the competitor portfolios from the user's perspective. The method involves determining elements within inputted patents that are relevant to the user's organization. These elements are used to classify the patents in a user-specific way. Then, analysis is performed on the classified patents to identify patterns and discrepancies between the user's portfolio and the competitors'. This provides insight into competitor technology strategies and gaps. The classified and analyzed patents can also be filtered and distributed to specific recipients within the user's organization.

Source

An industrialized technology direction navigation analysis system to help organizations find promising research directions by analyzing intellectual property data. The system provides modules to analyze patents, trademarks, copyrights, circuit designs, and intellectual property enforcement cases. It aims to provide comprehensive insights into intellectual property trends, emerging technologies, infringement risks, and related information to guide research and development strategies.

Source

Intellectual property data platform that provides enhanced intellectual property management services by processing, categorizing, and analyzing intellectual property data using natural language processing techniques. The platform collects intellectual property information from sources, preprocesses and analyzes the text, measures similarity, learns categories, and stores labeled data. It then offers services like schedule management, prior art search, and reports using the processed data. The platform customizes patent attorney recommendations and tech trend insights based on user fields.

Source

Intellectual property (IP) data platform that provides IP management automation services using collected and processed IP data. The platform collects IP information from various sources, processes it using natural language techniques like parsing, tokenization, and text mining, and labels the processed data based on similarity measurements. The labeled data is stored for IP schedule management, prior art search, and report generation services. The platform also classifies IP agents and trends by field using labeled data.

Source

Automatic generation of patent citation landscapes for comprehensive analysis of patent family connections and citations to identify problematic patents and analyze portfolio value. The system identifies related patents in a family, and searches and aggregates backward citations for analysis.

Source

Automated method to generate citation landscapes of patent families for analyzing patent portfolios. It involves selecting a patent, finding the family members, searching for backward citations of those members, aggregating them, and presenting the backwards citations in a portfolio for review. This allows real-time citation landscape analysis of related patents to understand references and potential issues like prior art or invalidity concerns.

Source

An AI system for tracking and identifying innovations in specific categories across multiple data sources using natural language processing. The system receives user queries with keywords related to a category, extracts relevant information from sources using NLP, and identifies innovations based on qualifying terms like "new," "improved," etc. It can search public websites and databases as well as private sources.

Source

Machine learning-based prediction of future innovation at the company level using big data and predictive analysis techniques. The method involves collecting patent, financial, news, and social media data for predetermined companies over a period, feature extraction, and prediction using machine learning models like logistic regression, naive Bayes, neural networks, support vector machines, and deep belief networks. The goal is to predict if a company will innovate in the future based on historical data.

Source

Intellectual property (IP) business recommendation system using enterprise knowledge maps to improve accuracy and efficiency of finding target customers for IP services like patent filings and trademark registrations. The system generates customer portraits based on enterprise data, policy information, and IP records. It matches customer features against screening rules to filter potentials. This provides more complete and precise customer lists compared to traditional methods.

Source

Intellectual property (IP) analysis platform that uses machine learning and data analytics to provide comprehensive IP portfolio analysis. It generates visual representations of IP portfolios of entities to help analyze and compare them. The platform analyzes IP assets using metrics like breadth, opportunity, and exposure scores. It combines these scores into a comprehensive metric. The visualization shows the scores in a circular graph where size represents the number of defendants, color indicates non-practicing entity status, and position shows time since second filing.

Source

A method and system to predict technology trends using natural language processing and time series analysis on large volumes of scientific papers. The method involves extracting technical features from the papers using NLP, creating a high-dimensional feature space, and then analyzing the relationships between technologies over time using time series techniques to predict their development trajectories. This allows objective analysis and prediction of technology trends without relying on expert opinions.

Source

Method, device, computer equipment and medium for panoramic analysis of intellectual property rights to obtain comprehensive patent information of an enterprise conveniently. It involves gathering individual patent data, collecting industry patent data using key keywords, cleaning the data, and generating a panoramic report using multiple analysis planes. This provides a holistic view of an enterprise's patent portfolio compared to the industry.

Source

Automatic classification of patents into industries using a transductive learning method that leverages patent text and classification information to minimize manual labeling. The method involves determining a set of target patents to classify, generating vectors representing the abstract, claims, and description sections of each patent, concatenating them into a hierarchical vector, computing similarities between patent vectors using an adjacency matrix, and iteratively updating the matrix to converge on accurate industry classifications. The method uses patent text structure, IPC codes, and hierarchical vector representation to extract full patent context for classification.

Source

Automatically reviewing and analyzing vendor assurance reports like SOC reports using natural language processing to extract exceptions, generate summaries, identify trends, and present visualized insights. The system trains a natural language engine on report data to detect exceptions, generate summaries, and identify topics. It extracts exceptions from new reports, summarizes them, finds trends across reports, and presents visualized insights.

Source

Intellectual property (IP) landscaping platform that uses user-driven searches to identify similar IP assets and generate visual representations of clusters of IP assets. The platform allows users to seed searches based on entities, publications, products, etc., and identify similar entities, assets, and technologies. It then clusters assets at varying levels of granularity and generates interactive maps to visualize the IP landscapes. The platform also assesses exposure levels by mapping IP assets to entities' funding.

Source

Electronically mining intellectual property by using text mining techniques to analyze enterprise documents and industry trends to identify correlations that lead to potential intellectual property. The process involves identifying an industry trend, extracting keywords from enterprise documents, determining relevance scores, optimizing weighting schemes, and ranking documents based on intellectual property potential.

Source