KLIPA: A Knowledge Graph and LLM-Driven QA Framework for IP Analysis

2025-09-09

AI Summary

Traditional intellectual property management relies on manual retrieval and rigid keyword matching, failing to uncover latent semantic associations within patent data and thereby hindering strategic decision-making. Method: This paper proposes a knowledge graph (KG)–large language model (LLM) collaborative intelligent agent framework. It employs a dynamic strategy selection mechanism to adaptively parse user queries and synergistically integrates KG-based structured reasoning with retrieval-augmented generation (RAG) for contextual understanding. Contribution/Results: To our knowledge, this is the first approach to jointly optimize multi-hop semantic reasoning and structured relational mining. Experiments on a real-world patent corpus demonstrate significant improvements in knowledge extraction accuracy and novel association discovery capability, while substantially reducing dependence on domain experts. The framework enables efficient, scalable, and automated patent analysis.

Technology Category

• Data Mining & Knowledge Management: Intelligent Query Processing
• Knowledge Representation and Reasoning: Knowledge Acquisition
• Natural Language Processing: Information Extraction

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Agentic search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Effectively managing intellectual property is a significant challenge. Traditional methods for patent analysis depend on labor-intensive manual searches and rigid keyword matching. These approaches are often inefficient and struggle to reveal the complex relationships hidden within large patent datasets, hindering strategic decision-making. To overcome these limitations, we introduce KLIPA, a novel framework that leverages a knowledge graph and a large language model (LLM) to significantly advance patent analysis. Our approach integrates three key components: a structured knowledge graph to map explicit relationships between patents, a retrieval-augmented generation(RAG) system to uncover contextual connections, and an intelligent agent that dynamically determines the optimal strategy for resolving user queries. We validated KLIPA on a comprehensive, real-world patent database, where it demonstrated substantial improvements in knowledge extraction, discovery of novel connections, and overall operational efficiency. This combination of technologies enhances retrieval accuracy, reduces reliance on domain experts, and provides a scalable, automated solution for any organization managing intellectual property, including technology corporations and legal firms, allowing them to better navigate the complexities of strategic innovation and competitive intelligence.

Problem

• Overcoming inefficiency in manual patent analysis methods
• Revealing complex relationships within large patent datasets
• Reducing reliance on domain experts for IP management

Innovation

• Knowledge graph maps explicit patent relationships
• RAG system uncovers contextual patent connections
• Intelligent agent dynamically optimizes query resolution