Patent Data Analysis Method Based on Knowledge Graph and Similarity Measurement

• Published in: Chinese Control Conference pp. 3484 - 3489
• Main Authors: Zhenxing, Zhan, Dongmei, Fu, Zhouke, Dawei, Zhang
• Format: Conference Proceeding
• Language: English
• Published: Technical Committee on Control Theory, Chinese Association of Automation 24.07.2023
• Subjects: Automation; Corrosion; Corrosion inhibitors; Data analysis; Graph distance; Knowledge graph; Knowledge graphs; Patent analysis; Patent intelligence mining; Semantics; Similarity measure
• ISSN: 1934-1768
• DOI: 10.23919/CCC58697.2023.10241165

Patent analysis is a technique that utilizes data analysis methods to explore potential content or relationships within patents. As one of the important techniques in patent analysis, patent similarity measurement is a subject of concern and research to determine what specific methods can comprehensively and accurately utilize patent information for similarity measurement. This paper proposes a knowledge graph-based patent similarity measurement method that simultaneously considers bibliographic information, content semantic information, global importance of knowledge, and prior knowledge in the field of patents. Finally, this paper uses this method to analyze and mine the characteristics of patent subclusters in the field of corrosion inhibitors, and compares the analysis results with those of the traditional SAO structure method to verify the effectiveness of the proposed method.