Extracting Characteristic Terms from Patent Documents

  • Kaito Takano Seikei University
  • Miryu Tanaka Seikei University
  • Hiroyuki Sakai Seikei University
  • Ryozo Kitajima Tokyo Polytechnic University
  • Takahisa Ota Showa Denko K.K.
  • Chinatsu Tanabe Showa Denko K.K.
  • Hiroki Sakaji The University of Tokyo
Keywords: characteristic terms, natural language processing, patent documents, text mining


We propose a method to automatically extract “characteristic terms” from several patent documents belonging to a certain technical field. The characteristic terms extracted by the proposed method are useful for technology trend analysis. For example, in the case of patent documents relating to organic electroluminescence, the characteristic term expresses the characteristic of the technology shown in the patents, e.g., “発光効率” (hakkoukouritu: luminous efficiency) or “輝度” (kido: brightness). The proposed method was evaluated and good results were obtained.


H. Nonaka, A. Kobayashi, H. Sakaji, Y. Suzuki, H. Sakai, and S. Masuyama, “Extraction of the Effect and the Technology Terms from a Patent Document,” Journal of Japan Industrial Management Associastion, vol.63, no.2E, 2012, pp.105-111.

H. Sakaji, H. Nonaka, H. Sakai, and S. Masuyama, “Cross-Bootstrapping: An Automatic Extraction Method of Solution-Effect Expressions from Patent Documents,” The IEICE transactions on information and systems, vol.J93-D, no.6, 2010, pp.742-755. (in Japanese)

H. Sakai, H. Nonaka and S. Masuyama, “Extraction of information on the technical effect from a patent document,” The Japanese Society for Artificial Intelligence, vol.24, no.6, 2009, pp.531-540. (in Japanese)

H. Li, F. Xu, and H. Uszkoreit, “TechWatchTool: Innovation and Trend Monitoring,” Recent Advances in Natural Language Processing, 2011, pp. 660-665.

M. Okamoto, Z. Shan, and R. Orihara, “Applying Information Extraction for Patent Structure Analysis,” the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2017, pp. 989-992.

S. Suzuki and H. Takatsuka, “Extraction of Keywords of Novelties From Patent Claims,” Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, 2016, pp. 1192-1200.

M. Vazquez, M. Krallinger, F. Leitner, and A, Valencia, “Text mining for drugs and chemical compounds: methods, tools and applications,” Molecular Informatics, Vol. 30, No. 6-7, 2011, pp. 506–519.

A. Ekbal and S. Bandyopadhyay, “Named entity recognition using support vector machine: A language independent approach,” International Journal of Electrical, Computer, and Systems Engineering, Vol. 4, No. 2, 2010, pp. 155-170.

L. Luo, Z. Yang, P. Yang, Z. Yin, L. Wang, H. Lin, and J. Wang, “An attention-based bilstm-crf approach to document-level chemical named entity recognition,” Bioinformatics, Vol. 34, No. 8, 2018, pp. 1381–1388.

T. Mikolov, K. Chen, G. Corrado, and J. Dean, “Efficient Estimation of Word Representations in Vector Space,” arXiv preprint arXiv:1301.3781, 2013.

S. Kitamori, H. Sakai and H. Sakaji, “Extraction of sentences concerning business performance forecast and economic forecast from summaries of financial statements by deep learning,” IEEE Symposium on Computational Intelligence for Financial Engineering & Economics, 2017, pp.67-73.