Automatic detection of treatment relationships for patent retrieval

ABSTRACT

We devise a method for automatically detecting treatment relationships using lexico-syntactic patterns and its application to medical-oriented patent retrieval. This process for detecting treatment relationships involves finding lexico-syntactic patterns that are highly indicative of treatment relationships and also producing classification rules for those patterns.

This treatment relationship detection process is then used in a system to find treatment relationships based on a user query in a medical patent source. The query will consist of terms that the user wants to find in the subject or object of a treatment relationship. This is of great interest to both patent examiners and patent applicants as they search for prior art. Through the use of classification rules, this system was able to achieve a precision of 85.81% on a set of 20 test queries.

REFERENCES

• 1: Roxana Girju, Automatic detection of causal relations for Question Answering, Proceedings of the ACL 2003 workshop on Multilingual summarization and question answering, p.76-83, July 11, 2003, Sapporo, Japan [doi>10.3115/1119312.1119322]
• 2: Leah S. Larkey, A patent search and classification system, Proceedings of the fourth ACM conference on Digital libraries, p.179-187, August 11-14, 1999, Berkeley, California, United States [doi>10.1145/313238.313304]
• 3: Leah S. Larkey , Margaret E. Connell , Jamie Callan, Collection selection and results merging with topically organized U.S. patents and TREC data, Proceedings of the ninth international conference on Information and knowledge management, p.282-289, November 06-11, 2000, McLean, Virginia, United States [doi>10.1145/354756.354830]
• 4: Sougata Mukherjea , Bhuvan Bamba, BioPatentMiner: an information retrieval system for biomedical patents, Proceedings of the Thirtieth international conference on Very large data bases, p.1066-1077, August 31-September 03, 2004, Toronto, Canada
• 5: USPTO, Performance and Accountability Report Fiscal Year 2007, Page: 109, 2007.
• 6: Leo Wanner, Ricardo Baeza-Yates, Sören Brügmann, Joan Codina, Barrou Diallo, Enric Escorsa, Mark Giereth, Yiannis Kompatsiaris, Symeon Papadopoulos, Emanuele Pianta, Gemma Piella, Ingo Puhlmann, Gautam Rao, Martin Rotard, Pia Schoester, Luciano Serafini, Vasiliki Zervaki, "Towards Content-Oriented Patent Document Processing", World Patent Information, Elsevier, Vol. 30(1), Page: 21--33, March 2008.
• 7: Tianhao Wu, Lars E. Holzman, William M. Pottenger, Daniel J. Phelps, "A Supervised Learning Algorithm for Information Extraction from Textual Data", In the proceeding of the workshop on Text Mining, Third SIAM International Conference on Data Mining, 2003.
• 8: Xuan-Hieu Phan. Jtextpro: A java-based text processing toolkit.
• 9: C. Fellbaum. Wordnet: an electronic lexical database. Mit Pr, 1998.
• 10: O. Bodenreider and O. Journals. The unified medical language system (umls): integrating biomedical terminology. Nucleic Acids Research, 32(90001):267--270, 2004.
• 11: DA Lindberg, BL Humphreys, and AT McCray. The unified medical language system. In Methods of Information in Medicine, volume 32, pages 281--91, 1993.
• 12: J. Ross Quinlan, C4.5: programs for machine learning, Morgan Kaufmann Publishers Inc., San Francisco, CA, 1993

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.1 Content Analysis and Indexing
          Subjects: Linguistic processing

General Terms:
Experimentation, Languages

Keywords:
lexico-syntactic patterns, patent retrieval, treatment relationships