Medical documents processing for summary generation and keywords highlighting based on natural language processing and ontology graph descriptor approach

Alexander Dudko, Tatiana Endrjukaite, Yasushi Kiyoki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper a new method of data retrieval from free text documents in medical domain is proposed. Presented approach gives the document summary and highlights important keywords in the text to support further analysis of multiple medical documents. The document is processed with natural language processing techniques to find medical keywords and assign them to concepts in the medical ontology. These concepts contribute to higher levels in the hierarchy and build the document descriptor as a graph with concepts in the nodes and corresponding relevance points. The descriptor is used to generate the summary in a form of tree. Finally, we highlight the most important keywords in the original text. Presented experiments demonstrate the proposed approach, which successfully summarizes and highlights meaningful medical information.

Original languageEnglish
Title of host publication19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017 - Proceedings
PublisherAssociation for Computing Machinery
Pages58-65
Number of pages8
VolumePart F134476
ISBN (Electronic)9781450352994
DOIs
Publication statusPublished - 2017 Dec 4
Event19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017 - Salzburg, Austria
Duration: 2017 Dec 42017 Dec 6

Other

Other19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017
CountryAustria
CitySalzburg
Period17/12/417/12/6

Keywords

  • Concept
  • Data mining
  • Document descriptor
  • Information retrieval
  • Medical documents processing
  • Ontology
  • Summary generation

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Fingerprint Dive into the research topics of 'Medical documents processing for summary generation and keywords highlighting based on natural language processing and ontology graph descriptor approach'. Together they form a unique fingerprint.

  • Cite this

    Dudko, A., Endrjukaite, T., & Kiyoki, Y. (2017). Medical documents processing for summary generation and keywords highlighting based on natural language processing and ontology graph descriptor approach. In 19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017 - Proceedings (Vol. Part F134476, pp. 58-65). Association for Computing Machinery. https://doi.org/10.1145/3151759.3151784