Download presentation
Presentation is loading. Please wait.
Published byGiles Norris Modified over 9 years ago
1
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Concept Frequency Distribution in Biomedical Text Summarization Advisor : Dr. Hsu Presenter : Yu-San Hsieh Author : Lawrence H. Reeve, Hyoil Han, Saya V. Nagori, Jonathan C. Yang, Tamara A. Schwimmer, Ari D. Brooks 2006. CIKM.604-611
2
Intelligent Database Systems Lab N.Y.U.S.T. I. M. 2 Motivation Objective Introduction Method Experiments Conclusions Outline
3
Intelligent Database Systems Lab N.Y.U.S.T. I. M. 3 Motivation The medical text summarization is particularly useful in the biomedical domain, where physicians must continuously find clinical trial study information in mass treatment information database,then to incorporate into their patient treatment efforts.
4
Intelligent Database Systems Lab N.Y.U.S.T. I. M. 4 Objective This paper has proposed a better method to identify important sentences within a full-text and generate text summaries.
5
Intelligent Database Systems Lab N.Y.U.S.T. I. M. 5 Introduction The approaches of generating summaries ─ Extractive and Abstractive UMLS Metathesaurus UMLS MetaMap Transfer Biomedical text concept distribution Full-textSentencesNoun phraseConcept Term abstracts full-text
6
Intelligent Database Systems Lab N.Y.U.S.T. I. M. 6 Method source-model Summary-output Candidate model Source-text Source-model Candidate-modelSentence-pool Candidate-model > best-score Best-sentence y n srcUIs : [12, 13, 14, 7, 10] sryUIs : [5, 9, 6, 7, 8] ui: unit item
7
Intelligent Database Systems Lab N.Y.U.S.T. I. M. 7 Experiments ─ Corpus A citation database of oncology clinical trial papers(1200) ─ Evaluation tool ROUGE-2 and ROUGE-SU4 ─ Model Summaries The first model is the abstract of the paper Three models from three different domain experts were generated ─ Summarizers used for evaluation BaseLine, FreqDist, MEAD, AutoSummarize, SumBasic, SWESUM
8
Intelligent Database Systems Lab N.Y.U.S.T. I. M. 8 Experiments
9
Intelligent Database Systems Lab N.Y.U.S.T. I. M. 9 Conclusions We developed a new algorithm based on frequency distribution modeling and evaluate it using terms as well as concepts.
10
Intelligent Database Systems Lab N.Y.U.S.T. I. M. 10 My opinion Advantage ─ …… Drawback ─ …… Application ─ Information Retrieval
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.