Evaluating Different Similarity Measures for Automatic Biomedical Text Summarization

Automatic biomedical text summarization is maturing and can provide a solution for biomedical researchers to access the information they need efficiently. Biomedical summarization approaches often rely on the similarity measure to model the source document, mainly when they employ redundancy removal or graph structures. In this paper, we examine the impact of the similarity measure on the performance of the summarization methods. We model the document as a weighted graph. Various similarity measures are used to build different graphs based on biomedical concepts, semantic types and a combination of them. We next use the graphs to generate and evaluate the automatic summaries. The results suggest that the selection of the similarity measure has a substantial effect on the quality of the summaries (≈37% improvement in ROUGE-2 metric, and ≈29% in ROUGE-SU4). The results also demonstrate that exploiting both biomedical concepts and semantic types yields slightly better performance.

[1]  Dongwook Shin,et al.  Clustering cliques for graph-based summarization of the biomedical research literature , 2013, BMC Bioinformatics.

[2]  Ruslan Mitkov,et al.  The Oxford handbook of computational linguistics , 2003 .

[3]  Xiaohua Hu,et al.  A coherent graph-based semantic clustering and summarization approach for biomedical literature and a new summarization evaluation method , 2007, BMC Bioinformatics.

[4]  Kamal Sarkar,et al.  Using Domain Knowledge for Text Summarization in Medical Domain , 2009 .

[5]  David Camacho,et al.  A genetic graph-based clustering approach to biomedical summarization , 2013, WIMS '13.

[6]  Hyoil Han,et al.  Concept frequency distribution in biomedical text summarization , 2006, CIKM '06.

[7]  Betsy L. Humphreys,et al.  Unified Medical Language System® (UMLS®) Project , 2017 .

[8]  Guilherme Del Fiol,et al.  Text summarization in the biomedical domain: A systematic review of recent research , 2014, J. Biomed. Informatics.

[9]  Jorge Carrillo de Albornoz,et al.  Evaluating the use of different positional strategies for sentence selection in biomedical literature summarization , 2012, BMC Bioinformatics.

[10]  Chin-Yew Lin,et al.  ROUGE: A Package for Automatic Evaluation of Summaries , 2004, ACL 2004.

[11]  Rada Mihalcea,et al.  TextRank: Bringing Order into Text , 2004, EMNLP.

[12]  M. Owen,et al.  Genomic insights into the overlap between psychiatric disorders: implications for research and clinical practice , 2014, Genome Medicine.

[13]  Panagiotis Stamatopoulos,et al.  Summarization from Medical Documents: A Survey , 2005, Artif. Intell. Medicine.

[14]  Dongwook Shin,et al.  Degree centrality for semantic abstraction summarization of therapeutic studies , 2011, J. Biomed. Informatics.

[15]  W. Alkema,et al.  Application of text mining in the biomedical domain. , 2015, Methods.

[16]  Hyoil Han,et al.  BioChain: lexical chaining methods for biomedical text summarization , 2006, SAC.

[17]  Vishal Gupta,et al.  Recent automatic text summarization techniques: a survey , 2016, Artificial Intelligence Review.

[18]  Halil Kilicoglu,et al.  Automatic summarization of MEDLINE citations for evidence-based medical treatment: A topic-oriented evaluation , 2009, J. Biomed. Informatics.

[19]  Karen Spärck Jones.  Automatic summarising: The state of the art , 2007, Inf. Process. Manag..

[20]  Laura Plaza.  Comparing different knowledge sources for the automatic summarization of biomedical literature , 2014, J. Biomed. Informatics.

[21]  Xiaojun Wan,et al.  Recent advances in document summarization , 2017, Knowledge and Information Systems.

[22]  Pablo Gervás,et al.  A semantic graph-based approach to biomedical summarisation , 2011, Artif. Intell. Medicine.

[23]  Jian Pei,et al.  Data Mining: Concepts and Techniques, 3rd edition , 2006 .

[24]  Hyoil Han,et al.  The use of domain-specific concepts in biomedical text summarization , 2007, Inf. Process. Manag..