Entity Extraction in Biomedical Corpora: An Approach to Evaluate Word Embedding Features with PSO based Feature Selection

Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 1159–1170, Valencia, Spain, April 3-7, 2017. c⃝2017 Association for Computational Linguistics Entity Extraction in Biomedical Corpora: An Approach to Evaluate Word Embedding Features with PSO based Feature Selection Shweta Yadav, Asif Ekbal, Sriparna Saha, Pushpak Bhattacharyya Indian Institute of Technology Patna Bihar, India {shweta.pcs14,asif,sriparna,pb}@iitp.ac.in Abstract Text mining has drawn significant atten- tion in recent past due to the rapid growth in biomedical and clinical records. Entity extraction is one of the fundamental com- ponents for biomedical text mining. In this paper, we propose a novel approach of feature selection for entity extraction that exploits the concept of deep learning and Particle Swarm Optimization (PSO). The system utilizes word embedding features along with several other features extracted by studying the properties of the datasets. We obtain an interesting observation that compact word embedding features as de- termined by PSO are more effective com- pared to the entire word embedding fea- ture set for entity extraction. The pro- posed system is evaluated on three bench- mark biomedical datasets such as GENIA, GENETAG and AiMed. The effective- ness of the proposed approach is evident with significant performance gains over the baseline models as well as the other ex- isting systems. We observe improvements of 7.86%, 5.27% and 7.25% F-measure points over the baseline models for GE- NIA, GENETAG, and AiMed dataset re- spectively. 1 Introduction The tremendous amount of information accumu- lated in the domains of molecular biology has drawn the attention of biomedical natural lan- guage processing (BioNLP) community in order to facilitate the development of various tools for various text processing applications, curation and organization of ever-growing biomedical literature etc. Entity extraction is crucial step for solving several pipelined applications such as information extraction, automatic summarization, question- answering, word sense disambiguation etc. Bio- medical entities mostly refer to the biological se- quences of protein & gene such as DNA, RNA, cell type, cell line etc. (Kim et al., 2004). The way of extracting these information from biomed- ical and clinical texts refers to as entity extraction. An automatic system which can extract biomed- ical names such as gene, protein or any disease name from text can substantially reduce the human efforts. However, extracting these entities from text poses several challenges which are presented as follows: 1. Named entities are very generative in nature, i.e. many new names are continuously being generated. Any dictionary can not capture all the various forms of a given name. 2. Similar words convey different meanings, and therefore, a word can have multiple NE types. For example, gene names often con- tain alphabets, digits, hyphens, and other characters, thus having many variants (e.g., “HIV-1 enhancer” versus “HIV 1 enhancer”). Moreover, many abbreviations (e.g., “IL2” for “Interleukin 2”) constitute integral parts of biomedical named entities (NEs). 3. Biomedical names are usually of long length, and contains different types of symbols, and hence boundary detection becomes problem- atic. 4. …

Content truncated to 500 words