Abbreviation Extraction Project: AbbrevExtractor

n Research Problem

n Understanding abbreviations is a difficult task for humans

n Extracting abbreviation is difficult for machines

n Context information is requires

n Well-known abbreviations are not always defined

n A method to identify senses in documents is required

n Abbreviations are highly ambiguous

n Our Solution

n A hybrid approach combining lexical analysis techniques and the Support Vector Machine (SVM): AKA AbbrevExtractor

n It incorporates lexical analysis techniques to supervised learning for extracting abbreviations

n It makes use of text chunking techniques to identify long forms of abbreviations

n It significantly improves recall compared to other techniques

· AbbrevExtractor for biomedical literature can be found at AbbrevExtractor