Keyphrase Extraction in Scientific Articles: A Supervised Approach
Pinaki Bhaskar, Kishorjit Nongmeikapam, Sivaji Bandyopadhyay · 2012
This paper contains the detailed approach of automatic extraction of Keyphrases from scientific articles (i.e. research paper) using supervised tool like Conditional Random Fields (CRF). Keyphrase is a word or set of words that describe the close relationship of content and context in the document. Keyphrases are sometimes topics of the document that represent the key ideas of the document. Automatic Keyphrase extraction is a very important module for the automatic systems like query or topic independent summarization, question-answering (QA), information retrieval (IR), document classification etc. The system was developed for the Task 5 of SemEval-2. The system is trained using 144 scientific articles and tested on 100 scientific articles. Different combinations of features have been used. With combined keywords i.e. both authorassigned and reader-assigned keyword sets as answers, the system shows a precision of 32.34%, recall of 33.09 % and F-measure of 32.71 % with top 15 candidates.