Research Proposal Research Question “ Can Named Entities improve

Research Proposal Research Question " Can Named Entities improve Query Precision in the Biomedicine domain? " Observations 1) Phrasing techniques have not been fine tuned to the requirements of the biomedicine domain. 2) Most of the phrasing techniques have been based on n-gram models. 3) Research in the area of phrasing and biomedicine specific IR systems has been done in isolation. 4) Due to the diversity of the vocabulary of the biomedicine domain, existing phrasing techniques may not improve query precision. 5) Biomedicine is a specialized domain with vocabulary rich in named entities. 6) Named Entity Recognition systems have shown remarkable precision in the news article domain. Hypotheses Adding named entities as phrases to the index of an IR system can increase the accuracy because: • The biomedicine domain has a rich and diverse collection of named entities. • Users of such systems invariably use named entities for their search. e.g. N-acetyl-cysteine, 84 kDa proteins Prior Work Named Entity Recognition started as task of the Message Understanding Conference in 1995, primarily to understand text at a higher level. Most of the research in the area of named entities (NE) has been focused on fine tuning extracting information from news articles. Many supervised learning algorithms have been employed to extract named entities from the text namely: • Support Vector machines • Hidden Markov Models • Maximum Entropy Models Recently, researchers have started recognize their relevance in the biomedicine domain. Although, not very encouraging results have been achieved due to lack of sufficient training data, evidence shows that with each passing year researchers are getting better results.