A Query Expansion Method Using Multinomial Naive Bayes

Information retrieval (IR) aims to obtain relevant information according a certain user need and involves great diversity of data such as texts, images, or videos. Query expansion techniques, part (IR), are used more items, particularly documents, that the requirements. The initial query is reformulated, adding meaningful terms with similar significance. In this study, supervised technique based on an innovative use Multinomial Naive Bayes extract from first documents retrieved by presented. proposed method was evaluated using MAP R-prec 5, 10, 15, 100 documents. improved performance expanded queries increased number in comparison baseline method. We achieved accurate document results (MAP 0.335, 0.369, P5 0.579, P10 0.469, P15 0.393, P100 0.175) compared top performers TREC2017 Precision Medicine Track.