Print Email Facebook Twitter Query Answerability Classifier for Direct Answer Module in Web Search Engines Title Query Answerability Classifier for Direct Answer Module in Web Search Engines Author Wang, Yiran (TU Delft Electrical Engineering, Mathematics and Computer Science; TU Delft Intelligent Systems) Contributor Hauff, C. (mentor) Iosifidis, G. (graduation committee) Degree granting institution Delft University of Technology Programme Computer Science and Engineering Project CSE3000 Research Project Date 2021-06-30 Abstract In order determine when we can show direct answer module to user queries in web search engine, an independent classifier is designed in this study to assess the answerability of each user query. Real user queries are sampled from MS MARCO Question Answering and Natural Langauge Generation dataset \cite{MSMARCO} and manually labelled with query answerability to train and evaluate the classifier. As a result, the XGboost model has an overall better performance than the random forest model with prediction accuracy score 0.83 and F1 score 0.89. Once the classifier determines the user query is answerable, a MRC model may be used to find the direct answer within provided passages. Else, no direct answer shall be provided to this query. Subject Information Retrievalsearch engineAnswerability of queriesMachine Reading Comprehension To reference this document use: http://resolver.tudelft.nl/uuid:c4464d76-d3ec-4d2f-af74-bcea54292315 Part of collection Student theses Document type bachelor thesis Rights © 2021 Yiran Wang Files PDF Research_Project_19_3_.pdf 560.92 KB Close viewer /islandora/object/uuid:c4464d76-d3ec-4d2f-af74-bcea54292315/datastream/OBJ/view