Naive Bayes Classification in The Question and Answering System

Abstract

Abstract—Question and answering (QA) system is a system to answer question based on collections of unstructured text or in the form of human language. In general, QA system consists of four stages, i.e. question analysis, documents selection, passage retrieval and answer extraction. In this study we added two processes i.e. classifying documents and classifying passage. We use Naïve Bayes for classification, Dynamic Passage Partitioning for finding answer and Lucene for document selection. The experiment was done using 100 questions from 3000 documents related to the disease and the results were compared with a system that does not use the classification process. From the test results, the system works best with the use of 10 of the most relevant documents, 5 passage with the highest score and 10 answer the closest distance. Mean Reciprocal Rank (MMR) value for QA system with classification is 0.41960 which is 4.9% better than MRR value for QA system without classification