Abstract

Research on Arabic Natural Language Processing is facing a lot of problems due to language complexity, lack of machine readable resources, and lack of interest among Arab researchers. One of the fields that research has started to appear in is the field of Question Answering. Although some research has been done in this area, few have proved to be effective in producing exact relevant answers. One of the issues that affected the accuracy of producing correct answers is proper tagging of entities and proper analysis of a user’s question. In this research, a set of 60+ tagging rules, 15+ Question Analysis rules, and 20+ Question Patterns were built to enhance the answer generation of Natural Language Questions posed over some corpora collected from different sources. A QA system was built and experimental results showed good results with an accuracy of 78%, a recall of 97%, and an F-Measure of 87%.

Key words: Arabic; Question Answering; Question Analysis; Tagging, NLP.