Using an Islamic Question and Answer Knowledge Base to answer questions about the Holy Quran

This paper presents the QAEQAS Quranic Arabic/English Question Answering System, which relies on a specialized search dataset corpus, and data redundancy. Our corpus is composed of questions along with their answers. The questions are phrased in many different ways in differing contexts to optimize Question Answering (QA) performance. As a complete question answering solution, the Python NLTK natural language toolkit has been used to process the user question as well as to implement the search engine to retrieve candidate results and then extract the best answer. The system takes and accepts a Natural Language (NL) question in English or Arabic from the user - through a GUI - as an input, then matches this question with the knowledge base questions, and then returns the corresponding answer. A keyword based search was used. First the user question was tokenized to get the keywords, and then the stop words were removed. The remaining keywords were used for searching the corpus looking for matched questions. After that, the system used scoring and ranking to find the best matched question and then return the corresponding answer for this question. QAEQAS deals with a wide range of question types including facts, definitions. It produces both short and long answers with a precision of 79% and a recall of 76 for Arabic version; and a precision of 75% and a recall of 73% for English version.