To search, Click below search items.


All Published Papers Search Service


A New Approach for Open-Domain Question Answering System


Ibrahim Mahmoud Ibrahim Alturani and Mohd Pouzi Bin Hamzah


Vol. 18  No. 6  pp. 100-103


Many people have many queries related to the deferent topics. They are inquisitive to find these answers using search engines. The most problem in the search engine is that they provide you with the link to the web page contains passages include answer instead of the exact answer. The goal of a question-answering system is to retrieve answers to questions rather than full documents or passages. QA systems can be running on a closed domain or open domain. In close domain they work on a specific field, like biology or medicine or other, they do not need to integrate different knowledge bases thus less from ambiguity, and they create faster and higher quality results. But closed-domain approaches, not flexibility and there are high costs when adopting such a system to a new domain or implementing a new system. Thus, the research focus has moved to domain-specific, adaptable extension or open-domain QA systems. This paper proposed a new approach architecture for open domain question-answering system depends on the ontology and wordnet to improve answer accuracy. The result of experiments generated a score of 84.7% for WH questions and 82.6% for Yes/No questions.


Question answering, Domain selection, Document processing, Answer processing, Ontology, WordNet, Named Entity Recognition.