Question-Response System Using the NooJ Linguistic Platform

Nowadays, the access to medical information is a crucial task given the large number of electronic documentation and its various sources, particularly on the internet. Search engines such as Google, Yahoo, etc. establish an effective solution to find documents corresponding to a user request but they provide imprecise and fast information corresponding to user needs. This research gap motivated us to develop a question-answer system allowing users to ask a question about the desired information using natural language without browsing through documents. The request is presented as few key words; our system responds by a precise and quick answer thanks to the various features provided by the development environment NooJ. The maturity and the efficiency of the search medical information tools are classified according to the level of complexity of the subject area and the target language. Despite of several researches, tools for Arabic language remain relatively lacking many features compared to other languages. The implementation of the proposed system is based on two processes: the first process consists of identifying the type and keywords of the question with the aim of limiting the number of responses. The second process consists of applying the appropriate transducer set to the corpora to extract responses.