Question Answering

Yassine Benajiba, Paolo Rosso, Lahsen Abouenour, Omar Trigui, Karim Bouzoubaa, Lamia Hadrich Belguith

Abstract: Question Answering (QA) is a task that aims at finding a precise answer to a specific user question. This task is significantly challenging because both the question and the answer are formulated in natural language. For this reason, in order to build an efficient QA system one has to rely on different NLP parsers to extract the necessary information to be used to compute the most relevant answer(s). Challenges are even higher when the target language is based on a rich/complex morphology. Not only the lack of resources and tools stymie the task but also the nature of the language requires some preprocessing for the statistical models to be able to operate efficiently. In this chapter, we describe in details how QA is more complex for rich morphology languages. We also summarize the literature that has been published around this task and describe in more detail some recent research work that has been conducted to build Arabic QA systems.