A New Question Answering System for the Arabic Language

Abstract

Abstract:We depict the architecture of a question answering system and methodically evaluate contributions of different system components to accuracy. The system differs from most question answering systems in its dependency on data redundancy rather than complicated linguistic analyses of either questions or contender answers. Because a wrong answer is often worse than no answer. A Question Answering (QA) system is a system that takes natural language questions expressed in the Arabic language then attempts to provide short answers. In order to handle this problem, traditional information retrieval techniques joined with a sophisticated natural language processing approach have been used in this research work. Using keyword matching, simple structures extracted from both the question and the candidate documents selected by the IR system were used in the process of identifying the answer. In order to perform this process, we used an existing tagger to identify proper names and other crucial lexical items and build lexical entries. Also provide an analysis of Arabic question forms and attempt to formulate better kinds of answers that users find more appropriate.

Arabic community question answering

“…Rosso et al (2005) analyzed Arabic questions by removing the stop words, extracting the named entities, and classifying the questions in four categories: Name , Date , Quantity , or Definition . Subsequent work further extracted question focus (the main noun phrase), and important noun phrases and named entities (Wissal et al 2009; Abdelbaki, Shaheen and Badawy 2011; Kanaan and Hammouri 2009). The latter three studies only used a small number of questions (e.g., hundred questions), which makes it hard to draw any general conclusions.…”

“…They assigned a higher rank to passages with a smaller distance between the keywords. Kanaan and Hammouri (2009) used stop word removal, root extraction, and term weighting, together with cosine similarity between document words and question words. Abouenour (2011) used a three-level approach in passage retrieval based on (i) query keywords, expanded with morphology and semantic concepts from the Arabic WordNet, (ii) n -gram structure to give higher weights to question words closer to each other, and (iii) semantics encoded in concept graphs for passage reranking.…”

Lemaza: An Arabic why-question answering system

“…The problem of automatically answering questions formulated in natural language has been studied in the field of IR since the mid 1990s (Verberne 2010). But unlike IR, QA system delivers simple and exact answer to a question instead of set of documents from a query (Kanaan et al 2009; Trigui et al 2010). Most of the QA systems are composed of three subtasks: question analysis, passage or document retrieval and answer extraction.…”

“…Arabic is a Semitic language and is one of the six official United Nations languages, and the fifth most spoken language in the world with approximately 300 million native speakers (Kanaan et al 2009; Akour et al 2011), along with over 1,500 million worldwide Muslims who use it in their regular daily prayers. Yet, few studies and little effort were directed toward the development of QA system for Arabic language, a modest effort when compared to other languages.…”

Development of Yes/No Arabic Question Answering System

“…Kanaan et el [15] described Arabic QA system which makes use of data redundancy rather than complicated linguistic analyses of either questions or candidate answers, to achieve its task. Akour et el [16] introduced a QA system (QArabPro) for Arabic Language.…”