A Review on Cooperative Question-Answering Systems

• Author(s): Melo, Dora ; Pimenta Rodrigues, Irene ; Beires Nogueira, Vitor
• Date: 2013
• Persistent ID: http://hdl.handle.net/10174/9452
• Origin: Repositório Científico da Universidade de Évora
• Subject(s): Question-Answering Systems; Information Retrieval; Information Extraction; Natural Language Processing
• Document Type: Article
• Language: Portuguese
• Editor(s): Ferreira, Lígia; Pedro, Vasco

Description

The Question-Answering (QA) systems fall in the study area of Information Retrieval (IR) and Natural Language Processing (NLP). Given a set of documents, a QA system tries to obtain the correct answer to the questions posed in Natural Language (NL). Normally, the QA systems comprise three main components: question classification, information retrieval and answer extraction. Question classification plays a major role in QA systems since it classifies questions according to the type in their entities. The techniques of information retrieval are used to obtain and to extract relevant answers in the knowledge domain. Finally, the answer extraction component is an emerging topic in the QA systems. This module basically classifies and validates the candidate answers. In this paper we present an overview of the QA systems, focusing on mature work that is related to cooperative systems and that has got as knowledge domain the Semantic Web (SW). Moreover, we also present our proposal of a cooperative QA for the SW.