SISTEM QUESTION ANSWERING MENGGUNAKAN PENDEKATAN BERBASIS PENGETAHUAN

• Main Authors: , I Wayan Dharma Suryawan, , Dr. tech Khabib Mustofa, S.Si., M.Kom
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2013
• Subjects: ETD
• Online Access: https://repository.ugm.ac.id/119070/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=59061
• Institution: Universitas Gadjah Mada

Human using several tools to communicate, but the most commonly used is natural language. If computer can understand human�s natural language, human can give command to the computer easily. This research attempt to develop a question answering system using knowledgebased approach. The syntactic and semantic knowledge are represented into grammatical rules and the world knowledge is represented into ontology serialized using Web Ontology Language (OWL). Meaning of the sentence used to ask question is gotten by transform the sentence into word sequences, represent the word sequence into a parse tree using the bottom-up, right-to-left, depth-first search, and extract semantic information from the parse tree using the rule-to-rule hypothesis. The developed system is able to understand question given using correct sentence according to the grammar and detect question given using uncorrect sentence according to the grammar. The system is also able to use thesaurus when retrieve informations. Tests conducted using questions provided by the respondents indicate that the success rate of applications in understanding question given using correct sentence according to the grammar or detect question given using uncorrect sentence according to the grammar are 80%.