Nabila N.F.Basir N.Zaizi N.J.M.Deris M.M.2024-05-292024-05-2920219.79E+122194535710.1007/978-981-15-5856-6_462-s2.0-85096409144https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096409144&doi=10.1007%2f978-981-15-5856-6_46&partnerID=40&md5=909b6179368865516cf1e8a9f4a48b4dhttps://oarep.usim.edu.my/handle/123456789/9529Ontology is used as knowledge representation of a particular domain that consists of the concepts and the two relations, namely taxonomic relation and non-taxonomic relation. In ontology, both relations are needed to give more knowledge about the domain texts, especially the non-taxonomic components that used to describe more about that domain. Most existing extraction methods extract the non-taxonomic relation component that exists in a same sentence with two concepts. However, there is a possibility of missing or unsure concept in a sentence, known as an incomplete sentence. It is difficult to identify the matching concepts in this situation. Therefore, this paper presents a method, namely similarity extraction method (SEM) to identify a missing concept in a non-taxonomic relation by using a rough set theory. The SEM will calculate the similarity precision and suggest as much as similar or relevant concepts to replace the missing or unclear value in an incomplete sentence. Data from the Tourism Corpus has been used for the experiment and the results were then evaluated by the domain experts. It is believed that this work is able to increase the pair extraction and thus enrich the domain texts. � 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.en-USConceptNon-taxonomic relationOntologyOntology extractionRough setSimilarity precisionConference Paper4704781183