Pengembangan Sistem Pencarian Buku Fiksi Berbasis Semantik Impresi dengan Pendekatan Ontologi

Abstract

Fiction book search systems continue to be developed for efficiency and relevance, but keyword-based approaches are still limited to capturing subjective impressions. This research develops a semantic impression-based fiction search system using an ontology approach. The methodology used is Design Science Research Methodology (DSRM), and the ontology is built with Methontology, then extended using semantic relations based on the wheel of emotions and synonyms from WordNet. Data collection from 60 respondents resulted in 183 unique impressions that formed the basis of the ontology and testing. Verification results using reasoner revealed conflict-free ontology, and 30 SPARQL scenarios showed stable and efficient search. System performance was evaluated through testing and Blackbox Testing, showing extensive, stable, and accurate search. User evaluation based on the Technology Acceptance Model (TAM) also found the system useful and easy to use, with an average Perceived Usefulness (PU) and Perceived Ease of Use (PEOU) above 5.70 which is measured on a Likert scale of 1-7. This research proves the ontology approach is effective for expanding the search of fiction books based on impressions.