Ontology-Based Conversational Recommender System for Motorcycle

Abstract

It is common for customers to face challenges when trying to choose a vehicle that fits their modern lifestyle. Even though there are many recommender systems available to assist with making informed decisions based on unique needs, these systems often lack direct user involvement. Additionally, their recommendations are primarily based on technical specifications rather than functional requirements. To address these limitations, a recent study aimed to create an ontology-based conversational recommender system. This system incorporates user preferences and offers personalized recommendations based on functional requirements. The study evaluated the system based on accuracy and user satisfaction metrics and found that it achieved an impressive recommendation accuracy rate of 87.84%. Furthermore, the study received positive feedback from users searching for motorcycles based on various functional requirements. This feedback is a testament to the system's effectiveness in aiding customers in making informed decisions.