Developing a Personal Recommender System for Homestay Services using TOPSIS Method
Homestay Perlis recommender system is a web-based application system developed for users to find the best homestay services in Perlis. This system functions as: i) recommending the best homestay in Perlis and ii) receive review of homestays from the users. User requirements were identified through a...
Gespeichert in:
Veröffentlicht in: | Journal of Computing Research and Innovation 2018-01, Vol.1 (1), p.60-66 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Homestay Perlis recommender system is a web-based application system developed for users to find the best homestay services in Perlis. This system functions as: i) recommending the best homestay in Perlis and ii) receive review of homestays from the users. User requirements were identified through a survey. A set of questionnaire was distributed, and several criteria to select a homestay were identified. The collected data from experienced users were used to calculate the result using TOPSIS method. From the requirements, an application of a recommender system was designed and developed. The advantage of this system is the recommendations for the homestay are based on the criteria that meet users’ priority. Besides that, the system able to help users to make a right decision in order to choose a homestay that meets their preferences and requirements. The usability test and the user acceptance test were conducted in the study. Results from usability test that focuses on the system’s design, navigation, content, understanding and interactivity show that the recommender system is easy to navigate, fast loading and easy to understand. The user acceptance test measures the aspects of developing a website which are reactions to the website, interface design, navigation and the content for developing a website, which indicates high mean values showing that the recommender system is accepted by the users. Generally, the objectives of this research are successfully achieved. |
---|---|
ISSN: | 2600-8793 |