Development and Evaluation of Viajefy: A Tourism Information System using TF-IDF Algorithm
DOI:
https://doi.org/10.61310/mjst.v23i1.2383Keywords:
confusion matrix, CRISP-DM, machine-learning, natural language processing, text miningAbstract
The study aimed to design, develop, and evaluate a tourism information system called "Viajefy," incorporating the TF-IDF algorithm and assessing its model performance. It employed Feature-Driven Development as the software development method. It utilized the Cross-Industry Standardized Process for Data Mining for the data mining process of the TF-IDF algorithm, serving as its recommender agent feature. The confusion matrix evaluation tool was used to assess the algorithm's performance, yielding an accuracy of 97%, a precision of 93%, and a recall of 90%. Results showed that the recommender agent of the software application was proven reliable based on the algorithm's performance criteria in terms of accuracy, precision, and recall, and the system received a "Very Highly Acceptable" rating of 4.74. This software application is one of the first studies along tourism information systems for Ilocos Sur, Philippines, to integrate a recommender agent to help the Provincial Government of Ilocos Sur advertise attractions and establishments to be managed by the said government, where one of the Seven Wonders of the World, Vigan City, is situated.