Design and Implementation of the Culinary Recommendation System Using Sentiment Analysis and Simple Adaptive Weighting in Bengkulu, Indonesia

Abstract

In 2017, the minister of Indonesia tourism stated that everyone who travels spends his time for culinary about 30-40%. The key point in increasing the tourism income; especially from the culinary sector is about how to inform and promote the wealth of Indonesia culinary to all travellers. The information system of Bengkulu tourism has been developed in the previous study. However, that system has not been able to provide the best culinary recommendations to the travellers. This study focuses on designing and implementing the recommendation system of Bengkulu culinary by using sentiment analysis and simple adaptive weighting (SAW). The recommendation offered is based on the user review and criteria as well. The user review will be classified into positive, negative and neutral reviews by the sentiment analysis method. If the user needs culinary information based on criteria, the system will provide a recommendation and rank of culinary by using simple adaptive weighting. These criteria used are the average price, opening hours, facilities, distance from a central city, and transportation as well. Sentiment analysis method obtains the accuracy of recommendation classification at 79% while the recommendation rank obtained by the SAW method is 90.83%. These results show that the proposed method has a potential for assisting the travellers to gain the best culinary recommendation, especially in the Bengkulu area.