Abstract

The classification of tourism attractions is becoming a more well admirable research trend. Various techniques have been taken on board to classify tourism attractions to provide information to the users based on their preferences. This study provides an in-depth insight into the entropic distance-based classification approach for the prediction of user attractions. In this study, a lazy classification technique, k-star, is implemented to predict the tourism places based on user ratings. The k-star algorithm is the nearest neighbor approach that discovers the nearest instances to the targeted instance. Unlike other nearest neighbor approaches, the k-star algorithm exploits entropic distance, which measures all the possible shortest paths to discover the nearest instances based on user ratings. Furthermore, the evaluation assessments are also carried out to justify the performance of the k-star algorithm.