Optimizing Sports Center Recommendation System in Malaysia Through Content-Based Filtering Technique and Web Application

Sports center is a place that provides services for sports court booking where people can play the sport together as a group of families or friends. However, with the traditional method of booking, which needs to book directly at the venue, many customers feel unsure about which sports court is available provided by the sports center because there are no recommendations. Thus, a new method is needed to recommend the customer according to their preferences: Term Frequency-Inverse Document Frequency and Cosine Similarity technique. This study proposed these methods for optimizing the sports center’s recommendation system, and both are content-based filtering. It is implemented using a web-based platform. Two types of testing were conducted: functionality and cosine similarity testing that tested for user profile, location and court. As a result, the functional testing results showed all requirements correctly and successfully. The cosine similarity score for the user profile location for Test 1 is between 0.2894 and 0.5031; meanwhile, Test 2 is between 0.2781 and 0.4557. The similarity between the court and location is 1.0. It shows that the details of the sports center recommendation are similar to the features input and correctness. Therefore, it can conclude that optimizing the sports center recommendation system through content-based filtering enables the customer to choose the sports center with the similarity features based on the user’s latest booking of location and court in the system.