Analysis and implementation of the bi-polar slope one algorithm with the content base filtering method in producing culinary place recommendations in kuningan regency

Abstract

The recommendation system is an analysis model to obtain recommendations that use filtering techniques to produce recommendations as desired. One of the prediction algorithms is Bipolar Slope One which is a development of the Slope One algorithm to predict an event. Previous studies have shown that the results of the recommendations on the Bipolar Slope One algorithm show a mismatch between the type of recommended item and the type of item that has been rated by the active user, because the rating prediction process in the Bipolar Slope One algorithm does not pay attention to the type or content of the item, but pay more attention to the similarity of rating patterns. This study uses the Content-Based Filtering method to form a profile based on the attributes of an item which will make an assessment based on the analysis of the similarity of the user profile with the vector space model. Application users get information about which culinary delights have an item profile that is expected by him. Culinary players get information about what item profile I expect by the User.