Enhancing Context-Aware Hybrid Collaborative Filtering Using DBSCAN Clustering Approach

With the increasing availability of travel-related data and the rising demand for personalized travel recommendations, context-aware recommender systems (CARS) are crucial for assisting travelers in discovering relevant and better experience. The technique known as density-based spatial clustering of application with noise (DBSCAN) is utilized by the framework that has been proposed to locate user clusters that have similar travel preferences and contextual signals. In addition to that, it utilizes a hybrid collaborative filtering strategy, which combines collaborative filtering. This combination improves the precision and variety of the recommendations, providing users with a wider array of pertinent options. By incorporating contextual data such as location, time, and user behavior, the proposed system CHCF_DBSCAN is able to generate recommendations that are specific to the user’s context and current circumstance. Experiments were conducted using Flicker City Platform travel dataset. The outcomes demonstrate the approach’s efficacy in enhancing the recommendations’ precision and relevance.