A Context Features Selecting and Weighting Methods for Context-Aware Recommendation

The notion of "Context" plays a key role in recommender systems. In this respect, many researches have been dedicated for Context-Aware Recommender Systems (CARS). Rating prediction in CARS is being tackled by researchers attempting to recommend appropriate items to users. However, in rating prediction, three thriving challenges still to tackle:(i) context feature's selection, (ii) context feature's weighting, and (iii) users context matching. Context-aware algorithms made a strong assumption that context features are selected in advance and their weights are the same or initialized with random values. After context features weighting, users context matching is required. In current approaches, syntactic measures are used which require an exact matching between features. To address these issues, we propose a novel approach for Selecting and Weighting Context Features (SWCF). The evaluation experiments show that the proposed approach is helpful to improve the recommendation quality.