A Novel Aggregation Technique for Multi-Criteria Recommendation

The traditional recommender system makes the recommendations using the overall ratings toward items provided by the users. However, the multi-criteria recommender system suggests that considering the effects of criteria ratings to the overall rating is the key to provide more personalized recommendations. In this work, a novel multi-criteria recommendation technique is proposed. The prediction from each criterion is made by considering the trade-off between the neighborhood-based and the model-based techniques. The effects of the criterion ratings to the overall rating are measured by the similarities among the user preference patterns, extracted from matrix factorization. The evaluation shows that our proposed method outperforms various well-known techniques on both single and multi-criteria recommendations.