Based on the similarity of interval-valued intuitionistic fuzzy sets defined by entropy in the application of commodity recommendation

The effective and accurate recommendation method is called for the fast development of E-commence. In the existed recommendation algorithms, there are two biggest problems: data sparseness and manipulation problem. In our paper, we propose a new view to recommend which based on similarity of two interval-valued intuitionistic fuzzy sets defined by entropy. And we have proved that the recommendation algorithm is feasible.