The optimization of weights in weighted hybrid recommendation algorithm
Wenfu Lin, Ying Li, Feng Shuang, Yongbin Wang · 2014
In the field of recommender systems, the performance of every single recommendation algorithm is limited and each has its own strengths and weaknesses, so more attentions are paid to the hybrid recommendation algorithms. There are various hybridization strategies, this paper is focused on the weighted hybridization. In the weighted hybridization, researchers are always stumped by a problem -how to optimize the weights of each algorithm. When the number of algorithms of the weighted hybridization is less then 3, then we can fine tune the weight through repeating experiment, but when the number is more then 3, it is hard to get the weights through the same method. And that is what is addressed by this paper.