Employing BP Neural Networks to Alleviate the Sparsity Issue in Collaborative Filtering Recommendation Algorithms

Improved Recommendation Sorting of Collaborative Filtering Algorithm

“…Item-based collaborative filtering algorithms improve scalability by focusing on the similarity among items using user ratings rather than on the similarity among users themselves (Sarwar, Karypis, Konstan, & Riedl, 2001). Huang et al [3] adaptively select the neighbors of the prediction target by the user and the similarity calculation of the product to improve the accuracy of the score prediction.Zhang et al [4] based on the cloud model, Zhang Feng et al [5] used BP neural network, Hou Cuiqin [6] proposed a compressed sparse user scoring matrix, through the above improvements to alleviate the affects of scoring data sparsity on collaborative filtering recommendation quality, and improve the accuracy of score prediction.However, the subsequent works on improvements of collaborative filtering mostly focused on the rating prediction of collaborative process, and lack the evaluation of information filtering effects on recommended items.So in this paper,we pay more attention on the sorting result of the recommendation list.…”

Research on User Clustering Collaborative Filtering Algorithm

“…Step 1 Utilize SVD to smooth original sparse matrix and fill up missing values In training dataset, as per [2], target user u's predicted evaluating score of item i can be obtained by the following equation:…”

“…To overcome problems mentioned above, some scholars suggested model-based collaborative filtering approaches, combing machine learning techniques and artificial intelligence to improve classical collaborative filtering algorithms e.g. Bayesian network [2][3], clustering [4][5][6], neural network [7], SVD [8][9], latent topic analysis and so on. The common point of them is to forecast pre-trained models with the use of historical rating matrix.…”

Effective Multifocus Image Fusion Based on HVS and BP Neural Network

“…The architecture of BP neural network in the paper is shown in Figure 1. According to [24], we also adopt empirical formula to determine the number of nodes of the hidden layer, and the formula is defined as follows: $\begin{array}{c}{n}_h=\text{sqrt}(n_i+n_o)+\mathrm{1},\end{array}$ where n h , n i , and n o are the number of nodes of the hidden layer, the number of nodes of the input layer, and the number of nodes of the output layer, respectively.…”