Computing User Similarity by Combining SimRank++ and Cosine Similarities to Improve Collaborative Filtering

2017 ◽  
Author(s):  
Xiuli Wang ◽  
Zhuoming Xu ◽  
Xiutao Xia ◽  
Chengwang Mao
Keyword(s):  
Collaborative Filtering ◽  
User Similarity

scholarly journals Collaborative Filtering Recommendation Algorithm Based on TF-IDF and User Characteristics

2021 ◽  
Vol 11 (20) ◽  
pp. 9554
Author(s):  
Jianjun Ni ◽  
Yu Cai ◽  
Guangyi Tang ◽  
Yingjuan Xie
Keyword(s):  
Collaborative Filtering ◽  
User Characteristics ◽  
Recommendation Algorithm ◽  
Data Set ◽  
Rating Data ◽  
User Similarity ◽  
Public Data ◽  
Weighted Algorithm ◽  
Collaborative Filtering Algorithm ◽  
The Impact

The recommendation algorithm is a very important and challenging issue for a personal recommender system. The collaborative filtering recommendation algorithm is one of the most popular and effective recommendation algorithms. However, the traditional collaborative filtering recommendation algorithm does not fully consider the impact of popular items and user characteristics on the recommendation results. To solve these problems, an improved collaborative filtering algorithm is proposed, which is based on the Term Frequency-Inverse Document Frequency (TF-IDF) method and user characteristics. In the proposed algorithm, an improved TF-IDF method is used to calculate the user similarity on the basis of rating data first. Secondly, the multi-dimensional characteristics information of users is used to calculate the user similarity by a fuzzy membership method. Then, the above two user similarities are fused based on an adaptive weighted algorithm. Finally, some experiments are conducted on the movie public data set, and the experimental results show that the proposed method has better performance than that of the state of the art.

Time-Weighted Collaborative Filtering Algorithm Based on Improved Mini Batch K-Means Clustering

2021 ◽  
Vol 105 ◽  
pp. 309-317
Author(s):  
Xue Han ◽  
Zhong Wang ◽  
Hui Jun Xu
Keyword(s):  
Collaborative Filtering ◽  
Sparse Matrix ◽  
Pearson Correlation ◽  
Absolute Error ◽  
User Interest ◽  
User Similarity ◽  
Filtering Algorithm ◽  
Collaborative Filtering Algorithm ◽  
Low Efficiency ◽  
Score Matrix

The traditional collaborative filtering recommendation algorithm has the defects of sparse score matrix, weak scalability and user interest deviation, which lead to the low efficiency of algorithm and low accuracy of score prediction. Aiming at the above problems, this paper proposed a time-weighted collaborative filtering algorithm based on improved Mini Batch K-Means clustering. Firstly, the algorithm selected the Pearson correlation coefficient to improve the Mini Batch K-Means clustering, and used the improved Mini Batch K-Means algorithm to cluster the sparse scoring matrix, calculated the user interest score to complete the filling of the sparse matrix. Then, considering the influence of user interest drift with time, the algorithm introduced the Newton cooling time-weighted to improve user similarity. And then calculated user similarity based on the filled score matrix, which helped to get the last predicted score of unrated items The experimental results show that, compared with the traditional collaborative filtering algorithms, the mean absolute error of Proposed improved algorithm is d, and the Precision, Recall and F1 value of MBKT-CF also get a large improvement, which has a higher rating prediction accuracy.

scholarly journals Enhancing Collaborative Filtering by User-User Covariance Matrix

2018 ◽  
Vol 2018 ◽  
pp. 1-9 ◽  
Author(s):  
Yingyuan Xiao ◽  
Jingjing Shi ◽  
Wenguang Zheng ◽  
Hongya Wang ◽  
Ching-Hsien Hsu
Keyword(s):  
Collaborative Filtering ◽  
Covariance Matrix ◽  
Personalized Recommendation ◽  
Similarity Matrix ◽  
User Similarity ◽  
Filtering Method ◽  
The Past ◽  
Changing Trends ◽  
Over Time

The collaborative filtering (CF) approach is one of the most successful personalized recommendation methods so far, which is employed by the majority of personalized recommender systems to predict users’ preferences or interests. The basic idea of CF is that if users had the same interests in the past they will also have similar tastes in the future. In general, the traditional CF may suffer the following problems: (1) The recommendation quality of CF based system is greatly affected by the sparsity of data. (2) The traditional CF is relatively difficult to adapt the situation that users’ preferences always change over time. (3) CF based approaches are used to recommend similar items to a user ignoring the user’s demand for variety. In this paper, to solve the above problems we build a new user-user covariance matrix to replace the traditional CF’s user-user similarity matrix. Compared with the user-user similarity matrix, the user-user covariance matrix introduces the user-user covariance to finely describe the changing trends of users’ interests. Furthermore, we propose an enhancing collaborative filtering method based on the user-user covariance matrix. The experimental results show that the proposed method can significantly improve the diversity of recommendation results and ensure the good recommendation precision.

Sign in / Sign up

Export Citation Format

Share Document