A New Weighted Similarity Method Based on Neighborhood User Contributions for Collaborative Filtering.
Xuefeng ZangTianqi LiuShuyu QiaoWenzhu GaoJiatong WangXiaoxin SunBangzuo ZhangPublished in: DSC (2016)
Keyphrases
- collaborative filtering
- similarity measure
- recommender systems
- high accuracy
- neighbor selection
- user similarity
- detection method
- end users
- clustering method
- pearson correlation coefficient
- segmentation method
- cost function
- objective function
- social networks
- probabilistic model
- computational complexity
- user preferences
- significant improvement
- matrix factorization
- collaborative filtering algorithms
- chronological backtracking
- user interface
- preprocessing