Confidence-weighted bias model for online collaborative filtering

摘要: Abstract Collaborative filtering (CF) is widely applied in recommender systems to predict user preferences or interests according a user’s historical information. Traditional CF methods mainly adopt batch-processing train models, which require prior preparation of all training data. However, always change with time, and it impossible prepare the data at once. In actuality, are obtained time certain sequence. Therefore, update model, trained model needs be re-trained datasets. As result, cost re-training very high, slows down makes new updates difficult capture. To solve these problems, online system emerged. this paper, we propose confidence-weighted bias (CWBM) for collaborative (OCF). This adds into further introduces confidence weights; thus, can improve stability accuracy OCF. A comparative experiment on two real datasets, Movielens100K Mvielens1M, show that method proposed paper superior other baseline methods.