A Dynamic Item-Based Weight Collaborative Recommendation Algorithm

In order to resolve collaborative filtering recommendation system recommended decline in the quality for the sparse dataset, a dynamic Item-based weight collaborative recommendation algorithm is presented, which user´s preference weight items vector set is constructed based on filtering user´s evaluating data and the data rate measured and time-weighted are done, then the Item-based & weighted collaborative filtering recommendation is achieved in the target user´s TOP-N similarity set. Experiments show that the algorithm is better than the traditional collaborative filtering algorithms in improving the recommendation dependability and accuracy.