In this work we present a novel item recommendation approach that aims atimproving Collaborative Filtering (CF) in social tagging systems using theinformation about tags and time. Our algorithm follows a two-step approach,where in the first step a potentially interesting candidate item-set is foundusing user-based CF and in the second step this candidate item-set is rankedusing item-based CF. Within this ranking step we integrate the information oftag usage and time using the Base-Level Learning (BLL) equation coming fromhuman memory theory that is used to determine the reuse-probability of wordsand tags using a power-law forgetting function. As the results of our extensive evaluation conducted on data-sets gatheredfrom three social tagging systems (BibSonomy, CiteULike and MovieLens) show,the usage of tag-based and time information via the BLL equation also helps toimprove the ranking and recommendation process of items and thus, can be usedto realize an effective item recommender that outperforms two alternativealgorithms which also exploit time and tag-based information.
展开▼