Collaborative filtering is the process of predicting how a user would rate a given item from other user ratings. we propose a new collaborative filtering algorithms, ratio-based collaborative filtering algorithms, by calculating the ratio between the ratings of one item and another for users who rated both to predict the ratings. Ratio-based collaborative filtering algorithms are easy to implement, and have reasonably accurate, by factoring in the weighted average methods and the preference parameter, we achieve results competitive with traditional memory-based algorithms over the Movielens data sets. The result is sufficient to support our claim.
Ratio-based collaborative filtering algorithms
01.12.2008
477602 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Incremental Collaborative Filtering for Highly-Scalable Recommendation Algorithms
British Library Conference Proceedings | 2005
|Common and Special Embedding-Based Collaborative Filtering
Springer Verlag | 2022
|Common and Special Embedding-Based Collaborative Filtering
British Library Conference Proceedings | 2022
|An Ontology-Based Collaborative Filtering Personalized Recommendation
British Library Conference Proceedings | 2013
|