Ghazanfar, Mustansar and Prugel-Bennett, Adam
An Improved Switching Hybrid Recommender System Using Naive Bayes Classi?er and Collaborative Filtering.
In, The 2010 IAENG International Conference on Data Mining and Applications, Hong Kong,
17 - 19 Mar 2010.
Recommender Systems apply machine learning and data mining techniques for ?ltering un- seen information and can predict whether a user would like a given resource. To date a number of rec- ommendation algorithms have been proposed, where collaborative ?ltering and content-based ?ltering are the two most famous and adopted recommendation techniques. Collaborative ?ltering recommender sys- tems recommend items by identifying other users with similar taste and use their opinions for recom- mendation; whereas content-based recommender sys- tems recommend items based on the content informa- tion of the items. These systems su?er from scalabil- ity, data sparsity, over specialization, and cold-start problems resulting in poor quality recommendations and reduced coverage. Hybrid recommender systems combine individual systems to avoid certain afore- mentioned limitations of these systems. In this paper, we proposed a unique switching hybrid recommenda- tion approach by combining a Naive Bayes classi?ca- tion approach with the collaborative ?ltering. Exper- imental results on two di?erent data sets, show that the proposed algorithm is scalable and provide better performance–in terms of accuracy and coverage–than other algorithms while at the same time eliminates some recorded problems with the recommender sys- tems.
Actions (login required)