Title :
Social tagging in Recommender Systems
Author :
Arabi, Hossein ; Balakrishnan, Vimala
Author_Institution :
Fac. of Comput. Sci., Univ. of Malaya, Kuala Lumpur, Malaysia
Abstract :
The Web 2.0 gave the Internet users a virtual life, in which they could shop online and try to socialize through Web. Recommender Systems (RS) improves users´ shopping experience by by recommending them a shopping item. Many techniques have been introduced to enhance RS algorithms, including social tagging. Social Tagging let users share resources and this lead to more personalized recommendation. There is a lack of overall information about RS algorithms that have implemented social tagging. Therefore, in this paper we compared and analysed some of the studies that have particularly used social tagging in recommender systems. Both Collaborative Filtering (CF) and Content-based filtering systems were compared, and results show that it is better to combine these algorithms for achieving higher personalized recommendation, and also to address the cold start issue.
Keywords :
collaborative filtering; content-based retrieval; recommender systems; retail data processing; social networking (online); CF system; Internet users; RS algorithms; Web 2.0; cold start issue; collaborative filtering system; content-based filtering system; personalized recommendation; recommender systems; resource sharing; shopping item recommendation; social tagging; user shopping experience improvement; virtual life; Accuracy; Algorithm design and analysis; Collaboration; Recommender systems; Tagging; Taxonomy; Collaborative Filtering; Content-based; Recommendation System; Social tagging;
Conference_Titel :
Computational Science and Technology (ICCST), 2014 International Conference on
DOI :
10.1109/ICCST.2014.7045009
