TY - GEN
T1 - An effective recommendation algorithm for clustering-based recommender systems
AU - Kim, Taek Hun
AU - Yang, Sung Bong
PY - 2005
Y1 - 2005
N2 - In this paper we present an effective recommendation algorithm using a refined neighbor selection and attributes information on the goods. The proposed algorithm exploits the transitivity of similarities using a graph approach. The algorithm also utilizes the attributes of the items. The experiment results show that the recommendation system with the proposed algorithm outperforms other systems and it can also overcome the very large-scale dataset problem without deteriorating prediction quality.
AB - In this paper we present an effective recommendation algorithm using a refined neighbor selection and attributes information on the goods. The proposed algorithm exploits the transitivity of similarities using a graph approach. The algorithm also utilizes the attributes of the items. The experiment results show that the recommendation system with the proposed algorithm outperforms other systems and it can also overcome the very large-scale dataset problem without deteriorating prediction quality.
UR - http://www.scopus.com/inward/record.url?scp=33745609807&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33745609807&partnerID=8YFLogxK
U2 - 10.1007/11589990_159
DO - 10.1007/11589990_159
M3 - Conference contribution
AN - SCOPUS:33745609807
SN - 3540304622
SN - 9783540304623
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 1150
EP - 1153
BT - AI 2005
T2 - 18th Australian Joint Conference on Artificial Intelligence, AI 2005: Advances in Artificial Intelligence
Y2 - 5 December 2005 through 9 December 2005
ER -