Name
Abstract | ||
|---|---|---|
With the rapid proliferation of diverse online social network sites, user recommendation has been received unprecedented attention. At present, the methods for user recommendation are mainly divided into two categories: recommending a new friend for a target user according to similar interest, or by friendships similarity between the two users. The first category methods have high recall but low precision, the second methods have high precision but low recall. In this paper, we proposed a new hybrid approach by incorporating users' interests and users' friendships together to recommend new friends for target users. Firstly, we use latent Dirichlet allocation (LDA) to model users' interests, and Weighted-PageRank Algorithm to model users' friendship network, and then merge these two factors into a hybrid model based on PageRank algorithm. This hybrid method models users' interests and users' friendships at the same time, and we demonstrate the effectiveness of our hybrid model by using some social network datasets. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/NAS.2013.38 | NAS |
Keywords | Field | DocType |
friendships similarity,diverse online social network,topic models,hybrid method models user,weighted-pagerank algorithm,new interest-sensitive,new hybrid approach,model user,target user,new friend,social network datasets,recommender systems,online social network sites,user friendships,friendship network,user recommendation,lda,network-sensitive method,pagerank,natural language processing,hybrid model,social networking (online),user interests,web graph,latent dirichlet allocation,interest-sensitive method,computational modeling,vectors,mathematical model,predictive models | Recommender system,PageRank,Latent Dirichlet allocation,Social network,Friendship,Information retrieval,Computer science,Topic model,Merge (version control),Recall | Conference |
Citations | PageRank | References |
1 | 0.37 | 2 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yanmin Shang | 1 | 6 | 5.92 |
| Peng Zhang | 2 | 478 | 39.61 |
| Ya-nan Cao | 3 | 131 | 19.42 |