Logo image
OmniRank: Learning to Recommend based on Omni-traversal of Heterogeneous Graphs
Edited book   Peer reviewed

OmniRank: Learning to Recommend based on Omni-traversal of Heterogeneous Graphs

Panagiotis Symeonidis and S Chairistanidis
Social Network Analysis and Mining, Vol.9(1), pp.39-63
9, Springer
2019
Handle:
https://hdl.handle.net/10863/53166

Abstract

In this paper, we propose a new node similarity measure, Omni-Rank, for multi-dimensional and heterogeneous social networks. In particular, we recursively propagate the structural similarity computation beyond the neighborhood of the nodes to the entire heterogeneous (e.g., user, item, tag)graph, which incorporates several unipartite and bipartite graphs. We have evaluated experimentally OmniRank and compare it against other state-ofthe-art algorithms (wRWR, SimRank and P-Rank) on two real-life datasets (HetRec 2011 and GeoSocialRec). Our experiments have shown that Omni-Rank outperforms its comparison partners in terms of eectiveness and recommendation accuracy, because it exploits information on both multi-step and omni-directional neighbourhoods (unipartite and bipartite).
url
https://doi.org/10.1007/s13278-019-0585-7View

Details

Metrics

1 Record Views
Logo image