SIG Social Computing
Loading...
Paper Type
Complete
Paper Number
1142
Description
With the swift proliferation of handheld mobile devices, location based social networking (LBSNs) services have gained immense attention allowing users to discover their point of interest (POI). Application of collaborative filtering techniques in POI recommendation becomes challenging due to the sparsity of large user-POI rating matrix. Further, in the context of LBSNs, the spatiotemporal information is pivotal to capture user's real-time preferences. In this work we propose a graph based POI recommendation approach, Context Aware POI with Social Trust (CAST) which integrates the geographical influence of the POIs and the influence of the social connections with the user rankings derived from the weighted bipartite graph. Experiments have been conducted with six state-of-the-art baselines using two real-world LBSN data sets. Findings reveal that user ranking on bipartite graph is a significant contributor to the performance along with social, geographical and spatial influence.
Recommended Citation
Manmath, Davangave Balaji and Pervin, Nargis, "Context Aware POI Recommendation using Bipartite Graph" (2022). AMCIS 2022 Proceedings. 5.
https://aisel.aisnet.org/amcis2022/sig_sc/sig_sc/5
Context Aware POI Recommendation using Bipartite Graph
With the swift proliferation of handheld mobile devices, location based social networking (LBSNs) services have gained immense attention allowing users to discover their point of interest (POI). Application of collaborative filtering techniques in POI recommendation becomes challenging due to the sparsity of large user-POI rating matrix. Further, in the context of LBSNs, the spatiotemporal information is pivotal to capture user's real-time preferences. In this work we propose a graph based POI recommendation approach, Context Aware POI with Social Trust (CAST) which integrates the geographical influence of the POIs and the influence of the social connections with the user rankings derived from the weighted bipartite graph. Experiments have been conducted with six state-of-the-art baselines using two real-world LBSN data sets. Findings reveal that user ranking on bipartite graph is a significant contributor to the performance along with social, geographical and spatial influence.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
Social