Abstract:With the rapid progress of Mobile Internet of Things (MIoT), Location-Based Services (LBSs) have become widely popular due to the deployment of diverse devices and facilities. The utilization of spatial–temporal information has led to a considerable improvement in service recommendation in the MIoT-enabled service platform. In practical scenarios, the interaction between the user and the service is often initiated by the joint effect of previous service invocations. Many-to-many and high-order spatial–temporal relations exist among services, which are inadequately reflected by prevailing methods that rely on ordinary graphs and consider spatial–temporal relations among services as pairwise connections. Additionally, most methods indiscriminately conflate the separate influences of spatial and temporal information on user preferences, which can lead to suboptimal service recommendation performance. To this end, we propose an approach powered by spatial–temporal hypergraphs (STH) for personalized service recommendation in the MIoT-enabled service platform, which is inspired by the deep learning paradigm. Specifically, we first construct spatial and temporal hypergraphs based on spatial–temporal information. Then, we develop the spatial–temporal hypergraph neural network-based encoders to learn spatial and temporal representations of services. Finally, we introduce an auxiliary task to differentiate between distinct spatial–temporal effects for improved service recommendations. We conduct extensive experiments on four real-world datasets, and the results demonstrate the effectiveness of STH. Thus, STH not only enhances the user experience but also improves service management in the MIoT-enabled service platform.

Service Recommendation Using Conditional Restricted Boltzmann Machines

Conditional Restricted Boltzmann Machines for Cold Start Recommendations

Content-Boosted Restricted Boltzmann Machine for Recommendation.

A study on time series prediction model based on CRBM algorithm

Recurrent Tensor Factorization for Time-Aware Service Recommendation

Refinements of Approximation Results of Conditional Restricted Boltzmann Machines

Exploiting Structural and Temporal Influence for Dynamic Social-Aware Recommendation

Representation Enhancement Based Cold Start Model for Elastic Compute Service Recommendation

Personalised Service Recommendation Process Based on Service Clustering.

CAMER: A Context-Aware Mobile Service Recommendation System.

A Temporal-Aware Hybrid Collaborative Recommendation Method for Cloud Service

Cloud Service Recommendation with Bi-LSTM and GLDA Based Approach

Service Recommendation Based on Social Balance Theory and Collaborative Filtering.

QF-RNN: QI-Matrix Factorization Based RNN for Time-Aware Service Recommendation

A conditional random field recommendation method based on tripartite graph

Service Recommendation for Mashup Creation Based on Time-Aware Collaborative Domain Regression.

Mobile Service Recommendation Based on Context Similarity and Social Network

A Social Balance Theory-Based Service Recommendation Approach.

SRM-TGA: A session-based recommendation model supported by temporal graph attention

Restricted Boltzmann Machines For Retweeting Behaviours Prediction

A Spatial–temporal Hypergraph Based Method for Service Recommendation in the Mobile Internet of Things-enabled Service Platform