Social Group oriented Recommendation Systems in Wireless Edge-Empowered Metaverse

Metaverse incorporates a variety of information technologies such as advanced wireless communications, edge computing and virtual reality to create a multi-dimensional, intelligent and powerful virtual world. However, the large number of products and user groups can lead to difficulties in matching users with Metaverse products. In this paper, we present a Metaverse-oriented user preference recommendation system that strengthens the connection between Metaverse users and products by matching the vast array of products in the Metaverse with user preferences. In order to solve the problem that the classical collaborative filtering recommendation algorithm is inefficient in dealing with a large amount of multidimensional information in Metaverse, the clustering algorithm is introduced into the recommendation algorithm, which is the key factor to deploy the recommendation system. This idea allows the users to be preclassified based on their social relationships, improving the efficiency of the recommendation algorithm. Moreover, we enhance the performance of the recommendation system with a Density-Sensitive Distance based Transformer (DSD-Transformer) framework in Metaverse. Extensive experiments demonstrate that our proposed algorithm can significantly improve the Metaverse experience over various state-of-the-art baselines, and can be used to create a variety of product recommendation systems.