A Hybrid Group Recommender System Using Fuzzy Clustering and GNN-Based Collaborative Filtering

Recommender systems matter on digital platforms because they provide suggestions for each user, but group recommender systems (GRS) have a hard time making different user choices equal and fair. This research tackles these issues through a hybrid GRS. It carefully integrates Fuzzy C-Means clustering to group users and also integrates Light Graph Convolutional Networks (LightGCN) for collaborative filtering. The method improves scalability throughout many recommendation settings while increasing recommendation precision as well as group consistency. Detailed, experimental evaluations on the MovieLens 1M and MovieLens 100 K datasets clearly show that the proposed approach considerably outperforms customary collaborative filtering model and all existing GNN-based models with regard to precision, recall, and each user's overall satisfaction. The findings indicate that hybrid GRS models have strong potential for real-world applications such as entertainment, e-commerce, and online learning, contributing to more accurate and equitable group recommendations.