GNN for QoS

using Python, I want to build a model which uses a graph neural network to enhance the web service quality-of-service prediction. The dataset is: WS-Dream (https://wsdream.github.io/) Evaluation metrics: Root mean square error (RMSE) and Mean Absolute error (MAE) Compared it with three other models: • UIPCC: A hybrid collaborative filtering method that combines data of similar users and services for QoS prediction [1]. • RANEP: A Web service QoS prediction method based on reputation-aware network embedding. This scheme considers the user’s reputation, employs reputation-aware network embedding to learn the user nodes’ vector representation, and finally adopts user-based collaborative filtering for prediction [2]. • TAP: A personalized reputation-aware QoS prediction method that employs k-means clustering and Beta distribution to calculate a user’s reputation. Service clustering is exploited to identify similar services, while the data contributed by reliable and similar users and services are utilized for QoS prediction [3]