QoS Prediction of Web Services Based on Reputation-Aware Network Embedding

Abstract

As the emergence of numerous services with similar functions, it is very helpful to recommend personalized services for users, and urgent to accurately predict the QoS(Quality-of-Service) values of Web services. Collaborative Filtering (CF) is a commonly-used method to handle above issues. However, it faces two common issues: data sparsity problem and trustworthiness issue, which greatly reduces its prediction accuracy. To address this problem properly and systematically, we introduce the network embedding learning into the QoS prediction process and propose an improved QoS prediction method based on the reputation-aware network embedding learning. Firstly, a two-phase K-means clustering is adopted to filter untrustworthy users. Next, the reputation of trustworthy users is calculated, and an attributed user-service bipartite network is constructed between trustworthy users and services while considering the user reputation. Then the reputation-aware network embedding is adopted to learn the hidden representations of users. Finally, user-based CF is adopted to predict the unknown QoS values. The experimental results show that our method has a significant improvement in accuracy compared with other methods.

QoS Prediction of Web Services Based on a Two-Level Heterogeneous Graph Attention Network

“…Furthermore, Gao et al [7] propose a Bipartite Network Embedding (BiNE) model to embed dichotic networks and verify their model's effectiveness on movie recommendation. Wang et al [2] propose a novel Web service QoS prediction method that combines reputation perception and network embedding.…”

“…preserving the information network's global structure and properties. Additionally, the learned representation can be effectively exploited to solve the prediction [2] and classification [3] problems. Representative models include Deepwalk [4], Line [5], and node2vec [6].…”

A new model for trustworthy web service QoS prediction

“…24 To address these issues, numerous research have been conducted. For example, 25,26 presented ways to address the data sparsity problem, 22,27,28 considered the cold-start problem, [29][30][31][32] and evaluated the credibility of user histories before being employed.…”

“…Some methods have used the reputation of the users to identify unreliable users and reduce their impact on the prediction accuracy. [24][25][26][27]33 Calculating the user's reputation is simple and does not require additional information about the users. Each prediction model presented in past studies combines the reputation of the users with a specific prediction method to have a new trust-aware prediction model.…”

“…Some methods have used the reputation of the users to identify unreliable users and reduce their impact on the prediction accuracy 24–27,33 . Calculating the user's reputation is simple and does not require additional information about the users.…”