Deep Learning Model for Personalized Web Service Recommendations Using Attention Mechanism
Résumé
Home Service-Oriented Computing Conference paper
Deep Learning Model for Personalized Web Service Recommendations Using Attention Mechanism
Marwa Boulakbech, Nizar Messai, Yacine Sam & Thomas Devogele
Conference paper
First Online: 20 November 2023
253 Accesses
Part of the Lecture Notes in Computer Science book series (LNCS,volume 14419)
Abstract
The big volume of candidate Web services and their differences make it hard for developers to discover a set of appropriate ones for mashup creation. Thus, recommending suitable services is a vital problem. Service recommendation methods should not only meet the functional needs of users but also consider contextual features like application domain and service performances to provide more personalized recommendations. In this paper, we propose an attention-based deep learning model for service recommendation. It makes service recommendation based on service characteristics and user feed-backs. Specifically, we build a service network, which learns to intelligently discover services with two attention mechanisms - a functional attention mechanism that takes tags as functional prior to mine the function-related features of services and mashups, and a non-functional attention mechanism that considers service qualities to guide the selection of the most appropriate ones and improves user satisfaction. Experiments are carried out on a real-world web API dataset crawled from ProgrammeableWeb.com.