Discovering web services in social web service repositories using deep variational autoencoders
作者：
Highlights：
• We explore the use of Variational Autoencoders for syntactic Web Service discovery.
• We evaluate our approach using a 17113-service dataset, the largest among the research community.
• Our approach outperforms service engines based on traditional dimensionality reduction techniques (LSA, LDA).
• Our approach outperforms service engines based on Word Embeddings.
• Average query processing times and VAE training times confirm that our approach is viable in practice.
摘要
•We explore the use of Variational Autoencoders for syntactic Web Service discovery.•We evaluate our approach using a 17113-service dataset, the largest among the research community.•Our approach outperforms service engines based on traditional dimensionality reduction techniques (LSA, LDA).•Our approach outperforms service engines based on Word Embeddings.•Average query processing times and VAE training times confirm that our approach is viable in practice.
论文关键词：Service-oriented computing,Web Services,Service discovery,Deep neural network,Variational autoencoder
论文评审过程：Received 1 October 2019, Revised 18 February 2020, Accepted 19 February 2020, Available online 6 March 2020, Version of Record 6 March 2020.
论文官网地址：https://doi.org/10.1016/j.ipm.2020.102231
