Semantic Sparse Service Discovery Using Word Embedding and Gaussian LDA

Gang Tian,Jian Wang,Shengtao Zhao,Junju Liu,Ziqi Zhao,Lantian Guo

doi:10.1109/access.2019.2926559

Abstract

Nowadays, a growing number of web services are offered in API marketplaces browsed by service developers or third-party registries. Under this situation, API marketplaces' users greatly rely on a search engine to find suitable web services. However, due to the fact that functional attributes of web services are usually described in short texts, the search engine-based discovery approach suffers from the semantic sparsity problem, which hinders the effect of service discovery. To address this issue, we propose a novel web service discovery approach using word embedding and Gaussian latent Dirichlet allocation (Gaussian LDA). Unlike most existing service discovery approaches, our approach first uses context information generated by word embedding to enrich the semantics of service descriptions and users' queries. Then, the enriched service description is loaded into the Gaussian LDA model to acquire service description representation. Finally, the services are ranked by considering the relevance between the extended user's query and service description representation. The experiments conducted on a real-world web service dataset and the results demonstrate that the proposed approach achieves superior effectiveness on web service discovery.

Highlights

Benefit from the development of Internet infrastructure and the advantages of service-oriented computing (SOC), an increasing number of enterprises tend to exploit or convert their business applications into distributed web or cloud services [1], [2]
We propose a method that integrating word embedding and Gaussian latent factor based approach (LDA) (GLDA, for short) model to improve the service discovery performance [18]
We integrate external context information learned by word embedding technique, which can boost the performance of some information retrieval (IR) tasks [17], such as short text similarity measurement [16], [30]

Summary

INTRODUCTION

Benefit from the development of Internet infrastructure and the advantages of service-oriented computing (SOC), an increasing number of enterprises tend to exploit or convert their business applications into distributed web or cloud services [1], [2]. These methods may suffer from dimensionality curse due to the sparse representation of short text [16], [17] To address this issue, we propose a method that integrating word embedding and Gaussian LDA (GLDA, for short) model to improve the service discovery performance [18]. In the embedded vector space, words with similar semantic and syntactic attributes tend to be close to each other [19] This characteristic can effectively model the context information such as word co-occurrence pattern, which is used for enriching the semantics of service descriptions, and is suitable for solving the problem of using synonyms/variants of keywords in queries.

RELATED WORK

QUERY MODELING

SERVICES RANKING

EXPERIMENTS

PERFORMANCE COMPARISON

CONCLUSION