Web service discovery method based on knowledge graph and similarity network

Abstract

The invention relates to a Web service discovery method based on a knowledge graph and a similarity network. The Web service discovery method mainly comprises the following steps: respectively obtaining a word embedding matrix, an entity embedding matrix and a theme embedding matrix by using Word2Vec, knowledge graph embedding and LDA; performing matrix alignment on the theme embedding matrix, theword embedding matrix and the entity embedding matrix; s300, taking the aligned theme embedding matrix, word embedding matrix and entity embedding matrix as input of a CNN (Convolutional Neural Network) so as to extract deep service description information, i.e., a feature vector of user service; and calculating similarity scores of the feature vectors of the user services and the feature vectorsof all the stored services, performing sorting from high to low according to the similarity scores, and outputting the stored services corresponding to the top k before the similarity scores as discovery results. Experiments show that the method provided by the invention is superior to the existing method in various evaluation indexes.

Description

technical field [0001] The present invention relates to the technical field of service computing, in particular to the technical field of service discovery, in particular to a Web service discovery method based on a knowledge map and a similarity network. Background technique [0002] Web service discovery is the process of finding and locating existing web services according to the needs of service requesters. When the service provider registers the service with the repository, it provides many similar functional descriptions, such as service information based on natural language description, service category, service provider name, etc. [0003] Existing service discovery methods mainly rely on keyword matching information retrieval techniques. However, due to the problem of grammatical sparsity (keyword sparsity) in the information retrieved by users, search engines may return a large number of irrelevant services. In order to solve the problem of keyword sparsity, some...

AI Technical Summary

Problems solved by technology

Services are described by feature vectors. Although the cosine similarity can be used to directly measure the similarity, it may ignore the data dependencies that may exist in the data set, and conventional measurement methods may not be able to capture this relationship.

Images