Graph embedding enhanced Web API (Application Program Interface) recommendation method and system

Abstract

The invention relates to a graph embedding enhanced Web API recommendation method and system, and the method comprises the steps: obtaining ID embedding vectors and text embedding vectors of a Mashup node and a Web API node, and calculating ID embedding vectors and text embedding vectors of neighbor nodes of the Mashup node and the Web API node; fusing the ID embedding vector and the text embedding vector, and the ID embedding vectors and the text embedding vectors of all neighbor nodes to obtain a fused embedding vector EM and a fused embedding vector EA; based on the EM and the EA, the matching degree of the Mashup node and the Web API node is calculated, and a Web API recommendation result is obtained. According to the method, the problem of poor recommendation precision caused by the problems of data sparsity and cold start when Web API recommendation is carried out by a method for generating an ID embedding vector purely based on a graph embedding method in a traditional Web API recommendation method is solved.

Description

technical field [0001] The invention relates to the technical field of automatic interaction, in particular to a method and system for Web API recommendation with enhanced graph embedding. Background technique [0002] Web service is a service-oriented architecture technology that is often used to automate interactions or link business processes between distributed and heterogeneous systems. However, it is difficult for Web services with a single function to meet some complex and changeable requirements. To solve this problem, a new enterprise-level application development technology, Mashup, which is different from traditional resource integration solutions, is proposed. This technology can integrate single-function services (ie Web API services: using REST style, HTTP protocol, JSON data format; application programming interface that can be used through the Internet; has many advantages such as easy access, extensibility, easy development and combination, etc.), Build mu...

AI Technical Summary

Problems solved by technology

[0004] The problem with the collaborative filtering recommendation method is that when the call relationship matrix data between Mashup and Web API is sparse, the performance of the recommendation algorithm is very poor

Although this graph-based embedding vector representation can alleviate the data sparsity problem existing in the collaborative filtering algorithm by superimposing high-order relationships, it has the same problem as the collaborative filtering method, that is, they are generated only by the calling relationship. ID embedding vector to calculate the embedding embedding vector corresponding to Mashup and Web API, without considering the text description information of Mashup and Web API itself

Images