A Graph Matching Model for Designer Team Selection for Collaborative Design Crowdsourcing Tasks in Social Manufacturing

In order to find a suitable designer team for the collaborative design crowdsourcing task of product, we consider matching problem between network graph and graph. Due difference in nodes edges two types graphs, propose model based on similar structure. The first uses Graph Convolutional Network extract features structure obtain node-level embeddings. Secondly, an attention mechanism considering differences importance different assigns weights aggregate embeddings into graph-level Finally, graphs be matched are input multi-layer fully connected neural similarity score pair after they obtained from concat operation. We compare our with basic four evaluation metrics datasets. experimental results show that can more accurately pairs crankshaft linkage produced by enterprise is taken as example verify practicality applicability method.