The role of machine learning techniques in optimising top management interlocking networks for green innovation

Social network structural characteristics of top management (TMT) are important variables that affect the outcome of team functioning, and variability in network structural characteristics leads to variability in TMT performance. This paper analyzes TMT social network structure characteristics based on TMT’s social relationship network using machine learning techniques. The top management interlocking network and technological innovation (machine learning technology) are divided into dimensions respectively, and the machine learning technology is used as a mediating variable to establish a model of the mediating effect of machine learning technology between top management interlocking network and green innovation. Statistical analysis of sample data and structural characterization of TMT social relationship networks by machine learning technology are conducted, and regression equations are used to verify the research hypotheses. The test results of the mediating effect of utilized innovation and exploratory innovation covered by the machine learning technology show that the overall regression effect of the model is good ( =0.537, =0.579, F-statistical test is significant), i.e., the mediating variables, utilized innovation and exploratory innovation, positively affect the green innovation performance and are significant. Meanwhile, the heterogeneity and size of TMT’s social relationship network, as well as relationship strength and relationship quality all have a significant and positive effect on green innovation.