Dividing Social Networks into Two Communities Using the Maximum Likelihood Method: Application to ESG

Abstract

This article explores the application of the Maximum Likelihood Estimation method (MLE) for community detection in environmental, social, and governance (ESG) networks. ESG factors are important in assessing the sustainability and ethical impact of investments. By understanding the structure of social networks that discuss and promote ESG practices, we can gain important insights. It proposes a probabilistic framework for identifying community structures by dividing the network into two distinct groups based on connectivity patterns using the MLE method. The network structure is analyzed, and the method identifies groups of united organizations such as companies, investors, and NGOs with similar ESG orientations and interaction patterns. The results reveal important insights into how ESG information flows within and between these communities, highlighting key influencers and central nodes whose connections play a key role in the diffusion of ESG practices. These conclusions can be important in developing targeted communication strategies, identifying potential opportunities for cooperation, and forming informed investment decisions. By providing a solid framework for analyzing ESG networks, this paper is relevant to a broader understanding of ESG dynamics and supports the development of a more sustainable and interconnected global ecosystem.