Skip to Main Content
Recently, there has been growing interests in exploiting stakeholder information to acquire essential knowledge about stock investments. More and more countries legislate for publicly-issued companies to provide such information. In this paper, we propose a new approach to exploit stakeholder information for constructing stake-based social networks. We devise three types of networks: StakeNet (a company-person network), StakeCompanyNet (a company-company network), and StakePersonNet (a person-person network). We also present a visualization tool to display socio-centric and ego-centric views of the networks. Furthermore, we investigate the static and dynamic properties of the StakeNet, and the results reveal that most of StakeNet's characteristics are similar to those of a typical social network, excluding that the in-degree does not follow a power law distribution. Finally, we show two applications of StakeNet by utilizing it to discover influential companies and business groups. The experiments suggest that our outcomes are highly consistent with the results generated by human experts.