Firm Profiling and Competition Assessment via Heterogeneous Occupation Network
Abstract
Extensive efforts have been made to understand interfirm competitiveness due to its profound significance in several key business objectives. However, it has never been an easy task to fully depict the landscape of competitiveness owing to its heterogeneity, multiformity, and dynamicity. In this paper, we design a novel firm profiling and competitiveness assessment system following the framework defined by Information System Design Theory (ISDT). Specifically, we build a Heterogeneous Occupation Network (HON) using heterogeneous information of employees’ occupation end education histories. We then employ a heterogeneous network embedding model Methpath2Vec to learn firms’ latent representations which are later used to train a supervised classifier model for downstream competitiveness estimation tasks. The efficacy and competency of our proposed system are validated by performance comparisons of multiple models in different settings as well as by formal statistical test analyses. Overall, our solutions shed new light on interfirm competitiveness assessment problems and are potentially beneficial to academic scholars as well as practitioners and decisionmakers in the industry.