Integrated news mining technique and AI-based mechanism for corporate performance forecasting

Abstract

The deterioration in a corporation's profitability not only threatens its interests and sustainable development but also causes tremendous losses to other investors. Hence, constructing an effective pre-warning model for performance forecasting is an urgent requirement. Most previous studies only analyzed monetary-based ratios, but merely considering such ratios does not depict the full perspective of a corporation's business conditions. This study thus extends monetary-based ratios to non-monetary-based ratios and aggregates them through the analytic network process (ANP) with a risk-adjusted strategy to establish performance ranks of corporations. Analyzing a corporation's business relationships can help it to react to changes in the market and improve profit margins, as it draws upon such relationship networks for the transfer of scarce resources and knowledge. We believe that no current study adopts such a method to construct a forecasting model. To fill this gap in the literature, this study implements the social network (SN) technique to examine a corporation's competitive edge from seemingly noisy big media data, which are subsequently fed into an artificial intelligence (AI)-based technique to construct the model. The introduced model, examined through real-life cases under numerous conditions, offers a promising alternative for performance forecasting.