Enterprise Working Capital Management by BP Neural Network under the Background of the Internet

Abstract

Today, with the rapid development of the Internet, the way of enterprise development has become more and more diversified. The development of the Internet not only brings opportunities for enterprise development but also certain risks. In order to deal with the uncertain risks in the future development of the enterprise, aiming at the operation and capital management problems of the enterprise, this study creates a scientific and effective performance evaluation system that conforms to the law of enterprise development to guide the enterprise's capital management problems in the operation process. This system can enable enterprises to improve economic efficiency and achieve sound development in the context of the Internet economy. This study sorts out the relevant theories of enterprise capital management and performance evaluation and refers to the existing performance evaluation system by analyzing the influencing factors of enterprise operation performance. The key factors of performance evaluation in the process of enterprise operation are extracted and processed by dimensionless normalization of sample data. The Pearson correlation test is used to eliminate data indicators with high correlation, and a model of enterprise operation performance evaluation based on the Back Propagation Neural Network (BPNN) is constructed. Valid panel data of some companies in 2015-2016 are selected. The constructed BPNN model is verified for feasibility. The results show that the constructed BPNN model has wide applicability and can be an effective tool for enterprise operation performance evaluation.