Research On Business Performance Of Internet Of Things Listed Companies Based On BP Neural Network

Since the 21st century,under the background of the large-scale development of the Internet of Things,IoT listed companies are the most important part of the Internet of Things industry.During the period of rapid development of the industry,they face many development problems,including long-term problems such as low-quality competition,lack of upgrades and confusion in development direction,short-term problems such as shortage of scale awareness and lack of operational confidence.This article starts with the operating performance of listed companies and refers to domestic and abroad research on the performance of listed companies in the Internet of Things industry and other industries.The 27 data of 75 listed companies in the Internet of Things industry in China are comprehensively evaluated and scored,and training an accurate and scientific BP neural network model,which can not only effectively predict enterprise performance scores and rankings,but also has good extensibility for enterprises in the Internet of Things industry outside the model.In the process of data selection and analysis,the data selected in this paper is the public data of the listed company's 2016-2017 annual report,which is divided into two categories:financial indicators and non-financial indicators.Including seven first-level categories such as risk control capability class,cash flow capability class,profitability class,operational capability class,expansion capability class,scientific research capability class,management competence class;and 27 specific categories such as net interest rate,main income growth rate,cash flow ratio,equity concentration,number of board members,and average executive compensation.After data standardization,pre-processing and factor analysis,they are finally aggregated into seven main factors,and its comprehensive score is the business performance score of the enterprise.Considering the scalability of other IoT enterprises that have not yet joined the model,the original data is used instead of the main factor to join the BP neural network model.After two optimizations,finally,an excellent model with high accuracy,low mean square error and low average error is formed.The excellent model,when applying the newly added data for testing,achieved 96% accuracy for the ranking prediction,and the average deviation of the score prediction was 0.0012 points,which fully reflected the feasibility and adaptability of the model.The research results show that the business performance of China's Internet of Things listed companies is mainly affected by seven factors:risk control,cash flow,profitability,operational capability,expansion capability,scientific research capability and management capability.There is a very significant correlation between these seven factors and business performance,with a correlation coefficient of 0.923.China's Internet of Things listed companies should identify their own advantages and disadvantages,grasp the dividend of the rapid development of the Internet of Things,steadily improve their competitiveness,adapt to the development of the times,find their own positioning,and strive to create higher value for enterprises,investors and society.