Article Preview
TopIntroduction
Information technology (IT) management has increasingly become an important factor affecting productivity (R. Baldwin, 2018;B.Begovi,2017). IT profoundly affects the economic development of a country or region. The emergence of informatization has changed the pattern of traditional enterprises and the form of competition. The emergence of informatization has led to the continuous expanding scope of competition between enterprises, the competitive environment has become more and more complex, and the competition has become more and more fierce (M. Chi& J. Zhao et al, 2017). Enterprises hope to effectively solve the difficulties through investment in IT management. The investment in IT management can not only help enterprises obtain good benefits but also help enterprises enhance the competitiveness of the industry (V., Sambamurthy&S., Venkataraman, etal, 2017). However, after a period of time, large-scale investment in IT management capabilities did not create sustainable competitive advantages for enterprises as expected (I. Muda&D. Y. Wardani etal, 2017; J. Luftman&K. Lyytinen etal, 2017). Therefore, the ability of IT management determines to a large extent whether an enterprise can have sustainable competitiveness. IT management capabilities will affect the development of a company’s sustainable competitiveness. This study uses stock information to evaluate the impact of IT management capabilities on the sustainable competitiveness of enterprises (J. A. Brink&R. L. Arenson etal, 2017;T. N. Varma & D. A. Khan, 2017).
Recently, the application of deep learning (DL) algorithms in business management has become more and more mature (Ilker and Ercanli, 2020). This type of algorithm can help companies better formulate development strategies and help companies predict future trends (L. A. Peng& J. B. Wei etal, 2020). In a general sense, the main content of stock analysis is to evaluate the fluctuations of stock prices in the stock market environment. In a broader sense, stock analysis is the analysis and evaluation of the overall change law and trend of stock prices within a certain period. Approaches to stock analysis are technical as well as statistical analysis (Q. Wang&Y. Zhu etal, 2017;A. S. Cordis & C. Kirby, 2017). These traditional methods play an active role in the development of stock analysis. Traditional stock analysis has promoted the development of the stock market to a certain extent. In recent years, stock analysis methods by computer technology have developed rapidly, and more and more scholars have begun to use DL neural networks (NNs) for stock analysis research. Deep neural network (DNN) is born since traditional NN. Compared with Artificial Neural Network (ANN), DNNs have more hidden layers (Alanis & Y. Alma, 2018;P.Bangalore & L. B. Tjernberg,2017). The principle of DNN is the multi-level processing of information by the human brain (E. Isik & M. Inalli, 2018). The biggest features of the raw data can be learned by the means of hierarchical training and solve classification problems. In addition, hierarchical initialization can be adopted to reduce the NNs’ training difficulty to a large extent (S. Li&M. Fairbank etal, 2017). As a kind of complex dynamic system, stock market contains many data that are generated during transaction. The structural features of NN can be adopted for the analysis of stock market, which is quite appropriate (M. Safa& S. Samarasinghe etal, 2018) and is unmatched by traditional analysis methods.