Optimized Deep Learning Model to Predict Business Bankruptcy

This article studies the ideal and optimized way to predict a company's bankruptcy according to the internal and external environment. Likewise, prognostic factors were identified through machine learning models that are ideal for this type of case. In this context, the choice of the ideal model is the basis for making improvements and updates that allow the optimization of a specific model. For this reason, the optimization of the model is considered a specialized process that allows the precise identification of the factors that lead to the bankruptcy of companies and the identification of the necessary correlations between relevant variables. In this sense, past works on similar contexts were considered for the present study, the analysis method where the methodology is required, and three stages: Data exploration, selection of models, and implementation. Likewise, to determine the results, training was considered to obtain results from the model with optimized characteristics, company bankruptcy factors and correlations, and finally the discussion of results and conclusions was specified.