Bankruptcy Prediction Using Machine Learning: A New Technological Approach to Prevent Corporate Bankruptcy Through Well Deployed Streamlit Based Application

Abstract

Corporate bankruptcy prevention and prediction is the most significant problem in the finance domain. The successful machine learning based predictive model allows the corporate stakeholders to check the status of their business. It claims that the person or the organization is the debtor. The proposed research study focused on building a machine learning model to predict the bankruptcy and deploy ML model by using Streamlit, the open-source Python library. The ML framework helps to accept all the values of independent parameters and predict the corporate bankruptcy which leads to early actions to avoid economic losses. Bankruptcy prediction is a classification problem (Bankrupt / Non-Bankrupt). Since the variable to predict is binary. The predictive models are built by applying several machine learning algorithms such as SVM, KNN, Naive Bayes and CART. We find that SVM model with Polynomial Kernel which achieves a high degree of accuracy in all applied ML models. The SVM model with 96.00% model accuracy and 4% error rate is selected for prediction purposes. The SVM model then deployed with the help of Streamlit library to check bankruptcy classification. This application helps stakeholders to prevent their business from bankruptcy by checking through it in early stages. User has to just input the values and our model immediately displays the prediction of bankruptcy either bankrupt(0) or non-bankrupt (1).