Predicting Bankruptcy and Suggesting Improvements on Financial Attributes using Machine Learning Models

Many financial players, including business investors, market analysts, banking experts, politicians, and shareholders, have been paying attention to the subject of corporate bankruptcy in recent years. Predicting company failure is crucial not just for financial institution decision-making, but it also influences the country’s financial suffering to some extent, since poor financial institution decision-making can have devastating consequences on a national or even global scale. The proposed system uses the Random Forest Model on the Polish Bankruptcy dataset to predict whether a company will go bankrupt in the near future. The dataset contains 10000 records and 64 attributes and which help in classifying a company as bankrupt or not bankrupt with an accuracy of 97.35%. The system has also incorporated the linear regression model to determine what should be the value of a particular attribute so that the financials of the company remain stable and it won’t go bankrupt. This system will help prevent bad-debts that have an impact on the economy while also making sure that the company remains afloat by checking its status regularly.