FINANCIAL DISTRESS PREDICTION USING MACHINE LEARNING

Financial distress occurs when a company or individual is unable to generate adequate revenue or money to fulfil or payback its financial commitments. This research looks at how Machine Learning can be used to identify personal financial distress. Financial frauds are a rising problem in the financial services industry with far-reaching implications, while numerous techniques have been developed, Machine Learning has been used to automate the processing of large volumes of complicated data in finance systems. In the identification of distress, Artificial Intelligence has played a significant role in the financial industry. Predicting different frauds or patterns is a big data challenge that is made more difficult by two factors: first, the profiles of normal and fraudulent behaviour vary regularly, and second, cybercrime data sets are highly skewed. This research explores and compares the performance of Different Machine Learning Models on publicly available dataset. Dataset of 15,000 individuals is sourced from public repository by Lending.com. The Algorithms are implemented on the raw and pre-processed data and the outcome of these Algorithms/Models is evaluated based on accuracy, sensitivity, specificity and precision.