Abstract: Financial Companies or firms try to determine if an individual or organization is worth lending specified amount of credit without any risk to its investors. If deemed eligible for it, they try to determine the risk associated (Probability of Default) with it. This Study compares Extreme gradient Boosting, Support Vector Machine, Naïve Bayesian, and Random Forest techniques for predicting the target variable efficiently with different strategies. This study tries to determine the risk using the person's assets, income, and various other parameters. Here, we are trying to calculate the home-credit risk factors using various parameters and compare various methods to try and determine which is more efficient and precise. Keywords: Probability of default, Credit Risk Analysis, Extreme gradient Boosting, Support Vector Machine, Naïve Bayesian, Random Forest