This project involves the creation of a banking application utilizing machine learning algorithms.
The project's objective, as outlined in the provided description, was to develop a predictive model capable of identifying potential defaulters among a bank's customers based on their EMI payment behavior.
The project's focus was on assessing the risk associated with transactions involving the transfer of funds from bank accounts.
Regarding the dataset, a comprehensive dataset was furnished, containing various columns, each accompanied by a corresponding description. Both training and testing datasets were made available. Preprocessing tasks, including data cleaning (removal of empty cells, etc.), were conducted using the Pandas library. Data visualization techniques were also applied to gain insights into the dataset's distribution and characteristics.
For the modeling phase, multiple algorithms were employed, including Decision Trees, Logistic Regression, and Hyperparameter tuning. After experimentation, it was deduced that Hyperparameter tuning resulted in higher accuracy compared to the other methods.
To access the dataset, please refer to the following link: Dataset Link.