GST Classification Model This repository contains a machine learning model developed to analyze and classify data related to Goods and Services Tax (GST). The model is designed to predict outcomes based on various financial and transactional features present in the dataset. Several data preprocessing techniques, feature selection methods, and machine learning algorithms were used to optimize the model's performance.

Key Features: Data Preprocessing: Handling missing values, outlier detection and removal, normalization using Yeo-Johnson transformation, and feature scaling. Feature Engineering: Multicollinearity check using VIF, correlation analysis, and transformation of skewed features. Modeling Techniques: Logistic Regression with hyperparameter tuning using GridSearchCV, followed by cross-validation for performance evaluation. Performance Metrics: Accuracy, Confusion Matrix, Precision, Recall, and F1-Score. The project also explores the impact of dropping correlated features on model performance, as well as how to balance model complexity and accuracy. This repository provides the necessary notebooks, data, and insights used in building this classification model.