GST is an important source of public revenue in India. Accurate forecasting of GST is important for budgeting, fiscal risk assessment, expenditure planning and policy formulation. In this research we provide an explainable machine learning framework for dependable GST revenue forecasting by integrating SHapley Additive exPlanations (SHAP) and Extreme Gradient Boosting (XGBoost). Historical GST collections were combined with economic, fiscal, administrative, policy and temporal elements. The process included data cleansing, missing-value treatment, outlier analysis, feature engineering, chronological division for training, validation and testing, hyperparameter optimisation, cross-validation, comparison modelling and explainability assessment. The suggested model was compared with Linear Regression, Decision Tree, Support Vector Regression, Random Forest and Gradient Boosting utilising Mean Absolute Error, Root Mean Squared Error, Mean Absolute Percentage Error and coefficient of determination. The results of the trials indicated that the best performing model overall was the XGBoost–SHAP model with an MAE of ?141.82 crore, RMSE of ?196.47 crore, MAPE of 4.36% and R2 of 0.928. The cross-validation R2 of 0.918 and the little performance difference between training and testing indicated stable generalisation with controlled overfitting. The SHAP analysis found that the greatest indicators were GST income from the previous month, taxable turnover, return-filing rate, GDP growth and registered taxpayers. The explanations made explicit how each variable had a role in each prediction and in what direction. The proposed framework is a suitable decision-support tool for revenue monitoring, budget preparation, compliance planning and fiscal policy evaluation due to its forecast accuracy, transparency, interpretability and stability. It is designed to supplement, not to replace, professional advice. It is sensitive to change in economic, budgetary and administrative situations.
GST Revenue Prediction; Explainable Artificial Intelligence; XGBoost; SHAP; Trustworthy Decision Support; Fiscal Forecasting.
International Journal of Trend in Scientific Research and Development - IJTSRD having
online ISSN 2456-6470. IJTSRD is a leading Open Access, Peer-Reviewed International
Journal which provides rapid publication of your research articles and aims to promote
the theory and practice along with knowledge sharing between researchers, developers,
engineers, students, and practitioners working in and around the world in many areas
like Sciences, Technology, Innovation, Engineering, Agriculture, Management and
many more and it is recommended by all Universities, review articles and short communications
in all subjects. IJTSRD running an International Journal who are proving quality
publication of peer reviewed and refereed international journals from diverse fields
that emphasizes new research, development and their applications. IJTSRD provides
an online access to exchange your research work, technical notes & surveying results
among professionals throughout the world in e-journals. IJTSRD is a fastest growing
and dynamic professional organization. The aim of this organization is to provide
access not only to world class research resources, but through its professionals
aim to bring in a significant transformation in the real of open access journals
and online publishing.