• Prerana Bharati
• Vishal D Hanchate
• Manoj D
• Sujit Gani
• Shakkeera L

The paper, titled "Income Tax Fraud Detection Using AI-ML," investigates the integration of Artificial Intelligence (AI) and Machine Learning (ML) to identify income tax fraud. With the rising challenge of tax evasion, advanced technologies are crucial for early detection. The study focuses on developing predictive models using supervised learning algorithms like Linear Regression, Decision Trees, Random Forest, SVM, KNN, Gradient Boosting, and Neural Networks. Feature engineering techniques, including label encoding and standardization, optimize model performance. The report includes exploratory data analysis, outlier detection, and correlation analysis to ensure dataset quality. Model evaluations, using metrics like Mean Squared Error and R-squared, provide insights into model accuracy. The Income Tax Fraud Detection system's user interface is implemented through Streamlit, enabling users to input financial parameters for predictions. The report concludes by identifying the best-performing model, deployed for real-time fraud detection. This research strengthens financial systems against fraud using AI and ML, providing valuable insights into the feasibility and effectiveness of predictive analytics for income tax fraud detection. Notably, XGBoost demonstrates exceptional accuracy, achieving 0.9973, surpassing all other models, which have a combined average accuracy of 0.7437.

Income Tax Fraud Detection; Artificial Intelligence(AI); Machine Learning(ML); Predictive Models; Decision Trees; Random Forest; Support Vector Machine(SVM); k-Nearest Neighbors(KNN); Anomaly Detection; Gradient Boosting.

"The Secure Framework to Develop Income Tax Fraud Detection using AI-ML Techniques", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 1, page no.b1-b8, January-2024, Available :http://www.jetir.org/papers/JETIR2401101.pdf

"The Secure Framework to Develop Income Tax Fraud Detection using AI-ML Techniques", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 1, page no. ppb1-b8, January-2024, Available at : http://www.jetir.org/papers/JETIR2401101.pdf