• Roshan Kumar Jha
• Mukesh Sah
• Rinku Kumari
• Abhishek Kumar
• Latha A

Abstract— Managing personal finances is increasingly difficult due to rising living costs, complex financial systems, and the lack of unified tracking tools. Traditional budgeting methods rely on manual input and offer only retrospective analysis, without forecasting future risks or supporting cryptocurrency assets. This work presents FINSIGHT, an AI-driven financial management platform that automates transaction processing, predicts expenses, and integrates blockchain-based asset tracking. The system parses data from PDFs, CSVs, and Excel files, and uses machine learning for transaction categorization. LSTM and GRU models were evaluated for forecasting, achieving 92% and 90% accuracy respectively, with LSTM delivering lower error rates and GRU offering faster inference. Integration with the Solana blockchain enables real-time monitoring of digital and traditional assets. A React–FastAPI dashboard provides visualizations and personalized financial insights. Experiments on 24-month synthetic datasets across 200 user profiles confirm strong performance in forecasting and spending pattern recognition, outperforming conventional finance tools. Index Terms— Financial forecasting, LSTM, GRU, blockchain, personal finance, predictive analytics.

Personal finance management, Financial forecasting, Deep learning, LSTM networks, GRU models, Predictive analytics, Blockchain integration, Cryptocurrency tracking, Automated budgeting, Time-series analysis, Transaction categorization, Web3.js, Solana blockchain, AI-driven financial systems, Financial data parsing

"FINSIGHT: An AI-Powered Personal Finance Management System with Predictive Analytics and Blockchain Integration", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 12, page no.a320-a325, December-2025, Available :http://www.jetir.org/papers/JETIR2512044.pdf

"FINSIGHT: An AI-Powered Personal Finance Management System with Predictive Analytics and Blockchain Integration", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 12, page no. ppa320-a325, December-2025, Available at : http://www.jetir.org/papers/JETIR2512044.pdf