AI-Powered Personal Finance Assistant for Intelligent Expense Analysis, Budgeting, and Personalized Financial Insights

Abstract

Individuals managing personal income and expenditure without structured tools often rely on manual record-keeping or generic spreadsheets, which provide limited categorization, no proactive budget feedback, and no personalized analysis of spending behaviour. This paper presents an AI-Powered Personal Finance Assistant, a web-based system integrating income and expense tracking, automated expense categorization, category-wise budget monitoring, savings-goal tracking, a transparent financial health score, data-grounded insight generation, a rule-based conversational query assistant, and general investment-education content. The system is implemented using a Flask backend, a React frontend, and a relational SQLite database. Every AI-labelled feature is built as explainable, rule-based logic rather than a trained model or a paid external API, avoiding the cold-start problem that limits machinelearning-based categorization for new users. The implemented prototype was functionally verified module by module; no controlled multi-user evaluation has yet been performed, and this is reported explicitly rather than replaced with fabricated metrics. The system is positioned as a decision-support and financial-literacy tool, not a substitute for professional financial advice, and makes no guarantee of investment return.