EXPENSE TRACKER USING MACHINE LEARNING

ABSTRACT: Effective expense tracking is vital for personal and business financial management, helping monitor spending, budget efficiently, and uncover cost- saving opportunities. Traditional methods, often manual and rule-based, are time-consuming, error-prone, and lack scalability. This paper introduces a machine learning-based expense tracking system that automates and enhances categorization using supervised learning algorithms. Specifically, the Naïve Bayes Algorithm processes unstructured transaction data to improve categorization accuracy. Additionally, unsupervised learning techniques such as clustering are applied to detect spending patterns and anomalies, offering deeper financial insights. The system significantly reduces manual effort, increases accuracy, and enables predictive forecasting of future expenses. Experimental evaluations show marked improvements over traditional approaches in efficiency and precision. This research demonstrates the transformative potential of machine learning in expense tracking, equipping users with intelligent, data-driven tools for better financial decision-making. KEYWORDS: Expense Tracking, Machine Learning, Naïve Bayes Algorithm,Transaction Categorization, Financial Behaviour Analysis, Predictive Analytics.