Published Paper Details:

The rapid expansion of the electric vehicle market has increased the need for transparent and accurate cost estimation methods for consumers, manufacturers, and policymakers. This paper presents a machine learning-based framework for the analysis and prediction of electric vehicle costs using technical and market-related vehicle attributes. A structured dataset containing features such as brand, acceleration, top speed, battery range, energy efficiency, fast-charging capability, plug type, body style, market segment, and seating capacity was used for model development. Data preprocessing techniques including missing value handling, feature encoding, normalization, and train-test splitting were applied to improve prediction quality. Multiple supervised learning algorithms, namely Linear Regression, Decision Tree, Random Forest, Support Vector Machine, and Artificial Neural Network, were trained and compared using standard regression metrics such as RMSE and R-squared. Experimental results showed that Random Forest and ANN models achieved superior predictive performance compared to traditional approaches, with Random Forest offering an effective balance between accuracy and computational efficiency. A web-based user interface was also developed to allow users to enter vehicle specifications and obtain real-time estimated costs. The proposed system demonstrates that machine learning can provide reliable and practical support for EV price estimation, helping reduce uncertainty in purchasing decisions and market analysis.