Analysis of Used Vehicle Value Prediction using Machine Learning

Vehicles of a specific brand, model, and year, equipped with specific features, initially have a manufacturer-set price. As they age and are traded as secondhand, their pricing is influenced by supply and demand for their specific features, along with their individual history. The more distinctive they are compared to similar cars, the more challenging it becomes to assess their value using conventional methods. Using Machine Learning (ML) algorithms to better use data on all of a car’s less prevalent attributes allows for a more accurate assessment of a vehicle’s worth. This research study analyzes the performance of Machine Learning (ML) algorithms such as Linear Regression, Ridge Regression, Lasso Regression, and Random Forest Regression in forecasting the used vehicle prices. A crucial aspect of an effective price prediction tool is its ability to incorporate depreciation, allowing for a more effective utilization of historical data in predicting current prices. This research study involved the analysis of an extensive public dataset comprising used vehicles.