Price prediction of used cars using linear regression

The study is aimed at the construction of a machine learning form which would be used for the forecasting of the second-hand car price.A huge dataset that was collected from the Quikr platform underwent deep cleaning and preprocessing to ensure the used data's accuracy.There were a variety of key features such as car model, company, year of manufacture, kilometers driven, fuel type, and the history of axles owners that were investigated.After that, a linear regression model was trained on this dataset that was fixed.The model's proper performance was assessed using the mentioned metrics such as the R2 score, mean squared error and the mean absolute error, achieving a resonant exactness.Moreover, a user-friendly Streamlit application was created.The user can insert the car's model, year, kilometers, fuel type, and the number of engines in the car, and the application will give the suitable price, the case of which could be the owner's one if ownership status is taken into account and the prediction becomes thereby more reliable.The output visuals proved that the model was very good at forecasting and capable of being used in the niche of used vehicles, and they also gave a lot of useful information for both buyers and sellers.This is an example of the successful implementation of machine learning to price estimation and consequently should be a priority objective for further studies refining the accuracy of prediction.