A Second User Automotive Value Prediction System for Consumer’s Purchasing Using Machine Learning Approach

Abstract

Determining the value of a used automobile
can be a challenging task due to the multitude
of factors influencing its market worth. The
objective of this project is to create machine
learning models capable of accurately
predicting the value of a secondhand vehicle
based on its various attributes, facilitating
informed purchasing decisions. The global
trend of rising interest in buying and selling
used cars, underscores the urgent need for a
reliable Secondhand Vehicle Value Prediction
system that can efficiently determine the
precise value of a car by considering a range
of features. Historically, this challenge was
addressed through arbitrary pricing decisions
by sellers, leaving buyers and sellers in the
dark regarding the actual value of the vehicle
in the current market. Sellers often lacked
knowledge about the vehicle's true worth and
what price to list it for. The project's
approach involves evaluating the
performance of different regression
algorithms, including linear regression, Ridge
Regression, and Lasso Regression. The
primary objective is to provide a reliable tool
that benefits both buyers and sellers by
bringing transparency and accuracy to the
valuation of secondhand vehicles, ultimately
aiding in more informed purchasing
decisions. In essence, this project seeks to
bring transparency and precision to the
valuation process.