PREDICTION OF USED CAR PRICES

Abstract

Over the last decade, the number of cars driving on Mauritius's roads has 
actually been steadily increasing, rising by 5% annually. The National Transport 
Authority received 173,954 vehicle registrations in 2014. As a result, 1 in 6 Mauritian 
adults own a vehicle, with many of those vehicles being pre-owned or replaced 
vehicles. In this study, we want to find out whether it's possible to utilise artificial 
semantic networks to forecast the value of used vehicles. As a result, four separate 
maker learning models were given data pertaining to 200 automobiles culled from 
diverse sources. Results were somewhat better using support vector equipment 
regression compared to semantic networks and straight regression, according to our 
findings. However, for more expensive vehicles in particular, several of the predicted 
values are far off from the actual pricing. Consequently, more investigations using a 
bigger dataset are necessary, as is a great deal of trial and error with other types of 
networks and frameworks, to get much improved predictions.