As always when starting to learn machine learning, the first model is always a regression model. In this project I was able to build a model that can predict Price of used cars. The dataset is from India and I wanted to see if certain features of a car can be used to predict its price. The pricing model that I created can effectively be used to predict used cars based on Kilometers driven, Fuel type, Transmission, Power, Engine and Mileage.
In this project I used the Random Forest Regressor model as it is able to capture complex relationships between features and the target variable without needing to specify a linear relationship. Then also with Random forest helps with reducing overfitting and providing a measure of feature importance which helps in understanding which features contribute most of to the prediction. Its good to note that Random Forests may not always be the best choice for every case particularly when the computational effciency is a major concern and also training time. The model used here takes about ten minutes averagely to train. the model had an accuracy of 76% of which was a ressult of a tuned random forest.
*For those who wish to help in making the model's accuracy to score high can email me *[email protected]. I would be grateful to learn and share ideas with you all.