Car Recommendation System

Predicting the value of used cars is a topic of considerable significance, as it requires specialized knowledge and expertise. There is a growing global trend in the sale of used cars, driven by increasing prices of new cars and consumers’ financial constraints, leading them to seek more affordable options in the used car market. This paper introduces a machine learning-based car recommendation system employing a Decision Tree Classifier. Utilizing a dataset encompassing diverse attributes such as price, engine power, body type, transmission type, year, and mileage. Categorical features undergo encoding via LabelEncoder to prepare the dataset for model training. The Decision Tree Classifier is trained using grid search and cross-validation techniques to fine-tune hyperparameters and enhance model performance. The model achieves an accuracy of $\mathbf{8 6. 4 3 \%}$ on the test set, highlighting its efficacy in predicting car models based on input features. Furthermore, precision, recall, and F1-score metrics are computed to provide a comprehensive evaluation of the model’s performance. Overall, the findings underscore the efficiency of the proposed car recommendation system in accurately predicting car models based on input features, offering valuable insights for automotive consumers.