A machine learning model in Python that classifies the relationship between the efficiency of different engine models and other car characteristics. This project also has a machine learning model that tries to predict the values of cars and the relationship between engine models and horsepower.
Importance score for each variable
The variable importance score in the context of the random forest algorithm indicates the importance of each feature (variable) in predicting the target variable, which, in this case, is the price of the car. In other words, a variable is considered important if removing it from the model results in worse predictions. The importance of the variable is calculated for all trees in the forest and then the average is taken 5.
![](https://github.com/Brevex/US-Used-Cars-Data-Analysis/raw/54f75e2de4f55d668ba495bef7ac14ec4011e24b/readme%20images/feature%20importance.png)
Average horsepower by engine cylinders
Data analysis showing the average horse power value of each engine model in the dataset
![](https://github.com/Brevex/US-Used-Cars-Data-Analysis/raw/54f75e2de4f55d668ba495bef7ac14ec4011e24b/readme%20images/horse%20power.png)
Average horsepower score per engine
Machine learning model that attempts to establish a relationship between the engine model and the amount of power. The model uses a random forest tree process to assign a score to each model.
![](https://github.com/Brevex/US-Used-Cars-Data-Analysis/raw/54f75e2de4f55d668ba495bef7ac14ec4011e24b/readme%20images/engines%20score.png)
Price prediction
The scatterplot shows the result of a machine learning model that tries to predict the price of cars. As shown in the graph, only a minority of the values deviate from those predicted by the prediction model, showing a good frequency of hits with approximate values.
![](https://github.com/Brevex/US-Used-Cars-Data-Analysis/raw/54f75e2de4f55d668ba495bef7ac14ec4011e24b/readme%20images/price%20prediction.png)