I. TASK I: ENSEMBLE LEARNING FOR CAR PRICE PREDICTION

In this task, you will explore the concept of ensemble learning and apply it to solve the car price

prediction problem. Ensemble learning involves combining the predictions of multiple individual models

to make a final prediction. For this task, you will use multiple linear regression models as the base models.

A$.$ Dataset and Analysis

$1)$ The dataset is provided in $$Dataset$/$CarPrice$.$csv$$;

$2)$ Print info, description, and pairwise correlation of columns.

$3)$ Extract the following columns as features:

$[$symboling$,$fueltype$,$doornumber$,$carbody$,$carheight$,$stroke$,$compressionratio$,$horsepower$$

$,$peakrpm$]$

$4)$ Extract $$price$$ column as labels;

$5)$ Transform non$-$numeric features to numeric features and discuss your solutions;

$6)$ Split the dataset into training and testing sets.

B$.$ Ensemble Learning with Linear Regression

$1)$ Perform bagging or boosting $($Choose one of them as an implementation for ensemble learning$)$:

$1.$ Bagging: Train multiple linear regression with different parameters as base models on the training

set and combine the predictions of all the individual regression models to make a final prediction

using a suitable bagging method, such as averaging the predictions or using a voting mechanism.

$2.$ Boosting: Applying a boosting algorithm to sequentially train a series of regression models, where

each subsequent model focuses on correcting the mistakes made by the previous models.

$2)$ Evaluate the performance of the ensemble model on the testing set using appropriate evaluation

metrics, such as mean squared error or mean absolute error.

C$.$ Analysis and Discussion

$1)$ Compare the performance of the ensemble model with the individual linear regression model.

$2)$ Discuss the benefits of using ensemble learning in this context.

$3)$ Reflect on the factors that may influence the performance of the ensemble model, such as the choice

of base models, diversity among the models, and the ensemble method used.