This experiment applies a supervised learning approach using Linear Regression to predict house prices from a mix of numerical and categorical features. Since machine learning models require numerical input, categorical variables were converted into numerical form using binary encoding and one-hot encoding techniques.

Before training the model, feature scaling was performed using StandardScaler to standardize the data. This ensures that all input features have a similar scale, improving model stability and performance.

The dataset was divided into two parts: 70% for training and 30% for testing. The training data was used to build the regression model, while the testing data was used to evaluate how well the model generalizes to unseen data.

Model performance was assessed using multiple evaluation metrics, including Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and relative error percentage. These metrics help measure the average prediction error and the overall accuracy of the model.

A scatter plot comparing actual house prices with predicted values was generated to visually assess performance. The plot indicates that the model performs reasonably well overall, with predictions closely following the actual values. However, there is some deviation in higher price ranges, suggesting reduced accuracy for more expensive properties.