Linear Regression is a supervised machine learning algorithm used for predicting a continuous dependent variable
based on one or more independent variables (features). It models the relationship between variables by fitting a
linear equation to observed data.
Objective: Import the dataset to work with.
Sources: CSV, Excel, SQL, or web-based datasets.
Tools: pandas.read_csv() or similar functions.
Objective: Understand the structure of data and clean it.
🔍 Data Exploration
View top rows: data.head()
Data types: data.dtypes
Shape: data.shape
Summary statistics: data.describe()
🧹 Data Cleaning
Handle missing values: fillna(), dropna()
Remove duplicates
Convert categorical to numeric (if needed): One-hot encoding or Label encoding
📐 Feature Selection
Choose relevant independent variables (features) and the dependent variable (target).
Objective: Separate data into training and testing sets to evaluate generalization.
Tool: train_test_split() from sklearn.model_selection
Objective: Train the model on the training data.
Tool: LinearRegression from sklearn.linear_model
Objective: Make predictions and compare them to actual values.
Objective: Quantify how well the model performs.
📏 Common Metrics:
Mean Absolute Error (MAE)
Mean Squared Error (MSE)
Root Mean Squared Error (RMSE)
R-squared Score (R²)
| Step | Description |
|---|---|
| Load Dataset | Import using pandas or other libraries |
| View/Preprocess | Clean data, handle nulls, transform features |
| Split Dataset | Training vs testing data |
| Build Model | Train Linear Regression on training data |
| Test/Evaluate | Predict and compare with actual test data |
| Performance Analysis | Use MAE, MSE, RMSE, R² for evaluation |