Principal Component Analysis (PCA)¶
Principal Component Analysis (PCA) is a statistical procedure that uses an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components.
In simpler terms, PCA is a technique used to reduce the number of dimensions (features) in a dataset while retaining as much of the original variability (information) as possible.
In [3]:
import pandas as pd
from sklearn.preprocessing import StandardScaler
from sklearn.decomposition import PCA
import matplotlib.pyplot as plt
# Load data
df = pd.read_csv('datasets/StudentsPerformance.csv')
# Select numeric columns only
numeric_df = df.select_dtypes(include='number')
# Standardize numeric data
scaler = StandardScaler()
scaled_data = scaler.fit_transform(numeric_df)
# Apply PCA (2 components)
pca = PCA(n_components=2)
pca_result = pca.fit_transform(scaled_data)
# Add PCA results back to a dataframe
pca_df = pd.DataFrame({
'PC1': pca_result[:, 0],
'PC2': pca_result[:, 1]
})
print(pca_df.head())
# Plot PCA scatter
plt.figure(figsize=(6, 5))
plt.scatter(pca_df['PC1'], pca_df['PC2'])
plt.xlabel('PC1')
plt.ylabel('PC2')
plt.title('PCA Scatter Plot')
plt.show()
PC1 PC2 0 0.560514 0.088285 1 1.719201 -0.910745 2 2.883135 -0.021999 3 -2.119921 -0.074994 4 0.988094 0.131914
