Introduction to Confusion Matrix in Python | Classification Evaluation

Learn the introduction to the confusion matrix in Python — what a confusion matrix is, how it evaluates classification models, and how to create one with examples using scikit‑learn metrics.

A confusion matrix is a matrix used to evaluate the performance of a classification model in machine learning, data analytics, and statistics.

It shows how many predictions the model got right and wrong by comparing actual values with predicted values.

Confusion Matrix (Binary Classification)

Meaning of Each Term

• True Positive (TP) Model correctly predicts Positive (e.g., predicts “spam” and it is spam)

• True Negative (TN) Model correctly predicts Negative (predicts “not spam” and it is not spam)

• False Positive (FP) Model predicts Positive, but actual is Negative (also called Type-I error)

• False Negative (FN) Model predicts Negative, but actual is Positive (also called Type-II error)

Why Confusion Matrix Is Important

It helps us calculate key evaluation metrics:

• Accuracy [ \frac{TP + TN}{TP + TN + FP + FN} ]

• Precision [ \frac{TP}{TP + FP} ]

• Recall (Sensitivity) [ \frac{TP}{TP + FN} ]

• F1-Score Harmonic mean of Precision and Recall

Simple Example

Suppose we have 100 students and predict whether they passed or failed:

• TP = 40 (correctly predicted pass)
• TN = 45 (correctly predicted fail)
• FP = 5 (predicted pass but actually failed)
• FN = 10 (predicted fail but actually passed)

This information forms the confusion matrix.

Where It Is Used

• Logistic Regression
• Decision Trees
• Random Forest
• Support Vector Machines (SVM)
• Deep Learning classification models

Example 1: Confusion Matrix using sklearn

Step 1: Import Libraries

from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay
import matplotlib.pyplot as plt

Step 2: Actual and Predicted Values

# Actual class labels
y_true = [1, 0, 1, 1, 0, 1, 0, 0, 1, 0]

# Predicted class labels
y_pred = [1, 0, 0, 1, 0, 1, 1, 0, 1, 0]

Step 3: Create Confusion Matrix

cm = confusion_matrix(y_true, y_pred)
print(cm)

Output

[[4 1]
 [1 4]]

This means:

• TN = 4
• FP = 1
• FN = 1
• TP = 4

Step 4: Visualize Confusion Matrix

disp = ConfusionMatrixDisplay(confusion_matrix=cm)
disp.plot()
plt.show()

Example 2: Confusion Matrix with a Classification Model

Using Logistic Regression

from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import confusion_matrix

Create Dataset

X, y = make_classification(n_samples=100, random_state=42)

X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)

Train Model and Predict

model = LogisticRegression()
model.fit(X_train, y_train)

y_pred = model.predict(X_test)

Generate Confusion Matrix

cm = confusion_matrix(y_test, y_pred)
print(cm)

Calculate Accuracy, Precision, Recall

from sklearn.metrics import accuracy_score, precision_score, recall_score

print("Accuracy:", accuracy_score(y_test, y_pred))
print("Precision:", precision_score(y_test, y_pred))
print("Recall:", recall_score(y_test, y_pred))

Key Takeaway

Confusion matrix is used to evaluate classification models by showing True Positives, True Negatives, False Positives, and False Negatives.

📘 Related Topics

• [Logistic Regression in Python /

  Learn Logistic Regression with Examples](/python/docs/statistics/logistic-regression.html)

• [Introduction to Regression & Regression Models in Python /

  Tutorial & Examples](/python/docs/statistics/regression-model.html)