Confusion Matrix
A Confusion Matrix summarizes the model's predictions by comparing the predicted labels with the actual labels.
It consists of four values:
- True Positive (TP): Fake reviews correctly classified as fake.
- True Negative (TN): Genuine reviews correctly classified as genuine.
- False Positive (FP): Genuine reviews incorrectly classified as fake.
- False Negative (FN): Fake reviews incorrectly classified as genuine.
The Confusion Matrix helps identify the types of errors made by the model.
Generate the Confusion Matrix
from sklearn.metrics import confusion_matrix
cm = confusion_matrix(y_test, y_pred)
cm
Visualize the Confusion Matrix
plt.figure(figsize=(6,5))
sns.heatmap(
cm,
annot=True,
fmt='d',
cmap='Blues',
xticklabels=['Genuine', 'Fake'],
yticklabels=['Genuine', 'Fake']
)
plt.xlabel("Predicted Label")
plt.ylabel("Actual Label")
plt.title("Confusion Matrix")
plt.show()
Explanation
The diagonal values represent correctly classified reviews, while the off-diagonal values indicate misclassified reviews.
A model with higher diagonal values generally performs better.
Fake Review Detection System using Machine Learning
VA









