The F1 Score is a model evaluation metric used mainly in classification problems to measure the balance between:

• Precision
• Recall

It is especially useful when:

• Data is imbalanced
• False positives and false negatives are both important

---

Simple Definition

The F1 Score is the harmonic mean of Precision and Recall.

Formula:

$F1 = 2 \times \frac{Precision \times Recall}{Precision + Recall}$F1=2×Precision+RecallPrecision×Recall​

---

Understanding Precision and Recall

Precision

Out of all predicted positive cases, how many were actually correct?

Formula:

$Precision = \frac{TP}{TP + FP}$Precision=TP+FPTP​

Where:

• TP = True Positives
• FP = False Positives

Example

If a fraud detection model flags 100 transactions as fraud, and only 80 are actually fraud:

Precision = 80%

---

Recall

Out of all actual positive cases, how many did the model correctly identify?

Formula:

$Recall = \frac{TP}{TP + FN}$Recall=TP+FNTP​

Where:

• FN = False Negatives

Example

If there were actually 100 fraudulent transactions, and model identified 80:

Recall = 80%

---

Why F1 Score is Important

Sometimes:

• Precision is high but recall is low
• Recall is high but precision is low

F1 Score balances both.

---

Example

Suppose:

Metric	Value
Precision	0.8
Recall	0.6

Then:

$F1 = 2 \times \frac{0.8 \times 0.6}{0.8 + 0.6}$F1=2×0.8+0.60.8×0.6​

Result:

F1 = 0.6857

---

Interpretation

F1 Score	Meaning
1.0	Perfect model
0.8+	Very good
0.5	Moderate
0	Poor

Higher F1 score means:

• Better balance between precision and recall
• Better classification performance

---

When to Use F1 Score

Use F1 Score When:

• Dataset is imbalanced
• Fraud detection
• Medical diagnosis
• Spam detection
• Risk prediction
• Rare event prediction

---

Real-Life Example

Healthcare Example

Suppose AI predicts cancer.

False Negative

Patient actually has cancer but model says no.

Very dangerous.

False Positive

Patient doesn’t have cancer but model says yes.

Still problematic.

You need both:

• High precision
• High recall

Hence F1 Score becomes important.

---

Difference Between Accuracy and F1 Score

Accuracy	F1 Score
Overall correctness	Balance of precision & recall
Bad for imbalanced data	Better for imbalanced data
Can be misleading	More realistic

---

Example of Accuracy Problem

Suppose:

• 990 non-fraud transactions
• 10 fraud transactions

Model predicts:

• Everything as non-fraud

Accuracy:

990 / 1000 = 99%

Looks great — but model found ZERO frauds.

F1 Score would expose this weakness.

---

Friendly Answer

“F1 Score is a machine learning evaluation metric used for classification problems. It is the harmonic mean of precision and recall and helps measure the balance between false positives and false negatives. F1 Score is especially useful when working with imbalanced datasets such as fraud detection, healthcare diagnosis, and anomaly detection.”