Skip to content

06 Feature Engineering

View code on GitHub Get the Video Course

Feature engineering creates new variables that help the model detect patterns more effectively. This section introduces amount_ratio, a feature that captures how much a transaction deviates from the customer's normal spending pattern. Fraudsters typically spend far more than the victim's historical average, making this ratio a potential fraud indicator.

What is Feature Engineering?

Section 06 adds amount_ratio to the 8 baseline features, making 9. Legitimate: $80 / $100 = 0.8x, normal spending. Fraud: $2,500 / $100 = 25x, 25 times normal spending.

We go from 8 input features (baseline) to 9 input features by adding amount_ratio. The diagram above shows the before/after comparison and how amount_ratio is calculated: transaction_amount / avg_transaction_amount. A legitimate transaction gives a ratio near 1.0x (normal spending), while a fraud transaction gives 25x (suspicious).

ML and CCFD Workflow

Two workflows. Top: Basic ML Workflow from Data Collection to Monitoring & Retraining, with a feedback loop. Bottom: our scripts p1_01 to p5_04, with APIs on Port 8000 and 8001.

Architecture

p4_01 adds amount_ratio = txn_amount / avg_amount (8 features to 9, for example $2500 / $100 = 25.0x). p4_02 compares legit vs fraud; p4_03 trains and saves model_latest.pkl.

Table of Contents

Step Topic
Step-01 Understand the amount_ratio Feature
Step-02 Create the Feature
Step-03 Analyze the Feature
Step-04 Train Model with amount_ratio
Step-05 Compare Results

Pre-requisite: Python Environment Setup

Terminal
# Create conda environment
conda create -n mlops-env1 python=3.14 -c conda-forge -y

# Activate environment
conda activate mlops-env1

# Install dependencies (locked versions)
cd ccfd-project
pip install -r requirements.txt

Note: All sections in this course use the same mlops-env1 environment. You only need to create it once. After that, just activate it with conda activate mlops-env1 before running any scripts.


Step-01: Understand the amount_ratio Feature

Formula

amount_ratio = transaction_amount / avg_transaction_amount

Why This Matters for Fraud Detection

  • A customer with avg spending of $100 making a $50 purchase → ratio = 0.5 (normal)
  • Same customer making a $2,500 purchase → ratio = 25.0 (suspicious!)
  • Fraudsters don't know the victim's spending patterns, so they tend to spend much more than normal

Example Calculations

Customer Avg Spend Transaction amount_ratio Interpretation
Alice $100 $50 0.5 Half normal spending
Alice $100 $100 1.0 Normal spending
Alice $100 $2,500 25.0 25x normal - suspicious!
Bob $500 $600 1.2 Slightly above normal
Bob $500 $3,000 6.0 6x normal - suspicious!

Step-02: Create the Feature

Prerequisite: Each section has its own ccfd-project/ folder. You must generate data first before running feature engineering.

File Description
ccfd-project/p4_01_feature_engineering_fe1.py Creates amount_ratio and saves new dataset
Terminal
cd 06_Feature_Engineering/ccfd-project
python p1_01_generate_initial_dataset.py    # Generate data (required first)
python p4_01_feature_engineering_fe1.py     # Create amount_ratio feature

Alternative: If you're using the shared folder, run all commands from ccfd-project-main/ instead. See Shared Project Folder in the root README.

Expected Output:

======================================================================
FEATURE ENGINEERING: Creating amount_ratio Feature
======================================================================

----------------------------------------------------------------------
STEP 1: Loading Raw Data
----------------------------------------------------------------------
Loaded 10,000 transactions (9 columns)

----------------------------------------------------------------------
STEP 2: Applying Feature Engineering
----------------------------------------------------------------------
Added: amount_ratio (9 total features)
Saved: data/credit_card_transactions_fe1_amount_ratio.csv

----------------------------------------------------------------------
STEP 3: Verifying Results
----------------------------------------------------------------------
Raw columns: 9 → FE columns: 10
New feature added: amount_ratio

======================================================================
FEATURE ENGINEERING COMPLETE!
======================================================================

   Output: data/credit_card_transactions_fe1_amount_ratio.csv

Step-03: Analyze the Feature

File Description
ccfd-project/p4_02_eda_amount_ratio_fe1.py Analyzes amount_ratio distribution and correlation
Terminal
python p4_02_eda_amount_ratio_fe1.py

Expected Output:

======================================================================
EDA FOR FEATURE ENGINEERING 1: AMOUNT RATIO
======================================================================

======================================================================
SECTION 2: DISTRIBUTION ANALYSIS
======================================================================

   Legitimate mean: 2.81
   Fraud mean:      12.10
   Difference:      9.29   GOOD separation!
   Saved: eda_plots/07_fe1_amount_ratio_distribution.png

======================================================================
SECTION 3: CORRELATION COMPARISON
======================================================================

   transaction_amount correlation: 0.3111
   amount_ratio correlation:       0.2415
   Saved: eda_plots/08_fe1_amount_ratio_fraud_comparison.png
   Saved: eda_plots/09_fe1_amount_ratio_correlation.png

======================================================================
FE1 EDA COMPLETE!
======================================================================

Feature Analysis Results

Metric Legitimate Fraud Difference
amount_ratio mean 2.81 12.10 +9.29
Interpretation ~3x normal spend ~12x normal spend Fraudsters spend 4x more relative to victim's average

Generated Plots

File Description
eda_plots/07_fe1_amount_ratio_distribution.png Distribution comparison: legit vs fraud
eda_plots/08_fe1_amount_ratio_fraud_comparison.png Fraud rate by range + scatter plot
eda_plots/09_fe1_amount_ratio_correlation.png Updated correlation heatmap

amount_ratio Distribution (Legit vs Fraud)

Histogram and box plot of amount_ratio for Legitimate and Fraud. Most Legitimate values are small. Fraud values spread much higher, with a higher median and a taller box.

Fraud Rate by amount_ratio Range

Fraud rate by amount_ratio range rises from 0.9% (2-3) to 5.5% (5-10) and 17.8% (10+). A scatter plot shows fraud points spread toward high amount_ratio.

Updated Correlation Heatmap (with amount_ratio)

Correlation heatmap with the new amount_ratio feature. It has 0.24 correlation with is_fraud, below transaction_amount at 0.31, and 0.66 with transaction_amount.


Step-04: Train Model with amount_ratio

File Description
ccfd-project/p4_03_train_model_with_fe1.py Trains Logistic Regression with 9 features (includes amount_ratio)
Terminal
python p4_03_train_model_with_fe1.py

Expected Output:

======================================================================
Logistic Regression (FE1: amount_ratio) - PERFORMANCE METRICS
======================================================================

TECHNICAL METRICS:
   • F1 Score:     0.2347
   • Precision:    0.1386
   • Recall:       0.7667
   • ROC AUC:      0.8837

BUSINESS METRICS:
   • Net Benefit:    $39,700
   • Fraud Caught:   46
   • Fraud Missed:   14
   • False Alarms:   286

======================================================================
TRAINING SUMMARY
======================================================================

   Model: Logistic Regression (FE1)
   Features: 9 (includes amount_ratio)

   F1 Score: 0.2347
   Net Benefit: $39,700

======================================================================
FE1 TRAINING COMPLETE!
======================================================================

Step-05: Compare Results - Baseline vs Feature Engineered

Metric Baseline (p3_01) With FE1 (p4_03) Change
F1 Score 0.2329 0.2347 +0.8%
Precision 0.1373 0.1386 +0.9%
Recall 0.7667 0.7667 0%
Net Benefit $39,625 $39,700 +$75
False Alarms 289 286 -3

Why Minimal Improvement?

The improvement is minimal (+$75) because:

  1. Logistic Regression already captures amount patterns - The raw transaction_amount feature already has strong predictive power (+0.31 correlation)
  2. amount_ratio provides redundant information - Since it's derived from transaction_amount and avg_transaction_amount, Logistic Regression can learn similar patterns from the raw features
  3. More value comes from hyperparameter tuning - Section 07 shows that threshold optimization provides much larger improvements than feature engineering alone
  4. Small dataset (10,000 synthetic transactions) - Real production fraud systems process millions of transactions per day. With only 60 fraud cases in our test set, small improvements from a single engineered feature get lost in statistical noise. At production scale, even a 0.5% lift translates to substantial business value (millions of transactions x ~$1,500 average fraud loss prevented).
  5. Limited feature space (8 base features, 1 engineered) - Real-world fraud systems use 100+ input features (device fingerprint, IP velocity, merchant risk scores, transaction sequences, user behavior history) and typically ship 10-20+ engineered features over months of iteration. Our amount_ratio is illustrative of the technique, not representative of what production feature engineering looks like end-to-end.

Next Steps

Next Topic What You'll Do
07_Hyperparameter_Tuning Optimize Parameters Search for best parameters, optimize threshold, train final model

Prefer to learn by watching?

The video course builds this whole project with you on screen, step by step.

Get the Video Course


05 EDA and Preprocessing Next: 07 Hyperparameter Tuning

🎉 New Course
Machine Learning End-to-End: Credit Card Fraud Detection
$15.99 $84.99 81% OFF
Get the Video Course