This project focuses on detecting fraudulent credit card transactions using machine learning techniques. The dataset used for this project is highly imbalanced, with a small percentage of fraudulent transactions.
The dataset contains the following features:
- Time: Number of seconds elapsed between this transaction and the first transaction in the dataset.
- V1-V28: Result of a PCA transformation.
- Amount: Transaction amount.
- Class: Class label (1 for fraudulent transactions, 0 otherwise).
-
Data Preprocessing:
- Handling missing values.
- Feature scaling.
- Balancing the dataset using techniques like SMOTE.
-
Exploratory Data Analysis (EDA):
- Visualizing the distribution of the data.
- Analyzing correlations between features.
-
Model Building:
- Training various machine learning models like Logistic Regression, Decision Trees, and Random Forest.
- Evaluating model performance using metrics such as accuracy, precision, recall, and F1-score.
-
Model Evaluation:
- Comparing different models.
- Selecting the best model based on evaluation metrics.
To run this project, ensure you have the required packages installed and execute the notebook.
Refer to the requirements.txt file for a list of dependencies.