-
Notifications
You must be signed in to change notification settings - Fork 337
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Anomaly Detection in Transactions using Deep Learning added
- Loading branch information
Showing
3 changed files
with
10,699 additions
and
0 deletions.
There are no files selected for viewing
660 changes: 660 additions & 0 deletions
660
...nsactions using Deep Learning/Anomaly_Detection_in_Transactions_using_Deep_Learning.ipynb
Large diffs are not rendered by default.
Oops, something went wrong.
38 changes: 38 additions & 0 deletions
38
Anomaly Detection in Transactions using Deep Learning/readme.md
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,38 @@ | ||
# Anomaly Detection in Transactions using Deep Learning | ||
|
||
## 🎯 Goal | ||
The main goal of this project is to develop models to detect anomalies in financial transactions using three different deep learning algorithms: Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN), and Multi-Layer Perceptrons (MLP). The aim is to identify fraudulent transactions effectively. | ||
|
||
## 🧵 Dataset | ||
The dataset used for this project is `transactions.csv`, which contains synthetic transaction data with features like transaction amount, transaction type, account age, transaction location, and a binary label indicating if the transaction is fraudulent. | ||
|
||
## 🧾 Description | ||
This project involves the implementation of three different deep learning algorithms for anomaly detection in transactions: RNN, CNN, and MLP. Each model is trained to identify patterns in the transaction data and detect anomalies. | ||
|
||
## 🧮 What I had done! | ||
1. Generated a synthetic dataset of transactions. | ||
2. Performed Exploratory Data Analysis (EDA) to understand the data distribution and characteristics. | ||
3. Implemented three different models: RNN, CNN, and MLP. | ||
4. Trained and evaluated each model. | ||
5. Compared the models based on their accuracy scores. | ||
|
||
## 🚀 Models Implemented | ||
- **Multi-Layer Perceptron (MLP):** A basic neural network architecture used for classification tasks. | ||
- **Recurrent Neural Network (RNN):** An LSTM-based RNN model used for sequential data analysis. | ||
- **Convolutional Neural Network (CNN):** A 1D CNN model used for pattern recognition in sequential data. | ||
|
||
## 📚 Libraries Needed | ||
- pandas | ||
- numpy | ||
- seaborn | ||
- matplotlib | ||
- scikit-learn | ||
- tensorflow | ||
|
||
## 📊 Exploratory Data Analysis Results | ||
- Basic dataset information and statistics. | ||
- Distribution of transaction amounts, transaction types, account ages, and transaction locations. | ||
- Correlation matrix to identify relationships between features. | ||
- Visualization of fraudulent vs non-fraudulent transactions. | ||
|
||
|
Oops, something went wrong.