mlex-lib 0.0.1

Money Laundering Expert System - A machine learning framework for financial fraud detection

Money Laundering Expert System (MLEX)

A comprehensive machine learning framework for financial fraud detection and money laundering prevention.

Features

• Neural Network Models: GRU, LSTM, and RNN implementations optimized for sequence data
• Evaluation Framework: Comprehensive evaluation metrics and visualization tools
• Data Processing: Advanced preprocessing and feature engineering capabilities
• Model Pipeline: End-to-end machine learning pipelines for fraud detection
• Visualization: Interactive plotting and analysis tools

Installation

pip install mlex

For development installation:

pip install mlex[dev]

Quick Start

import pandas as pd
import numpy as np
from mlex.models import GRU, LSTM, RNN
from mlex.utils import DataReader, FeatureStratifiedSplit
from mlex.evaluation import StandardEvaluator, F1MaxThresholdStrategy

# Load and preprocess data
reader = DataReader('path/to/your/data.csv', target_columns=['fraud_label'])
X = reader.fit_transform()
y = reader.get_target()

# Split data
splitter = FeatureStratifiedSplit(column_to_stratify='account_id', test_proportion=0.3)
splitter.fit(X, y)
X_train, y_train, X_test, y_test = splitter.transform(X, y)

# Train model
model = GRU(
    target_column='fraud_label',
    validation_data=(X_test, y_test),
    input_size=10,
    hidden_size=64,
    epochs=50
)
model.fit(X_train, y_train)

# Evaluate
scores = model.score_samples(X_test)
evaluator = StandardEvaluator("fraud_detection", F1MaxThresholdStrategy())
evaluator.evaluate(y_test, [], scores)
print(evaluator.summary())

License

This project is licensed under the MIT License - see the LICENSE file for details.

Citation

If you use MLEX in your research, please cite:

@software{mlex2024,
  title={Money Laundering Expert System (MLEX)},
  author={Pinheiro, Diego},
  year={2024},
  url={https://github.com/IoTDataAtelier/mlex}
}

Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions