Amaretto: An Active Learning Framework for Money Laundering Detection

Abstract

Monitoring financial transactions is a critical Anti-Money Laundering (AML) obligation for financial institutions. In recent years, machine learning-based transaction monitoring systems have successfully complemented traditional rule-based systems to reduce the high number of false positives and the effort needed to review all the alerts manually. Unfortunately, machine learning-based solutions also have disadvantages: while unsupervised models can detect novel fraudulent patterns, they are usually characterized by a high number of false alarms; supervised models, instead, usually offers a higher detection rate but require a large amount of labeled data to achieve such performance. In this paper, we present Amaretto, an active learning framework for money laundering detection that combines unsupervised and supervised learning techniques to support the transaction monitoring processes by improving the detection performance and reducing the fraud management costs. Amaretto exploits novel selection strategies to target a subset of transactions for investigation, making more efficient use of the feedback provided by the analyst. We perform the experimental evaluation on a synthetic dataset provided by the industrial partner, which simulates the profiles of clients trading in international capital markets. We show that Amaretto outperforms state-of-the-art solutions by reducing money laundering risk and improving detection performance. In particular, we compare state-of-the-art unsupervised and supervised techniques commonly used in the AML domain with the ones implemented in this work. We show that the Isolation and Random Forests of Amaretto perform best in the task under analysis, with an AUROC of 0.9 for the first (20% better on average) and a detection rate of 0.793 for the second (30% better on average). In addition, both methods are characterized by a lower costs computed in terms of the daily number of transactions to be examined and the number of false positives and negatives. Finally, we compare Amaretto against a state-of-the-art active learning fraud detection system, achieving better detection performances and lower costs in all the scenarios under analysis. Worth mentioning, Amaretto improves the detection rate up to 50% and reduces the overall cost by 20% in the most realistic scenario under analysis.

A Novel Money Laundering Prediction Model Based on a Dynamic Graph Convolutional Neural Network and Long Short-Term Memory

“…Luo et al [27] demonstrated the effectiveness of using a GRU to extract deep temporal features from information flows, contributing to the exploration of dynamic temporal dataset features. Labanca D et al [28] proposed using a GRU to automatically extract implicit temporal features from bank transaction data and built a fusion model with features extracted using self-attention mechanisms as an alternative to traditional rulebased methods. Additionally, several works have proposed using transformer models [39], which were originally designed for processing language sequences and have achieved good results.…”

“…They address the vanishing and exploding gradient issues faced by traditional RNNs by introducing gate mechanisms that capture long-term dependencies. Furthermore, a GCN [28] is a deep learning model for processing graph data that captures the relationships between nodes by performing convolution operations on the input graph structure. GCN-GRU and MGC-LSTM [33] are fusion models that execute graph convolution operations on a static graph constructed from the entire training set in combination with GRUs and LSTM networks, respectively.…”

“…Due to their excellent sequence modeling capabilities, recurrent neural networks (RNNs) [22] can effectively capture the temporal dependencies in input data. To overcome the issues faced by traditional RNNs when handling long sequences, improved structures such as long short-term memory (LSTM) [23][24][25] and gated recurrent units (GRUs) [26][27][28] have been proposed, achieving better results in tasks involving time series data.…”

Fraud Detection and Anti-Money Laundering Applying Machine Learning Techniques in Cryptocurrency Transactional Graphs

“…Effective AML systems are necessary for organizations to reduce the risk of financial losses from money laundering. AML involves identifying suspicious activities through customer transaction analysis using software, databases, and analytical tools (Labanca et al, 2022). Cryptocurrencies and blockchain technology present challenges in implementing traditional AML practices due to their decentralized and transparent nature.…”

Anomaly Detection in Cryptocurrency Transactions With Active Learning

“…Training high-performance supervised classifiers for FD is crucial, but the lack of labeled data makes it challenging to train such ML models (Labanca et al, 2022). The process of labeling an entire dataset requires a vast sample of manually reviewed transactions, which is often unpractical due to the high costs involved and the limited investigative time and budget available for companies (Barata et al, 2021).…”