Suspected Money Laundering Detection Using Unsupervised Hidden Markov Chains
This master's thesis contains a method for detecting transactions with money laundering suspicion, using a hidden Markov model and DBSCAN (Density-based spatial clustering of applications with noise) unsupervised machine learning algorithm. The aim of the work is to replace widely used rule-based money laundering systems risk score with clusters of the DBSCAN algorithm. The clusters found are used as the money laundering risk score, which will be hidden Markov´s model observable layer.
Graduation Thesis language
Graduation Thesis type
Master - Data Science