The manipulation of the Euro Interbank Offered Rate (Euribor) was an affair which had a great impact on in ternational financial markets. This study tests whether advanced data processing techniques are capable of classifying Euribor panel banks as either manipulating or non-manipulating on the basis of patterns found in quotes submissions. For this purpose, panel banks’ daily contributions have been studied and monthly variables obtained that denote different contribution patterns for Euribor panel banks. Thus, in accordance with the court verdict, banks are categorized as manipulating and non-manipulating and Machine Learning classification techniques such as Supervised Learning, Anomaly Detection and Cluster Analysis are applied in order to discriminate between convicted and acquitted banks. The results show that out of seven manipulative banks, five are detected by Machine Learning using Deep Learning algorithms, all five presenting very similar contribution patterns. This is consistent with Anomaly Detection which confirms that several manipulating banks present similar levels of abnormality in their contributions. In addition, the Cluster Analysis facilitates gathering the five most active banks in illicit actions. In conclusion, administrators and supervisors might find these techniques useful to detect potentially illicit actions by banks involved in the Euribor rate-setting process.