Our work focuses on the detection of anomalies in traces of the communication infrastruc- ture of the Information System (IS) of the SNCF. Two recent and independent techniques seem particularly appropriate in our case. The first is the storage and indexation of time series in a tree called iSAX tree, and the second is an anomaly detection score named CFOF, score that has been proven to resist to the concentration phenomenon in high dimension. In this article, we show that it is possible to use the structuration of information in the iSAX tree to quickly determine an approximation of the CFOF score. We show that the approximated score is close to the exact score on synthetic and real datasets, and the first feedbacks indicate that this score seems relevant for triggering alarms related to the anomalies detected in the activity traces of SNCF IS.