Abstract
Recently, Internet of things (IoT) has become one of the hot topics of research. Things are, in most cases, deployed in unmonitored fields. So, ensuring the reliability of collected data is becoming a challenging issue. One of the most common problem affecting IoT is caused by intrusion, which could alter, delete or modify data collected by things. It could affect the whole functionality of IoT by causing faulty situations like taking wrong decisions. On one hand, the intrusion is among the hardest attack that could to be detected. On the other, Artificial Intelligence (AI) tools are powerful and emerging techniques that could be used to achieve this purpose. In this paper, we propose to use classification techniques to deal with the problem of intrusion detection. More precisely, we applied a set of classification tools on a real IoT dataset to detect intrusion. The comparison of classification results is shown through an experimental study.