What's my App?

With the spread of broadband Internet, Real-Time Communication (RTC) platforms have become increasingly popular and transformed way people communicate. Thus, it is fundamental that network adopts traffic management policies ensure appropriate Quality Experience to users RTC applications. A key step for this identification applications behind traffic, which in turn allows allocate adequate resources make decisions based on specific application's requirements. In paper, we introduce a machine learning-based system identifying It builds domains contacted before starting call leverages techniques from Natural Language Processing (NLP) build meaningful features. Our works real-time robust peculiarities RTP implementations different applications, since uses only control traffic. Experimental results show our approach classifies 5 well-known meeting with an F1 score 0.89.