An application of spectral clustering approach to detect communities in data modeled by graphs

Graph clustering is a popular classification technique with numerous algorithms, with a high number of published proposals, this field keeps expanding. Spectral clustering is one of graph clustering algorithms and one of the most active tools in machine learning community in general and unsupervised classification methods especially, with several applications in different fields this technique has shown its performance and its ability to deal with different data formats. In this paper we present an application of spectral clustering to detect communities in data from real world after modeling those data by graphs. We present also a comparison between the obtained results from the two most known families of spectral clustering using the unnormalized and the normalized algorithms. We finally discuss the obtained results in the output of this application and present our future works.