Computer Science and Information Technologies, Computer Science and Information Technologies 2016

Font Size: 
Generic Profiling
A. V. Gorshechnikova

Last modified: 2020-12-20


The article proposes an analytics tool implementation created to cluster and profile users given a set of URLs they visit using their mobile phones. This analysis is carried out in an aggregated and anonymous way by the regions related to the origin of mobile traffic. We aim at employing a machine learning approach which relies on a comprehensive set of features derived from the content of the users navigations. As initial data we take charging data records of users activity in the mobile internet provided by Telecom Italia company (TIM).


Generic Profiling; cluster; mobile phones; machine learning approach


1. M.Y. Kan and H. O. N. Thi, "Fast webpageclassification using URL features" in Proc. of the 14thACM international conference on Information andknowledge management (CIKM '2005). New York,NY, USA: ACM, 2005, pp. 325-326;

2. E. Baykan , M. Henzinger, L. Marian and I. Weber,"Purely URL-based Topic Classification", PosterSessions: 2009

3. X. Qi and B. D. Davison, "Web page classification:Features and Algorithms," ACM Comput. Surv., vol.41, no. 2, pp. 12:1-12:31, 2009.

4. Leskovec J, Rajaraman,A., “Mining of MassiveDatasets”, 2011

5. R. Rajalakshmi, "Supervised term weighting methodsfor url classification", Journal of Computer Science10 (10): 1969-1976, 2014, Department of ComputerScience and Engineering, SSN College ofEngineering, Chennai, India

6. Pavel Berkhin, “Survey of Clustering Data MiningTechniques”, Accrue Software, Inc.

Full Text: PDF