In recent years the data companies collect from their business have increased exponentially and new frontiers of information extraction has started the so called Big Data techniques. In fact the real value that companies can obtain from data measured not only upon the quantity of collected data but also upon the process they implement to exploit them.
In Big Diamond project, data are provided by Fastweb S.p.A., a major Italian Internet Service Provider, and contain information about users Internet navigations in an anonymous form to preserve users’ privacy.
The first part of the project consisted in an analysis of Big Data related business model, identifying the driver for creating value, the future trends, the type of business in which Fastweb S.p.A. should operate and a strategic analysis. The web usage mining field turned out to be the best market Fastweb S.p.A. could enter in order to keep a competitive advantage.
After having inspected the raw data, a pre-processing has been performed in order to generate and visualize a graph which represents the paths users follow from one website to another. This abstraction model, referred to as a Webgraph, allows us to perform useful analyses.
To better understand and inspect the Webgraph created, a dynamic visualization has been developed. In particular, a force-directed layout is at the base of the visualization in order to avoid and reduce overlapping effects, which may create a visual clutter phenomenon.
To prove the potentialities of this analysis, a case study on a web usage dataset from a Bologna sample in April 2015 has been conducted: statistics on incoming and outgoing traffic have been compared among different websites, in order to identify common trends and differences. This information has been analysed, allowing us to create a description of both users’ habits and websites’ marketing strategies with particular reference to their competitors’ market positions.
Principal Academic Tutor
Marco Mellia, Department of Electronics and Telecommunications, Politecnico di Torino
Elena Baralis, Department of Electronics and Telecommunications, Politecnico di Torino
Chiara Francalanci, Department of Electronics Information and Bioengineering, Politecnico di Milano
Mauro Gasparini, Department of Mathematical Sciences, Politecnico di Torino
Paolo Giacomazzi, Department of Electronics Information and Bioengineering, Politecnico di Milano
Paolo Neirotti, Department of Management and Production Engineering, Politecnico di Torino
Francesco Vaccarino, Department of Mathematical Sciences, Politecnico di Torino
Fabrizio Callari, Electrical Engineering, Politecnico di Torino [Team Controller and Communication Coordinator]
Sebastiano Castello, Automation Engineering, Politecnico di Milano
Filippo Garolla, Computer Engineering, Politecnico di Milano
Dario Passi, Design & Engineering, Politecnico di Milano
Milica Pavlovic, Interior Design , Politecnico di Milano
Enrico Scopelliti, Computer Engineering, Politecnico di Torino