Graduate School and Research Center in Digital Sciences

Xiaolan SHA

Xiaolan SHA
Xiaolan SHA
Eurecom - Digital Security 
Phd Student ( 2009 - 2013)


Design, analysis and implementation of distributed recommendation engines



The objective of this Thesis is to design, analyze and implement distributed recommendation engines for highly volatile environments and applications, in which end-users connect to the Internet using a combination of wired and wireless access networks.
The goal of this work is to advance the state‐of‐the‐art within cooperative information filtering and recommendation systems, both at fundamental theoretical and/or algorithm design levels. The theoretical research will use emerging ideas from spectral clustering and principal component analysis and apply them to a distributed setting. The algorithmic research will use ideas akin to economic modeling to design novel utility functions describing the core mechanisms of a recommender system.
Furthermore, a fundamental objective of this Thesis is to develop new measurements techniques to perform user-profiling in a lightweight and non-intrusive manner, and to design new performance metrics to assess the quality of recommendations.
The ultimate goal of this Thesis will be a real-life implementation and experimentation of the devised system in the context of a "Mobile Advertisement" application, in cooperation with PlayAdz, a company affiliated with EURECOM, who will be funding the Thesis according to the CIFRE scholarship program.