PORTALE DELLA DIDATTICA

Ricerca CERCA
  KEYWORD

Artificial intelligence for the estimation of web browsing QoE

keywords BIG DATA, INTERNET TRAFFIC, MACHINE LEARNING

Reference persons MARCO MELLIA, MARTINO TREVISAN

External reference persons The thesis will be held in collaboration with a large telecom operator in the United States.
There is also an ongoing collaboration with University of Torino - Prof. Idilio Drago

Research Groups SmartData@PoliTO, Telecommunication Networks Group

Thesis type EXPERIMENTAL RESEARCH

Description Understanding the Quality of Experience (QoE) enjoyed by users of web browsing and mobile apps is key to optimize services and keep users’ loyalty. Quality is subjective, and the complexity of today’s webpages and mobile apps challenges its measurement.
The goal of this thesis is the realization of automatic systems to monitor and characterize the performance of both web pages and mobile apps, using passive measurement only. The student will design a system that automatically creates a model from the traffic generated by browsers or mobile apps, and use it to analyze users' QoE. To this end, it is necessary to leverage machine learning techniques to create accurate models as well as big data approaches to make them work at scale.
The thesis will be in collaboration with international Internet Service Operators, with which the research group has been collaborating since long time.

The current work of the research group on the topic can be found in the following papers:
https://www.sciencedirect.com/science/article/abs/pii/S138912861830358X
https://ieeexplore.ieee.org/abstract/document/7840921/

Required skills Networking and network protocols (HTTP, TLS, DNS)
Machine Learning, especially supervised learning models
Big data techniques (Spark, Hadoop)


Deadline 02/12/2020      PROPONI LA TUA CANDIDATURA




© Politecnico di Torino
Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY
Contatti