Alessandra Sala: Abstract

Network operators are looking into novel economic models to face the increasing demand in connectivity and network bandwidth raised by the adaption of technologies such as smart phones and social networking websites. Customer¹s data are targeted as precious source of information that should shed light into novel and better services for costumers that would also result in increasing profits for operators. Unfortunately, the available data to researchers, marketers and analysts are often partial and the experimental analysis curried on these partial data may conceal important features upon with operators may build new revenue streams.

This talk will discuss the hidden effects in the evaluation of marketing campaign in social networks when the spread of information is estimated from a partial view of the network. In particular, this presentation will discuss a methodology to estimate the systematic error found in the outcome of information diffusion models when run on partial networks.

Organised by BCS SGAI The Specialist Group on Artificial Intelligence http://www.bcs-sgai.org