You will investigate social structures through the use of networks and graph theory.
The goal will be to describe and analyse the network, its parameters and clusters in order to find the optimal algorithm for identifying the leading nodes (influences). This studies and methods could be used in later stages in Risk assessment, Fraud investigation and also Viral Marketing areas.
The Research Problem
Inputs: Telenor will provide aggregate data of profile and users behaviour.
The dataset is in csv format and contains 1,5M rows of data.
The attributes of the data are the connections between A and B nodes (not unique) and the connection strength represented in 2 different variables.
Output: Telenor will expect characterized network structures in terms of nodes and the ties or links that connect them. Examples of social structures commonly visualized through social network analysis (sociograms) include friendship and acquaintance networks.
Other important goal, not only for the case giver but for the society, will be to verify the algorithms of cluster and leading nodes selections.
Due to the sensitive nature of the data, more details of the case will be available only to the participants that have chosen to work on it.