Analytical methods for customer relationship management (CRM) have gained increasing importance in today's businesses. Some industry sectors such as the telecommunication industry accumulate huge amounts of data not only about the usage behaviour of individual customers, but also about how customers interact. In addition to traditional data mining and statistical techniques, methods from the field of social network analysis (SNA) are essential to leverage this special set of data. For example, call detail records of telephone operators can be used to evaluate the network of customers and derive measures for the influence of persons in such a network. This information is relevant to viral marketing, as well as various other forms of advertising and campaign management. Research in network analysis has led to a number of different centrality measures, which are potentially useful statistics for such purposes. In this paper, we compare different centrality measures based on a variety of different network topologies and model assumptions
[1]
Lada A. Adamic,et al.
Friends and neighbors on the Web
,
2003,
Soc. Networks.
[2]
N. Lin.
Foundations of social research
,
1976
.
[3]
Stanley Wasserman,et al.
Social Network Analysis: Methods and Applications
,
1994,
Structural analysis in the social sciences.
[4]
Albert-László Barabási,et al.
Statistical mechanics of complex networks
,
2001,
ArXiv.
[5]
Lada A. Adamic,et al.
Information flow in social groups
,
2003,
cond-mat/0305305.
[6]
Leonard M. Freeman,et al.
A set of measures of centrality based upon betweenness
,
1977
.
[7]
U. Brandes.
A faster algorithm for betweenness centrality
,
2001
.
[8]
P. Lazarsfeld,et al.
Friendship as Social process: a substantive and methodological analysis
,
1964
.
[9]
M. A. Beauchamp.
AN IMPROVED INDEX OF CENTRALITY.
,
1965,
Behavioral science.
[10]
Gert Sabidussi,et al.
The centrality index of a graph
,
1966
.