2004
DOI: 10.1111/j.1467-842x.2004.00325.x
|View full text |Cite
|
Sign up to set email alerts
|

Using Multinomial Mixture Models to Cluster Internet Traffic

Abstract: The paper considers the clustering of two large sets of Internet traffic data consisting of information measured from headers of transmission control protocol packets collected on a busy arc of a university network connecting with the Internet. Packets are grouped into 'flows' thought to correspond to particular movements of information between one computer and another. The clustering is based on representing the flows as each sampled from one of a finite number of multinomial distributions and seeks to identi… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0

Year Published

2006
2006
2024
2024

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 12 publications
(5 citation statements)
references
References 7 publications
0
5
0
Order By: Relevance
“…Grouping the areas into a moderate number of clusters with differing patterns of cell proportions provides a way to understand the "message" of the data, especially where the names and locations of the small regions are known to the analyst. A similar methodology was used by Jorgensen (2004) to cluster packet size distributions in packet flows over the internet between computers. That analysis was not as easy to interpret in the absence of background information about the various different packet flows.…”
Section: Resultsmentioning
confidence: 99%
“…Grouping the areas into a moderate number of clusters with differing patterns of cell proportions provides a way to understand the "message" of the data, especially where the names and locations of the small regions are known to the analyst. A similar methodology was used by Jorgensen (2004) to cluster packet size distributions in packet flows over the internet between computers. That analysis was not as easy to interpret in the absence of background information about the various different packet flows.…”
Section: Resultsmentioning
confidence: 99%
“…Elsewhere multinomial mixture modeling has been used to cluster different types of internet traffic [32]. These methods have also found use in topic modeling or text clustering [42], [46], [52], however a great deal of topic modeling focuses on mixed membership models such as Latent Dirichlet Allocation [12] or a Dirichlet-multinomial model.…”
Section: A Applications: Multinomial Mixture Modelingmentioning
confidence: 99%
“…While there is a wealth of choices for univariate non-continuous distributions, the use of multivariate non-continuous distributions for the definition of mixture models is limited due to the difficulty in constructing easy to work with models that allow practical flexibility on the dependence structure. Some successful, but limited in application examples, are finite mixtures of multivariate Poisson distributions (Karlis and Meligkotsidou, 2007), finite mixtures of multinomial distributions (Jorgensen, 2004) and models based on conditionally independent Poisson distributions (see, for example Alfo et al, 2011). Mixture models with latent structures have been considered in Browne and McNicholas (2012), but these can have limitations because of assumptions like conditional independence.…”
Section: Finite Mixture Models For Clustering Non-continuous Datamentioning
confidence: 99%