Hierarchical Clustering for Network Traffic Analysis,

Designing for Reliability and Maintainability in Service-Oriented Architectures, and

Topological Expression of Fingerprint Minutiae for Template Protection

Dr. Abdun Mahmood, Dr. Mikhail Perepletchikov and Dr. Fengling Han

Date and time: 11.30am - 13.00pm, Friday 6 November, 2009

Venue: 10.08.03 (Building 10, Level 8, Room 3)

Abstracts:

Hierarchical Clustering for Network Traffic Analysis (Dr. Abdun Mahmood)

An important problem in Network Management is how to understand the pattern of usage of a network by its users. One problem faced when analyzing network traffic data traces is how to use the mix of different types of attributes present in the data in order to better understand the underlying traffic patterns. In general, data mining techniques can be made more efficient by exploiting the underlying structure of the data. In this talk, we briefly explore how to use hierarchical clustering techniques for efficient representation, analysis, and reporting of network traffic data.

 

Designing for Reliability and Maintainability in Service-Oriented Architectures (Dr. Mikhail Perepletchikov)

Dr Mikhail Perepletchikov will provide a brief overview of the ARC Discovery Grant "Designing for Reliability and Maintainability in  Service-Oriented Architectures". Additionally, Mikhail will present a suite of empirically-evaluated software metrics designed to predict the maintainability of Service-Oriented software in the early stages of the SDLC. Such metrics can be incorporated into a service-oriented software design methodology in order to allow quantitative comparison and selection of alternative design structures, as well as detection of potential quality problems.

 

Topological Expression of Fingerprint Minutiae for Template Protection (Dr Fengling Han)

Fingerprint biometrics has long been used in identity identification. Minutiae representation of fingerprints is utilized by forensic experts, and has been adopted by most of the commercially available automatic fingerprint matching systems. Fingerprint minutiae are unique but unreliable, and prints of the same finger are rarely identical, which are called intra-user variation. The contribution of this research is describing the information of minutiae > with an equivalent vector to account for intra-user variations. The relative relation of minutiae is used as a feature, which is rotation invariant and translation invariant. The orientation information can help to overcome the elastic distortion. To improve the uniqueness of the vector feature, a dual layer structure check is deployed. In this way, the biometric template protection and cancellable biometrics can be achieved.

 


Seminar Organisation

Seminars are free and open to the general public. No booking is necessary. If you are interested in giving a presentation in this seminar series, or to make suggestions for speakers, please contact Xiaodong Li, the seminar co-ordinator.