Pattern Recognition and Applications Group

PRA Group works on the development of next generation pattern recognition systems for real applications such as biometric authentication, text categorization, and intrusion detection in computer networks.

PRA mission is to address fundamental issues for the development of future pattern recognition systems, in the context of real applications. 

 

Highlights 

 

Biometric Technologies for Computer Security Course. Registrations are now open.

A brief introduction to biometric technologies for personal recognition. Main algorithms for feature extraction, matching score computation are shown in this seminar. Read more...

New accepted paper: Security evaluation of pattern classifiers under attack

Sometimes our pattern classifiers have to be deployed in hostile environments, where intelligent and malicious adversaries (spammers, hackers) put them under attack. What about the design of classifiers that can resist under attack?

A new PRA paper, coming soon on IEEE Transactions on Knowledge and Data Engineering, tells you how to design this kind of pattern classifier.  Read the preprint copy. 

Seminar by prof. Riccardo Manzotti: "Machine consciouness: theoretical and empirical issues"

On April 16 (4.30 P.M.) Prof. Riccardo Manzotti held a seminar titled Machine consciouness: theoretical and empirical issues.

The aim of the seminar is to discuss the state of the art on artificial consciousness by understanding the nature of consciousness. Venue: Room 4B - Department of Pedagogy, Psychology, and Philosofy - University of Cagliari.

New PRA Lab publication on Information Sciences: Adversarial attacks against intrusion detection systems: Taxonomy, solutions and open issues.

The PRA Lab is happy to announce a new journal publication that uncover critical vulnerabilities of Intrusion Detection Systems (IDSs). Dr. Igino Corona, Prof. Giorgio Giacinto and Prof. Fabio Roli highlight, through both practical examples and thorough analysis of state-of-the-art research work, how skilled attackers can defeat IDSs, leveraging on intrinsic vulnerabilities of their design. They also critically investigate possible countermeasures against IDS attacks, and point out future, interesting research paths in the field. Read more about this publication.