Improving PLDA speaker verification with limited development data

This paper analyses the probabilistic linear discriminant analysis (PLDA) speaker verification approach with limited development data. This paper investigates the use of the median as the central tendency of a speaker’s i-vector representation, and the effectiveness of weighted discriminative techniques on the performance of state-of-the-art length-normalised Gaussian PLDA (GPLDA) speaker verification systems. The analysis within shows that the median (using a median fisher discriminator (MFD)) provides a better representation of a speaker when the number of representative i-vectors available during development is reduced, and that further, usage of the pair-wise weighting approach in weighted LDA and weighted MFD provides further improvement in limited development conditions. Best performance is obtained using a weighted MFD approach, which shows over 10% improvement in EER over the baseline GPLDA system on mismatched and interview-interview conditions.

Toward understanding social engineering

There is no doubt that social engineering plays a vital role in compromising most security defenses, and in attacks on people, organizations, companies, or even governments. It is the art of deceiving and tricking people to reveal critical information or to perform an action that benefits the attacker in some way. Fraudulent and deceptive people have been using social engineering traps and tactics using information technology such as e-mails, social networks, web sites, and applications to trick victims into obeying them, accepting threats, and falling victim to various crimes and attacks such as phishing, sexual abuse, financial abuse, identity theft, impersonation, physical crime, and many other forms of attack. Although organizations, researchers, practitioners, and lawyers recognize the severe risk of social engineering-based threats, there is a severe lack of understanding and controlling of such threats. One side of the problem is perhaps the unclear concept of social engineering as well as the complexity of understand human behaviors in behaving toward, approaching, accepting, and failing to recognize threats or the deception behind them. The aim of this paper is to explain the definition of social engineering based on the related theories of the many related disciplines such as psychology, sociology, information technology, marketing, and behaviourism. We hope, by this work, to help researchers, practitioners, lawyers, and other decision makers to get a fuller picture of social engineering and, therefore, to open new directions of collaboration toward detecting and controlling it.

Demonstrating accountable-eHealth systems

The security and privacy of patient information is one of the biggest hindrances to the wide adoption of eHealth systems. For eHealth systems to be successful they must provide protection for patients’ privacy while ensuring healthcare professionals are able to access the information necessary to provide appropriate care. Accountable-eHealth systems are a proposed solution to these potentially competing concerns by enforcing appropriate use and after-the-fact accountability measures. We have developed a Web-based prototype to demonstrate scenarios of how both appropriate and inappropriate use of patient information would be handled in an Accountable-eHealth system.

Flexible IBE and beyond in the commutative-blinding framework

The cryptographic community has, of late, shown much inventiveness in the creation of powerful new IBE-like primitives that go beyond the basic IBE notion and extend it in many new directions. Virtually all of these “super-IBE” schemes rely on bilinear pairings for their implementation, which they tend to use in a surprisingly small number of different ways: three of them as of this writing. What is interesting is that, among the three main frameworks that we know of so far, one has acted as a veritable magnet for the construction of many of these “generalized IBE” primitives, whereas the other two have not been nearly as fruitful in that respect. This refers to the Commutative Blinding framework defined by the Boneh-Boyen [Bscr ][Bscr ]1 IBE scheme from 2004. The aim of this chapter is to try to shed some light on this approach's popularity, first by comparing its key properties with those of the competing frameworks, and then by providing a number of examples that illustrate how those properties have been used.