Nikolaos Alexandris

Neural network based person identification using EEG features

A direct connection between the electroencephalogram (EEG) and the genetic information of an individual has been suspected and investigated by neurophysiologists and psychiatrists since 1960. However, most of this early as well as more recent research focuses on the classification of pathological EEG cases, aiming to construct tests for purposes of diagnosis. On the contrary, our work focuses on healthy individuals and aims to establish an one-to-one correspondence between the genetic information of the individual and certain features of his/her EEG, as an intermediate step towards the further goal of developing a test for person identification based on features extracted from the EEG. Potential applications include, among others, information encoding and decoding and access to secure information. At the present stage the proposed method uses spectral information extracted from the EEG non-parametrically via the FFT and employs a neural network (a learning vector quantizer-LVQ) to classify unknown EEGs as belonging to one of a finite number of individuals. Correct classification scores ranging from 80% to 100% in experiments conducted on real data, show evidence that the EEG indeed carries genetic information and that the proposed method can be used to construct person identification tests based on EEG features.

Comparing the Trust and Security Models of Mobile Agents

Achieving privacy preservation in a data-sharing computing environment is a challenging problem. The requirements for a privacy preserving data access policy should be formally specified in order to be able to establish consistency between the privacy policy and its purported implementation in practice. Previous work has shown that when specifying a privacy policy, the notion of purpose should be used as the basis for access control. A privacy policy should ensure that data can only be used for its intended purpose, and the access purpose should be compliant with the datas intended purpose. This paper presents a mechanism to specify privacy policy using VDM. The entities in the purpose-based access control model are specified, the invariants corresponding to the privacy requirements in privacy policy are specified, and the operations in the model and their proof obligations are defined and investigated.

A Classroom Observation Model Fitted to Stochastic and Probabilistic Decision Systems

This paper focuses on solving the problems of preparing and normalizing data that are captured from a classroom observation, and are linked with significant relevant properties. We adapt these data using a Bayesian model that creates normalization conditions to a well fitted artificial neural network. We separate the method in two stages: first implementing the data variable in a functional multi-factorial normalization analysis using a normalizing constant and then using constructed vectors containing normalization values in the learning and testing stages of the selected learning vector quantifier neural network.

A scenario for the development and use of teaching oriented ontologies

In this paper we propose a methodology for the construction and use of Teaching-Oriented Ontologies and present how such ontologies can support the lesson planning phase and the teaching process. We present their pilot use with focus on the Greek secondary-education system. We then present our arguments supporting their use, suggest a methodology for the development of such ontologies and explain how teaching-oriented ontologies can aid the teaching process. Finally, we demonstrate an example of the use of our methodology and discuss the findings of an empirical evaluation of its use by teachers during the lesson planning phase.

New Malicious Agents and SK Virii

In this paper possible trends in virus creation are described that might appear in the close future. Specialy, we show that a computer virus can be updated, just like Scipt-Kiddies do. The aim of this paper is mainly to state the problem of the evolution of computer virii, and the need to take precautional measures, if possible.

A New SETUP for Factoring Based Algorithms

When using cryptography, there is frequently the misconception that if the keys are properly stored and are not given away, then the user is safe, if he uses a strong encryption algorithm. SETUPs exploit this common misconception, be embedding secret trapdoors in the key creation procedure of public key algorithms. The essential part of a SETUP, is to provide good statistical properties for the used parameters, while maintaining the security of the encryption algorithm. In this work, a new method for creating SETUPs for factoring based encryption algorithms is presented. The SETUP takes advantage of Coppersmiths theorem in order to efficiently recover the needed parameters.

Notice of Retraction Towards a formalistic measuring of security using stochastic calculus

In todays digital era, one major problem for a corporation is securing its Information Systems (IS). So, now more than ever, we realize how significant it is, to both know as well as measure, the security risks of a corporation. The aim of this study is to try and zoom in, into the corporate risks taken when deciding for IS security and provide a new model for measuring security. It will attempt to be both, qualitative by providing a series of references to the world wide bibliography having dealt with security risks and quantitative by using Stochastic Processes. At the end, we will be able to provide a more accurate and tangible method for measuring security risks as well as, the surrounding factors that are involved while trying to achieve security. Furthermore, through stochastic processes, we will study the impact of the “time factor” in conjunction with the variation of IS security level and suggest a methodology for measuring security using stochastic calculus.

Enhancing 2D environments with 3D data input

In this paper we study how we can enhance the two dimensional environment that a user has, using a three dimensional input device. We mainly focus on the Wiimote device developed for Nintendo Wii console, but the results can be generalized for other 3D input devices.

Histographic Steganographic System

In this paper we propose a new steganographic algorithm for jpeg images named HSS, with very good statistical properties. The algorithm is based on a previous work of Avidan and Shamir for image resizing. One of the key features of the algorithm is its capability of hiding the message according to the cover image properties, making the hidden message as traceable as possible.

Fault Tolerant Production Infrastructures in Practice

This paper applies the theory of transformation of existing High Availability Standby Systems to fully Fault Tolerant Production Infrastructures in order to increase productivity, effectiveness and availability. Bringing together the concepts of change management and the experience from IT industry, a case study of a financial institution is presented and analysed, illustrating the increased capabilities and practical limitations of a Fault Tolerant Production Infrastructure.