Wyniki wyszukiwania - BazTech

1

Systemy wykrywania Intruzów wykorzystujące metody sztucznej Inteligencji

Kukiełka R., Kotulski Z.

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2011

|

nr 4

114-121

PL

W ostatnich latach sztuczna inteligencja znajduje zastosowanie w wielu dziedzinach. Jedną z nich są również systemy wykrywania intruzów IDS (Intrusion Detection System). Dzięki zdolności generalizacji metody sztucznej inteligencji umożliwiają klasyfikację ataków nie tylko według nauczonych wzorców, ale również wszelkich ataków podobnych do nich oraz niektórych nowych typów. IDS stosujące takie metody mogą się również w sposób dynamiczny dostosowywać do zmieniającej się sytuacji w sieci (np. uczyć się nowych zachowań użytkowników lub nowych ataków). Ich zaletą jest to, że nie wymagają budowy skomplikowanych zbiorów reguł i sygnatur odrębnych dla każdej instancji ataków, ponieważ dane niezbędne do wykrycia ataku są uzyskiwane automatycznie w procesie nauki. Artykuł zawiera podstawowe pojęcia związane z systemami wykrywania włamań oraz przegląd wyników dotyczących zastosowania w IDS metod sztucznej inteligencji, takich jak: drzewa decyzyjne, algorytmy genetyczne, systemy immunologiczne, sieci Bayesa oraz sieci neuronowe.

EN

Last years one of the most extensively studied field of research is artificial intelligence. It is used in many practical applications, one of them is Intrusion Detection Systems (IDS). Thanks to their generalization feature artificial intelligence methods allow to classify not only the learned attacks patterns but also their modified versions and some new attacks. They could dynamically adapt to changing situation in the network (eg., learn new users' behaviors or new attacks). Ań advantage of application of the artificial intelligence methods in IDS is that they do not require generation of the rule or the signature for each new instance of an attack because they automatically update the IDS knowledge in the learning phase. The first part of this paper includes basic information about intrusion detection system. In the next sections we present application in IDS such artificial intelligence methods like: decision trees, genetic algorithms, immunology systems, Bayes networks, and neural networks.

2

Security of information in university e-Learning systems

Grzybowski R.

Journal of Applied Computer Science

|

2011

|

Vol. 19, nr 2

7-18

EN

The increase in popularity of university eLearning platforms, supporting traditional methods of education, as well as new communication tools development, make the security issues of information stored in these systems becoming increasingly important. The general discussion on security issues in e-learning was inaugurated by M. K. Littman [1] already in 1996. This document has been prepared to identify data security threats existing in a very specific group of computer systems - university eLearning platforms. As a result of the security incidents database analysis and user behavior described in the literature, thirteen security areas have been defined. Activities undertaken in particular areas, described during the data analysis, have a significant impact on at least one of three basic elements of information security.

3

Nowy, bezpieczny Internet

Rawiński T.

Nowa Elektrotechnika

|

2005

|

nr 2 (6)

2--3

4

Intrusion detection systems: the fundamentals, problems and the way forward

Sodiya A. S., Longe H. O. D.

Journal of Applied Computer Science

|

2004

|

Vol. 12, nr 1

83-95

EN

Intrusion detection systems become very important computer security mechanisms as computer break-ins are getting more common everyday. Intrusion detection system (IDS) monitors computers and networks for any set of actions that attempt to compromise the integrity, confidentiality or availability of computer resources. The goal of this paper is to discuss the fundamentals of IDS and to create awareness on why IDS should be embraced. A users study was carried out to understand the perceptions of individuals: organisations and companies on the use of IDS. Summary of problems of en/rent IDS designs and the challenges ahead are presented. We also look tit what should be the new approaches or future directions m IDS design so as to eliminate these shortcomings.

5

A new combined strategy to intrusion detection

Sodiya A. S., Longe H. O. D.

Journal of Applied Computer Science

|

2004

|

Vol. 12, nr 1

97-116

EN

Researchers haw used so many techniques in designing intrusion detection systems (IDS) and we still do not have an effective one. The interest in this work is to combine techniques of data mining and expert system in designing an effective anomaly-based IDS. We believe that combining methods may give better coverage, and wake the detection more effective. The Idea is to mine system audit data for consistent and useful patterns of users behaviour, and then keep these normal behaviours in profiles. We then use expert system as our detection system that recognizes anomalies, and raises an alarm. The evaluation of the intrusion detection system designed allows that detection efficiency and false alarm rate problems of IDS are greatly unproved upon.

6

SMSFC - Secure Mailing System for Corporation

Grzybowski R., Cebula P.

Journal of Applied Computer Science

|

2003

|

Vol. 11, nr 1

39-46

EN

The objective of this article is to present a novel system for a secure mail exchange within a corporation. The applied solution is general enough to be implemented using a standard POP3 mail exchange protocol. Despite the fact that POP3 protocol is considered to be unsafe, in our approach no mail message in the corporate network is ever left in the plain format. It is encrypted at the moment of its transmission and while it is stored in user mailboxes. Hence, it is useless for an intruder who could possibly intercept encrypted data. The system under consideration been devised and realized at the Institute of Computer Science at the Technical University of Lodz, Poland.

7

Internet payments developing prognosis

Pietruszka M., Wojciechowski A.

Journal of Applied Computer Science

|

2000

|

Vol. 8, nr 1

63-76

EN

In this article it will be presented several possibilities of carrying out secure transactions via Internet. In the beginning, this paper describes historical conditions, methods used at present and visualisation of still developing solutions. It also treats about X.509 certificates and SET protocol - probably the best present solution. At the end of the article it is possible to accustom with software available at the market and firms dealing with this problem.