Yangın ve Güvenlik Dergisi 181. Sayı (Mart 2016)

YANGIN ve GÜVENL ø K SAYI 181 85 GÜVENL ø K - MAKALE edilebilir seviyede performans oldu ù u bulunmu ü tur. Ancak, modern bilgisayar sisteminin karma ü × k davran × ü ü ekilleri ta- mamen anla ü × lamam × ü t × r. Xie ve ba ü kala- r × [21] en yak × n K-kom ü uyu kullanm × ü , bura- da k, frekansta boyutland × rmay × azaltmak üzere ADFA-LD baz × nda kümele ü meyi temsil ederken, optimum mesafe fonksi- yonu tan × mlanmaktad × r. 5.ADFA-LD K × stas Veri Setinin K × s × tla- malar × ve ú yile ü tirme ú çin Öneriler ADFA-LD siber güvenlik veri setinin tüm ni- teliklerini anlayabilmek biraz zor olmaktad × r zira di ù er ara ü t × rmac × lar taraf × ndan da ko- layca anla ü × labilir bir dilde aç × klama yap × l- mam × ü t × r. Veri taraf × ndan sürülen tecavüz alg × lama sistemlerinin tasar × m × nda veri set- lerinin girdi ve ç × kt × lar × n × n aç × k bir biçimde tan × mlanmalar × çok kritiktir. Kendi kendine ö ù renme ve veri irdeleme camias × n × AD- FA-LD veri seti kullanmak üzere çekmek için, veri seti üreticisi için girdi ve ç × kt × nite- liklerinin çok net bir ü ekilde tan × mlanmas × gereklidir. Veri setinin sütun ve kiri ü leri aç × k bir biçimde tan × mlanmam × ü t × r. ADFA-LD niteliklerinin de, UCI Kendi Kendine Ö ù ren- me Kayna ù × k × stas × veri setine [23] benzer bir biçimde tan × mlanmas × yap × lmal × d × r. Bu takdirde, ara ü t × rma camias × ndan benzeri görülmemi ü bir ilgi olu ü acakt × r. Veri haz × r- lama ve mühendisli ù i, very irdeleme sü- recinin %80’ini olu ü turmaktad × r [24]. Veri setleri tam olarak anla ü × lamad × ù × takdirde, ADFA-LD siber güvenlik veri seti kullan × larak veri taraf × ndan sürülen tecavüz alg × lama sistemleri çok da anlaml × olmayacakt × r. Mevcut bilgisayar teknolojisine uyumlu ol- mas × na ra ù men, önerilen tecavüz alg × la- ma sistemlerinin de ù erlendirilmesinde veri setlerinin potansiyel kullan × c × lar × n × n ü evkini k × rabilecektir. UCI Kendi Kendine Ö ù renme Kayna ù × gibi, ADFA-LD veri setleri sütun ve kiri ü leri de net bir biçimde dosya formatla- r × nda tasarlanmal × d × r. Referanslar 1. Joycee, KAM, Parkavi R, Senthikumari R. Network Intrusion Detection & Pre- vention. Automat Auton Syst 2014; 5:244-245. 2. Chiroma H, Abdulhamid SM., Gital YA, Usman AM, Maigari TU. Academ- ic community cyber cafes - A Per- petration point for cyber crimes in Nigeria. Int J Inf Sci Comp Eng 2011; 2:7-13. 3. Abdulhamid SM, Haruna C, Abubakar A. Cybercrimes and the Nigerian Ac- ademic Institution Networks. IUP J Inf Tech 2011; 7: 47-57. 4. Longe OB, Chiemeke SC. Cyber- crime and criminality in nigeria -what roles are internet access points in playing?. Eur J of Social Sci 2008; 6:132-139. 5. Metasploit Penetration Testing Soft- ware, http://www.metasploit.com, Accessed November 24, 2014. 6. Creech G, Hu J. Generation of a new IDS test dataset: Time to retire the KDD collection. In: 2013 IEEE Confer- ence on Wireless Communications and Networking (WCNC), 2013; p. 4487-4492. 7. Hu J. Host-based anomaly intrusion detection, In: Handbook of Informa- tion and Communication Security, P. Stavroulakis and M. Stamp, editors. Springer Berlin Heidelberg, 2010; p. 235-255. 8. Murtaza SS, Khreich W, Hamou-Lhadj A Couture M. A Host-based Anomaly Detection Approach by Represent- ing System Calls as States of Kernel Modules. In: 24th International Sym- posium on Software Reliability Engi- neering, Pasadena, 2013, p. 431- 440. 9. Lin YD, Lai YC, Ho CY, Tai WH. Cred- itability-based weighted voting for re- ducing false positives and negatives in intrusion detection. Comput Se 2013; 39:460-474. 10. Sinclair C, Pierce L, Matzner. An ap- plication of machine learning to net- work intrusion detection,” In: Comput- er Security Applications Conference, 1999.(ACSAC’99) Proceedings. 15 th Annual, 1999; p. 371-377. 11. Chiroma H Abdul-Kareem S, Abuba- kar A. A Framework for Selecting the Optimal Technique Suitable for Appli- cation in a Data Mining Task. In: Park J.J et al. Future Information Technol- ogy, 2014; 276: p.163-169. 12. Julisch K. Data mining for intrusion detection. In: Barbara D, Jajodia S, editors. Applications of data mining in computer security, 2002; p. 33-62. 13. Moradi M, Zulkernine M. A neural net- work based system for intrusion detec- tion and classification of attacks. In: Pro- ceedings of the 2004 IEEE international conference on advances in intelligent system-theory and applications, 2004; p. 142-152. 14. Kim DS, Nguyen HN, Park J. Genetic algorithm to improve SVM based net- work intrusion detection system, AINA 2005. In: 19th International Confer- ence on Advanced Information Net- working and Applications, 2005; 2. p.155-158. 15. Helali RGM. Data mining based net- work intrusion detection system: A survey, Novel Algorithms and Tech- niques in Telecommunications and Networking. pp.501-505. J Inf Tech 2010; 7:47-57. 16. Internet crime complaint center (IC3). The 2013 Internet Crime report . [Online]. Available: http://www.ic 3. gov/default.aspx. Accessed Novem- ber 21, 2014. 17. McHugh J. Testing Intrusion Detec- tion System: a critique of the 1998 and 1999 DARPA Intrusion Detection System evaluations as performed by Lincoln Laboratory. ACM T Inf Syst Se 2000; 4: 262-294. 18. Computer Science Department, “University of New Mexico Intrusion Detection Dataset. [Online]. Avail- able: http://www.cs.unm.edu/_im- msec/systemcalls.htm 19. Brown C, Cowperthwaite A Hijazi A SoMayaji A. Analysis of the 1999 DARP A/Lincoln Laboratory IDS eva- luation data with NetADHICT. In: IEEE Symposium on Computational Intelli- gence for Security and Defense App- lications, 2009. p. 1-7. 20. School of Engineering and Informati- on Technology, UNSW, Australia. ADFA Linux data set (ADFA-LD) cyber se- curity benchmark dataset. [Online]. Available: http://www.cybersecurity . unsw.adfa.edu.au/ADFA%20IDS%20 Datasets/