APT Eğitimi Sunumu Alper BaşaranAPT (Advanced Persistent Threat - Gelişmiş Devamlı Tehdit) saldırıları konusunda düzenlediğim seminere ait sunumun bir kısmıdır. Sunum içerisinde yer alan konu başlıkları buradadır, sunumun tamamını dağıtımını kontrol edebilmek ve içerisindeki hassas bilgilerin korunması amacıyla paylaşmıyorum.
Eşle/İndirge Yöntemi Kullanılarak Destek Vektör Makinesi Algoritması ile Yü...Ferhat Ozgur CatakEşle/İndirge Yöntemi Kullanılarak Destek Vektör Makinesi Algoritması ile Yüksek Boyutlu Sosyal Medya Mesajlarının Kutupsal Değerinin Ölçülmesi
Rassal Bölümlenmiş Veri Üzerinde Aşırı Öğrenme Makinesi ve Topluluk Algoritma...Ferhat Ozgur CatakRassal Bölümlenmiş Veri Üzerinde Aşırı Öğrenme Makinesi ve Topluluk Algoritmaları ile Sınıflandırma
Robust Ensemble Classifier Combination Based on Noise Removal with One-Class SVMFerhat Ozgur CatakThe document describes a proposed approach for robust ensemble classifier combination based on noise removal with one-class SVM. The approach partitions an input dataset into sub-datasets, applies noise removal to each sub-dataset using one-class SVM, creates local classifier ensembles for each sub-dataset, and combines the ensemble classifiers using weighted voting. It aims to improve classification accuracy by reducing noise and training ensemble classifiers on partitions of the data. The document outlines the basic idea, discusses preliminaries like one-class SVM and AdaBoost, and describes experiments to evaluate the proposed approach.
MapReduce based SVMFerhat Ozgur CatakThis document discusses training a support vector machine (SVM) classifier in a cloud computing system using MapReduce. It introduces SVMs and issues with training them, describes MapReduce as a cloud computing algorithm, develops a CloudSVM system model using MapReduce functions, presents simulation results on UCI datasets, and concludes with recommendations.
Fuzzy Analytic Hierarchy Based DBMS Selection In Turkish National Identity Ca...Ferhat Ozgur CatakDatabase Management Systems (DBMS) play an important role to support
enterprise application developments. Selection of the right DBMS is a crucial decision for
software engineering process. This selection requires optimizing a number of criteria.
Evaluation and selection of DBMS among several candidates tend to be very complex. It
requires both quantitative and qualitative issues. Wrong selection of DBMS will have a
negative effect on the development of enterprise application. It can turn out to be costly and adversely affect business process. The following study focuses on the evaluation of a multi criteria
decision problem by the usage of fuzzy logic. We will demonstrate the methodological considerations
regarding to group decision and fuzziness based on the DBMS selection problem. We developed a new
Fuzzy AHP based decision model which is formulated and proposed to select a DBMS easily. In this
decision model, first, main criteria and their sub criteria are determined for the evaluation. Then these
criteria are weighted by pair-wise comparison, and then DBMS alternatives are evaluated by assigning a
rating scale.
Eşle/İndirge Yöntemi Kullanılarak Destek Vektör Makinesi Algoritması ile Yü...Ferhat Ozgur CatakEşle/İndirge Yöntemi Kullanılarak Destek Vektör Makinesi Algoritması ile Yüksek Boyutlu Sosyal Medya Mesajlarının Kutupsal Değerinin Ölçülmesi
Rassal Bölümlenmiş Veri Üzerinde Aşırı Öğrenme Makinesi ve Topluluk Algoritma...Ferhat Ozgur CatakRassal Bölümlenmiş Veri Üzerinde Aşırı Öğrenme Makinesi ve Topluluk Algoritmaları ile Sınıflandırma
Robust Ensemble Classifier Combination Based on Noise Removal with One-Class SVMFerhat Ozgur CatakThe document describes a proposed approach for robust ensemble classifier combination based on noise removal with one-class SVM. The approach partitions an input dataset into sub-datasets, applies noise removal to each sub-dataset using one-class SVM, creates local classifier ensembles for each sub-dataset, and combines the ensemble classifiers using weighted voting. It aims to improve classification accuracy by reducing noise and training ensemble classifiers on partitions of the data. The document outlines the basic idea, discusses preliminaries like one-class SVM and AdaBoost, and describes experiments to evaluate the proposed approach.
MapReduce based SVMFerhat Ozgur CatakThis document discusses training a support vector machine (SVM) classifier in a cloud computing system using MapReduce. It introduces SVMs and issues with training them, describes MapReduce as a cloud computing algorithm, develops a CloudSVM system model using MapReduce functions, presents simulation results on UCI datasets, and concludes with recommendations.
Fuzzy Analytic Hierarchy Based DBMS Selection In Turkish National Identity Ca...Ferhat Ozgur CatakDatabase Management Systems (DBMS) play an important role to support
enterprise application developments. Selection of the right DBMS is a crucial decision for
software engineering process. This selection requires optimizing a number of criteria.
Evaluation and selection of DBMS among several candidates tend to be very complex. It
requires both quantitative and qualitative issues. Wrong selection of DBMS will have a
negative effect on the development of enterprise application. It can turn out to be costly and adversely affect business process. The following study focuses on the evaluation of a multi criteria
decision problem by the usage of fuzzy logic. We will demonstrate the methodological considerations
regarding to group decision and fuzziness based on the DBMS selection problem. We developed a new
Fuzzy AHP based decision model which is formulated and proposed to select a DBMS easily. In this
decision model, first, main criteria and their sub criteria are determined for the evaluation. Then these
criteria are weighted by pair-wise comparison, and then DBMS alternatives are evaluated by assigning a
rating scale.
9. • Hedef
– Sunucu/Bilgisayar
– Ağ bileşenleri
– Uygulamalar
– Web Siteleri
• Yaklaşım
– Band Genişliği
• Kurallara uygun olmayan
yüksek trafik isteği
– Bağlantı
• CPU/Memory Kaynaklarını
tüketme
• Yüksek bağlantı isteği
• Sonuç
– Hizmet dışı kalma
Dağıtık Servis Dışı Bırakma Saldırıları (DDoS)
9
18. • Zararlı ağ trafiğinin analiz edilmesi
– Model değerlendirme. Precision & Recal & F1
– PCAP: Ağ trafiğinin kayıt edilmesi
– Nitelik Çıkarımı
– Çatak, Ferhat Özgür. "Topluluk Yöntemlerine Dayalı Dağıtık Hizmet Dışı Bırakma
Saldırılarının Algılanması." Süleyman Demirel Üniversitesi Fen Bilimleri Enstitüsü
Dergisi (2017): DOI-10.
Zararlı Ağ Trafiği – Topluluk Yöntemleri
18