Researcher:
BAYDOĞMUŞ, GÖZDE KARATAŞ

Profile Picture
ORCID

Job Title

Dr. Öğr. Üyesi

First Name

GÖZDE KARATAŞ

Last Name

BAYDOĞMUŞ

Name

Name Variants

Email Address

Birth Date

Filters

2014 - 2019132020 - 20224

Settings

Researcher Of Publications: search.filters.isAuthorOfPublication.4e820274-4a42-44ba-aced-ca58912c0424

Search Results

Now showing 1 - 10 of 17
  • Thumbnail Image
    PublicationEmbargo
    Mobil Cihazlarda Güvenlik Tehditler Temel Stratejiler
    (Istanbul Commerce University, 2016-09-30) Zaim, Abdül Halim; AKBULUT, AKHAN; BAYDOĞMUŞ, GÖZDE KARATAŞ; 110942; 116056
    The most common form of today’s consumer electronics is the use of mobile devices. The technological developments in this area are increasing in such a way as to affect all aspects of our lives and give direction to our life. The use of mobile devices, which can now have the same hardware features as computers, is not only for communication but also enriched by applications in the areas of internet use, business, hobby and health. Increased usage rate and the need for information and communication security are beginning to be needed and it is necessary to ensure the security of the information carried against the attacks against these devices. Due to security vulnerabilities in mobile devices and malicious software loaded applications by end users, there are situations that threaten personal information and communication security. This study describes security vulnerabilities, attacks in mobile applications and precautions for those problems. It summarizes not only the end-user's recommendations, but also the points to note for app developers. It is evaluated that end users can increase personal security by learning basic info for attack methods of mobile systems.
  • Thumbnail Image
    PublicationOpen Access
    A Systematic Mapping Study on Software Architecture Recovery
    (2016-11) Çatal, Çağatay; BAYDOĞMUŞ, GÖZDE KARATAŞ; ; 108363; 110942
    In this study, we investigated the approaches used in software architecture recovery papers, identify the current status of paper distributions in terms of year, publication channel, electronic databases, and journals. We executed a mapping study to cluster the software architecture recovery research papers. Papers published since 2000 have been used for this study. The following databases were investigated: Wiley, IEEE, ACM, Science Direct. Our search accessed 250 papers, but after in-depth analysis, 60 papers were found to be related to the software architecture recovery area. Our study shows that there exist many architecture recovery approaches in the literature, with machine learning-based techniques dominating the field. On the basis of this study, we suggest researchers develop more model centric software architecture recovery approaches because of model driven development’s popularity in software engineering field.
  • Placeholder
    PublicationMetadata only
    Intrusion Detection Systems with GPU-Accelerated Deep Neural Networks and Effect of the Depth
    (2018) Reis, Buminhan; Kaya, Semi Berk; BAYDOĞMUŞ, GÖZDE KARATAŞ; ŞAHİNGÖZ, ÖZGÜR KORAY; 110942; 214903
    With the extended use of the Internet, which connects millions of computers across the world, there is a growing number and types of intrusions which complicate ensuring the security of information and computers. Although Firewalls and rule/signature base Intrusion Detection Systems (IDSs) are used as the first line of the defense of networks, they cannot be sufficient for detecting the zero-day type attacks, which are not previously encountered. For this type of attacks, Anomaly-Based Intrusion detection systems arise as an acceptable solution which models the normal communication behavior of the network and identifies the others as a suspicious transaction. To classify the normal behavior, usage of neural networks and machine learning approaches are accepted as powerful solutions. However, due to the lack of computation power, generally single hidden layer approach is preferred. With the enhancement of the parallel computation technology, especially in Graphics Processing Units (GPUs), it will be easy to implement a multi-layer approach in Deep Neural Network concept which has a great deal of attention within Deep Learning approach. Therefore, better accuracy rate could be reached. In this paper, we aimed to implement a Deep Neural Network-based Intrusion Detection System. Moreover, we also study the performance of the proposed model in binary classification with a different number of layers, neurons and parameters. Additionally, the acceleration of the GPU usage is also measured and presented with a comparison. To measure the performance of the proposed system the NSL-KDD data set, which is a 'cleaned' data set of the KDD data set, is preferred. The experimental results showed that the proposed multi-layer Deep Neural Network model produces an acceptable performance in its classification with a high accuracy rate with the design of a 64x32 hidden layer structure depending on the data set NSL-KDD.
  • Thumbnail Image
    PublicationOpen Access
    Increasing the Performance of Machine Learning-Based IDSs on an Imbalanced and Up-to-Date Dataset
    (IEEE, 2020) BAYDOĞMUŞ, GÖZDE KARATAŞ; Demir, Önder; ŞAHİNGÖZ, ÖZGÜR KORAY
    In recent years, due to the extensive use of the Internet, the number of networked computers has been increasing in our daily lives. Weaknesses of the servers enable hackers to intrude on computers by using not only known but also new attack-types, which are more sophisticated and harder to detect. To protect the computers from them, Intrusion Detection System (IDS), which is trained with some machine learning techniques by using a pre-collected dataset, is one of the most preferred protection mechanisms. The used datasets were collected during a limited period in some specific networks and generally don & x2019;t contain up-to-date data. Additionally, they are imbalanced and cannot hold sufficient data for all types of attacks. These imbalanced and outdated datasets decrease the efficiency of current IDSs, especially for rarely encountered attack types. In this paper, we propose six machine-learning-based IDSs by using K Nearest Neighbor, Random Forest, Gradient Boosting, Adaboost, Decision Tree, and Linear Discriminant Analysis algorithms. To implement a more realistic IDS, an up-to-date security dataset, CSE-CIC-IDS2018, is used instead of older and mostly worked datasets. The selected dataset is also imbalanced. Therefore, to increase the efficiency of the system depending on attack types and to decrease missed intrusions and false alarms, the imbalance ratio is reduced by using a synthetic data generation model called Synthetic Minority Oversampling TEchnique (SMOTE). Data generation is performed for minor classes, and their numbers are increased to the average data size via this technique. Experimental results demonstrated that the proposed approach considerably increases the detection rate for rarely encountered intrusions.
  • Placeholder
    PublicationMetadata only
    Neural network based intrusion detection systems with different training functions
    (IEEE, 345 E 47th St, New York, Ny 10017 USA, 2018) BAYDOĞMUŞ, GÖZDE KARATAŞ; ŞAHİNGÖZ, ÖZGÜR KORAY; 110942; 214903
    In the last decades, due to the improvements in networking techniques and the increased use of the Internet, the digital communications entered all of the activities in the global marketplace. Parallel to these enhancements the attempts of hackers for intruding the networks are also increased. They tried to make unauthorized access to the networks for making some modifications in their data or to increase the network traffic for making a denial of service attack. Although a firewall seems as a good tool for preventing this type of attacks, intrusion detection systems (IDSs) are also preferred especially for detecting the attack within the network system. In the last few years, the performance of the IDS is increased with the help of machine learning algorithms whose effects depend on the used training/learning algorithm. Mainly it is really hard to know which learning algorithm can be the fastest one according to the problem type. The algorithm selection depends on lots of factors such as the size of data sets, number of nodes network design, the targeted error rate, the complexity of the problem, etc. In this paper, it is aimed to compare different network training function in a multi-layered artificial neural network which is designed for constructing an effective intrusion detection system. The experimental results are depicted in the paper by explaining the efficiency of the algorithms according to their true-positive detection rates and speed of the execution.
  • Placeholder
    PublicationMetadata only
    Genetic algorithm for intrusion detection system
    (IEEE, 345 E 47th St, New York, Ny 10017 USA, 2016) BAYDOĞMUŞ, GÖZDE KARATAŞ; 110942
    Intrusion detection systems are systems that prevent or slow the serving of different types of server data traffic caused by intensive use of networks that spread by internet. Particularly in recent years, because of increase in data density, need for these systems is increased thus different detection algorithms are being developed. In this study, information about the different detection algorithms using genetic algorithm, which are made of IDS algorithms is given and literature search's been made.
  • Placeholder
    PublicationMetadata only
    Multi-tenant architectures in the cloud: a systematic mapping study
    (IEEE, 345 E 47Th St, New York, Ny 10017 USA, 2017) Can, Ferit; Doğan, Gamze; Konca, Cemile; AKBULUT, AKHAN; BAYDOĞMUŞ, GÖZDE KARATAŞ; 110942; 116056
    Unlike traditional web applications, cloud-based applications can provide services for large number of tenants using the same hardware and software by implementing multi-tenant architectures. 407 papers have been examined by using systematic mapping method to evaluate the publications related to this architecture, which have been used increasingly in the Software-as-a-Service (SaaS) model. The goal of the study is to determine which storage strategies were used most, which criteria were taken into account in selecting the preferred storage strategy, and the most searched topics under multi-tenant architecture model. Primary researches which are conforming to specified review rules have been obtained from electronic databases (IEEE, ACM, Springer, ScienceDirect, Wiley, Scopus) and classified by research topic and content.
  • Thumbnail Image
    PublicationOpen Access
    Dağıtık veritabanları için test uygulaması
    (İstanbul Kültür Üniversitesi / Fen Bilimleri Enstitüsü / Bilgisayar Mühendisliği Anabilim Dalı, 2015-06) BAYDOĞMUŞ, GÖZDE KARATAŞ; Akbulut, Akhan
    İnternetin yaygınlaşması ile birlikte büyük veri kullanımında İlişkisel Veritabanı Yönetim Sistemleri (İVTYS), ölçeklenebilirlik konusunda yetersiz kalmaya başlamıştır. Son 10 yılın yükselmeye başlayan yeni veri saklama teknolojisi Dağıtık Veritabanları İVTYS‟lerin sunamadığı hizmetleri sağlamaktadır. Bu çalışmanın amacı, yazılım mühendisliği kapsamında önemli bir konu olan yazılım testlerinin, kullanımı artan bu sistemlere uygulanmasıdır. Bir dağıtık veritabanı uygulaması olan MongoDB üzerinde farklı sınamaların gerçeklendiği test uygulaması geliĢtirilmiş olup, bu uygulamaya "MongoDB Tester" ismi verilmiştir. Bu uygulama kullanılarak Dağıtık Veritabanları üzerinde yapılan iyileştirmelere ait sonuçlar paylaşılmıştır. MongoDB Tester uygulamasının önerisi ile yapılan iyileştirmelerle elde edilen sonuçlarda sistemde iyileştirmeler olduğu görülmüştür.
  • Placeholder
    PublicationMetadata only
    Deep Learning in Intrusion Detection Systems
    (2018) Demir, Önder; BAYDOĞMUŞ, GÖZDE KARATAŞ; ŞAHİNGÖZ, ÖZGÜR KORAY; 110942; 170651; 214903
  • Thumbnail Image
    PublicationEmbargo
    Automated testing for distributed databases with fuzzy fragment reallocation
    (TUBİTAK Scientific & Technical Research Council Turkey, Ataturk Bulvarı No 221, Kavaklıdere, Ankara, 00000, Turkey, 2018) AKBULUT, AKHAN; BAYDOĞMUŞ, GÖZDE KARATAŞ; 116056; 110942
    As the Internet and big data become more widespread, relational database management systems (RDBMSs) become increasingly inadequate for web applications. To provide what RDBMSs cannot, during the past 15 years distributed database systems (DDBSs) have thus emerged. However, given the complicated structure of these systems, the methods used to validate the efficiency of databases, all towards ensuring quality and security, have become diversified. In response, this paper demonstrates a system for performing automated testing with DDBSs, given that testing is significant in software verification and that accredited systems are more productive in business environments. The proposed system applies several tests to MongoDB-based NoSQL databases to evaluate their instantaneous conditions, such as average query response times and fragment utilizations, and, if necessary, suggest improvements for indexes and fragment deployment. Within this context, autogenerated data, replica, meta, system, fragment, and index tests are applied. Clearly, the system's most important feature is its fuzzy logic-enabled fragment reallocation module, which allows the creation and application of reallocation strategies that account for data changes in query frequency.