The multi-label customer reviews classification task aims to identify the different thoughts of customers about the product they are purchasing. Due to the impact of the COVID-19 pandemic, customers have become more prone to shopping online. As a consequence, the amount of text data on e-commerce is continuously increasing, which enables new studies to be carried out and important findings to be obtained with more detailed analysis. Nowadays, e-commerce customer reviews are analyzed by both researchers and sector experts, and are subject to many sentiment analysis studies. Herein, an analysis of customer reviews is carried out in order to obtain more in-depth thoughts about the product, rather than engaging in emotion-based analysis. Initially, we form a new customer reviews dataset made up of reviews by Turkish consumers in order to perform the proposed analysis. The created dataset contains more than 50,000 reviews in three different categories, and each review has multiple labels according to the comments made by the customers. Later, we applied machine learning methods employed for multi-label classification to the dataset. Finally, we compared and analyzed the results we obtained using a diverse set of statistical metrics. As a result of our experimental studies, we found the Micro Precision 0.9157, Micro Recall 0.8837, Micro F1 Score 0.8925, and Hamming Loss 0.0278 to be the most successful approaches.

Unmanned Aerial Vehicles (UAVs) are air vehicles that can be controlled by themselves or via a ground station, can operate with mission-route definition, and can stay in the air in a limited range and time. Today, UAVs are used for various missions in line with military and civilian purposes such as reconnaissance, observation, research, search, control and transportation. In the models used for military purposes, the gun can be mounted and turned into an armed UAV. Attacks on hardware and software systems of UAVs, which are increasingly used in line with developing life conditions and technologies, are also increasing. Especially attacks on communication systems and coordinate information come to the fore. Cyber attacks on UAV systems in general include Data transmission link attacks, GPS (Global Positioning System) Scam attack, authentication attacks such as Brute Force, attacks against source code vulnerabilities, and hardware port and protocol attacks. It is known that attacks on UAV systems cause loss of life and property. It also results in other damages such as task delay, data breach, task failure, and loss of prestige. In this study, after the components, features and introduction of UAV systems, security threats to UAV systems and cyber attacks are defined. Then, the precautions to be taken against threats and attacks are listed. Some of these, importance are subjecting UAV software, hardware and transmission systems to vulnerability and risk scanning, ensuring the security of transmission with data encryption methods, controlling data traffic with a firewall application and making access control secure.

Recent technological advancements, including internet-based distance education and artificial intelligence-supported learning analytics, have significantly impacted the field of education. These advancements not only enhance the efficiency of education but also broaden access to learning while mitigating barriers to implementation. AI-supported learning analytics emerges as a pivotal tool for interpreting data gleaned from educational processes and stakeholders, thereby enhancing educational processes and outcomes. This tool streamlines the measurement, analysis, and evaluation of learning processes, encompassing a wide array of factors and parameters. Moreover, it contributes to the development of personalized and adaptive learning environments. In this study, a predictive model utilizing the XGBoost algorithm has been developed to analyze students' academic achievements. The model forecasts final exam grades based on various student characteristics, including age, participation rate, and exam scores. Evaluating the performance of the AI model involves metrics such as Mean Squared Error, Mean Absolute Error, and R² score. In findings indicate a strong prediction performance, with an R² score of 0.819. As a result of underscore the potential of AI-supported learning analytics as an effective tool for predicting and enhancing student academic performance.

Some measures can be taken by analyzing network traffic on small and medium-sized computer networks and using open source intrusion detection systems (IDS) on embedded microprocessor hardware instead of high-capacity firewalls in the attack detection phase. At this stage, Raspberry Pi and open source IDSs are the first to come to mind because of the cheap and easy installation. In this research, a scenario is created on a local network to listen to the network traffic and Snort and Suricata are installed on two different Raspberry Pi, to compare the performance of the IDSs. For this purpose, hping attacks have been made by a user and IDSs have been trying to obtain the performance of catching packets.

In small and home office (SOHO) computer networks, an intrusion detection system can be activated to intercept the possible attack. Network traffic analysis is the basis of attack detection methods. In the intrusion detection process, both hardware and software solutions are available, as well as in systems that are bundled together. Hosting these systems in SOHO networks is usually negligible because it is considered a costly and laborious investment. In order to overcome this disadvantage open source code Intrusion Detection Systems (IDS) such as snort and strata can be used together with raspberry pi computers. In this study, a raspberry pi device was prepared on a small scale network and a network traffic analysis application was performed with the snort IDS module installed on it. It has been determined that some of the malicious traffic on the network by activated developing system.

Yaşamın giderek dijitalleştiği bilgi çağında, bilginin değeri ve önemi her geçen gün artmaktadır. Bilginin değerini ve önemini artırmanın esas yolu; onun gizliliği, güvenliği ve bütünlüğü unsurlarını korumaktan geçmektedir. Bu unsurların tümü pek çok bileşenin ve faktörün bir arada olduğu, sürekli ve çok yönlü etkileşimde bulundukları bir sistemi oluşturmaktadır. Bu sistemin yapısı gereği pek çok riski de beraberinde getirmektedir. Bu risklerin önceden belirlenmesi, hesaplanması ve analiz edilmesinde iyi bir risk yönetim anlayışına ihtiyaç vardır. Bilişim teknolojileri bu yönetim anlayışına destek olmak için pek çok yeni yöntem ve teknik ortaya koymaktadır. Yapay zekâ yöntem ve teknikleri buna örnek olarak verilebilir. Bilgi güvenliğini sağlama aşamasında ortaya çıkan açıkların, eksiklerin ve risklerin yapay zekâ ile önceden belirlenerek önlemlerin alınmasında önemli roller üstlendiği görülmektedir. Özellikle veri toplama, işleme ve karar verme süreçlerini kapsayan veri analitiği ile tahmin etme ve karar vermeyi kolaylaştırmaktadır. Ayrıca, veri iletimi ve erişimi sırasında oluşan anormal durumların tespitinde makine öğrenimi ve doğal dil işleme algoritmaları önemli başarılar elde etmektedir. Bu çalışma, kurumsal bilgi güvenliği yönetiminde yapay zekâ destekli risk analizine değinmektedir. Bu kapsamda kullanılan yapay zekâ uygulamalarının özellikleri açıklanırken, yapay zekâ destekli örnek bir risk analizi modelini oluşturulmuştur. Bu model içerisinde yapay zekâ yöntem, teknik ve araçları belirtilmiştir. Ayrıca, yapay zekânın, bilgi güvenliği alanında risk analizi süreçlerine entegrasyonu ve potansiyel faydaları üzerinde durulmaktadır. Çalışmada, önerilen modelde yer alan yapay zekâ tekniklerinin ve modellerinin risk analizi aşamalarına uygunluğu araştırılırken uygulama temelli ilişkilendirmeler yapılmıştır. Çalışmanın diğer bir amacı ise, geleneksel risk analizi yöntemlerine kıyasla yeni nesil bilişim teknolojileri desteğiyle oluşturulan yöntemlere yönelik farkındalığın artırılmasıdır.