Yazar "Güven, Ebu Yusuf" seçeneğine göre listele

Listeleniyor 1 - 4 / 4
  • Küçük Resim
    Öğe
    ICON: Instagram Profile Classification Using Image and Natural Language Processing Methods
    (Institute of Electrical and Electronics Engineers Inc., 2023) Güven, Ebu Yusuf; Boyacı, Ali; Sarıtemur, Fatma Nur; Türk, Zehra; Sütçü, Gizem; Turna, Özgür Can
    The use of social media has grown significantly, and businesses are now using these platforms to promote their products and services. To do this, companies have created business accounts on social networks. However, social media platforms can also be a breeding ground for unwanted behaviors such as cyberbullying, sexual content, and promotional comments. To address this issue, a study was conducted to create a system that could classify public accounts on Instagram by analyzing comments, profile pictures, bios, and posts shared by users with business accounts. First, a crawler was developed, and data were collected using this crawler and then anonymized. Next, the collected data were processed using natural language processing (NLP) techniques for text and image processing methods for images to extract features and create a dataset. Nearly 10 000 profiles and 30 000 comments from public accounts were manually tagged to create the classification model. The final model had an accuracy rate of 95% on the dataset, allowing for the effective identification of different types of business accounts on Instagram.
  • Küçük Resim
    Öğe
    Investigation of the effectiveness of edible oils as solvent in reactive extraction of some hydroxycarboxylic acids and modeling with multiple artificial intelligence models
    (Springer, 2023) Sevindik, Yunus Emre; Gök, Aslı; Lalikoglu, Melisa; Gülgün, Sueda; Güven, Ebu Yusuf; Gürkaş-Aydın, Zeynep; Yağcı, Mehmet Yavuz; Turna, Özgür Can; Aydın, Muhammed Ali; Aşçı, Yavuz Selim
    This study investigated the usability of diferent vegetable oils as solvents for separating citric, malic, and glycolic acids from aqueous solutions by reactive extraction method. A machine learning model was developed to predict intermediate values from the dataset created using the experimental results using multiple linear regression (MLR) and extreme gradient boosting (XGB). We used sunfower oil, corn oil, linseed oil, sweet almond oil, sesame oil, and castor oil in six types of vegetable oil. Trioctylamine (TOA) was used as an extractant in reactive extraction studies. The results obtained showed that approximately 99% of acids can be separated from their aqueous solutions when suitable mixtures of organic phases are used. Based on the results, we discovered that the XGB method outperforms the MLR method for each dataset. Thanks to the high-performance prediction model developed, it was possible to reach higher separation efciencies by determining the optimum experimental conditions. In addition, the costs and wastes associated with experiments decreased due to the developed high-performance estimation model. The reactive extraction estimation model was publicly available on GitHub and open to other researchers.
  • Küçük Resim
    Öğe
    A novel password policy focusing on altering user password selection habits: A statistical analysis on breached data
    (Elsevier, 2022) Güven, Ebu Yusuf; Boyacı, Ali; Aydın, Muhammed Ali
    Online services generally employ password-based systems to enable users to access personal/private con- tent. These services also force their users to change their passwords periodically under specific policies to increase security. However, analysis of breached data reveals that current policies do not consider user password selection habits and pose critical security and privacy concerns. Additionally, when passwords are leaked, attackers have the opportunity to study - and possibly identify - the structure or pattern of the user password selection set. This way, attackers could predict the next password or reduce the search space considerably in their attacks. Therefore, this study proposes a novel behavior-based pass- word policy to increase the present security level and avoid further exploitations if a breach occurs. This study uses statistical methods and visualization techniques to examine the password selection behaviors of over ten million UserID-password pairs collected from anonymously shared data breaches. The data set is anonymized while keeping the uniqueness of userID-password pairs and shared with other researchers along with extracted features. Results show that user password selection patterns can be generalized and used to increase the success rate of attacks.
  • Küçük Resim Yok
    Öğe
    A survey on backbone attack
    (Institute of Electrical and Electronics Engineers Inc., 2019) Güven, Ebu Yusuf; Ya?cı, Mehmet Yavuz; Boyacı, Ali; Yarkan, Serkan; Aydın, Muhammed Ali
    The Internet is the universal network infrastructure that surrounds the Earth with thousands of devices and connections that make up it. The communication of various technologies from data centers to personal smartphones is provided through this infrastructure. While end devices are renewed as technology and product in short periods, network devices such as switches and routers, where communication is provided, can work for many years and work with out-of-date software and protocols. Therefore, it is clear how important the weaknesses are. Internet communication protocols are designed with security concerns in mind instead of communication speed and bandwidth. Even though researchers work intensively on wireless networks, the security of the infrastructure that connects wireless networks is ignored. In this study, we examined the attacks on OSI layer 2 and layer 3 layers made to the devices that constitute the backbone of the Internet infrastructure. Although several security measures and updates have been published for some of these attacks, the vulnerabilities that may occur in outdated devices are revealed. © 2019 IEEE.