Yazar "Erten, Yusuf Murat" seçeneğine göre listele

Listeleniyor 1 - 4 / 4
  • Küçük Resim
    Öğe
    DGStream: High quality and efficiency stream clustering algorithm
    (Pergamon-Elsevier Science Ltd, 2020) Ahmed, Rowanda; Dalkılıç, Gökhan; Erten, Yusuf Murat
    Recently as applications produce overwhelming data streams, the need for strategies to analyze and cluster streaming data becomes an urgent and a crucial research area for knowledge discovery. The main objective and the key aim of data stream clustering is to gain insights into incoming data. Recognizing all probable patterns in this boundless data which arrives at varying speeds and structure and evolves over time, is very important in this analysis process. The existing data stream clustering strategies so far, all suffer from different limitations, like the inability to find the arbitrary shaped clusters and handling outliers in addition to requiring some parameter information for data processing. For fast, accurate, efficient and effective handling for all these challenges, we proposed DGStream, a new online-offline grid and density-based stream clustering algorithm. We conducted many experiments and evaluated the performance of DGStream over different simulated databases and for different parameter settings where a wide variety of concept drifts, novelty, evolving data, number and size of clusters and outlier detection are considered. Our algorithm is suitable for applications where the interest lies in the most recent information like stock market, or if the analysis of existing information is required as well as cases where both the old and the recent information are all equally important. The experiments, over the synthetic and real datasets, show that our proposed algorithm outperforms the other algorithms in efficiency. (C) 2019 Elsevier Ltd. All rights reserved.
  • Küçük Resim
    Öğe
    HSV color histogram based ımage retrieval with background elimination
    (Institute of Electrical and Electronics Engineers Inc., 2019) Erkut, U.; Bostancıoğlu, F.; Erten, Yusuf Murat; Özbayoğlu, A.M.; Solak, E.
    In this study, a new content based image retrieval (CBIR) method, which uses HSV histogram data is proposed. The model uses the HSV histogram to find the background from the image by analyzing the peaks in the histogram data and performing a moving window algorithm to identify the region within the histogram that belongs to the background colors. After identifying the background information, the sections of the image that are part of the background are removed from the original image and the remaining foreground or content information is stored for comparison with other images. In order to verify the methodology, a graphical user interface is developed and 1000 different images from 10 different groups from the coral database are put into the image database for comparison. The analysis and preliminary tests show that comparing only the foreground information of the images pro-vided better results than comparing images themselves, especially when searching for particular objects within the images. This algorithm can also be used as a background elimination technique to reduce the storage requirements of images and the comparison time between images can be reduced significantly. © 2019 IEEE.
  • Küçük Resim
    Öğe
    Localized identification of malicious nodes in wireless sensor networks
    (IEEE, 2020) Akram, Vahid Khalilpour; Erten, Yusuf Murat
    Securing a wireless sensor network against the various attacks with minimal energy consumption is a challenging task because the nodes usually have limited energy source, are distributed in harsh environments and communicate over standard radio channels. Adding malicious nodes to an existing wireless sensor network is one of the common types of attacks. The malicious nodes can reduce the reliability of a network by flooding heavy traffic, sending fake data or counterfeiting the paths. This study proposes a new localized algorithm for identifying the malicious nodes in WSNs where the existing verified nodes detect the malicious nodes using neighborhood information and signature of messages. The testbed experiments and simulation results show that the proposed algorithm is a feasible and efficient approach for detecting the malicious nodes.
  • Küçük Resim
    Öğe
    Loyalty program using blockchain
    (IEEE Computer Soc, 2020) Sönmeztürk, Osman; Ayav, Tolga; Erten, Yusuf Murat
    The traditional loyalty systems usually offer people benefits in a specific sector. The users usually need to stay within the loyalty system for a long time and accumulate points in order to win rewards which may not be very interesting for them most of the time. Additionally, users usually do not prefer to share their personal information to join these loyalty systems due to privacy concerns. It has, therefore, been observed that the number of customers in the loyalty systems is decreasing day by day. To reduce these drawbacks a loyalty program which complies with ERC20 standards was proposed in this study using tokens based on the Ethereum blockchain. Using this new generation loyalty system, users can convert their earned tokens to Ether in the market and they can receive services or products with the accumulated tokens according to their interests from any supplier that has been contracted by the manufacturer. Additionally, users in the designed system do not need to carry many different cards, it is adequate to have only one Ethereum wallet. Furthermore, users do not need to share any personal data to join the loyalty system. Suppliers can also request Ether from the manufacturer for the tokens they have accumulated from the members of the loyalty system. The proposed loyalty system has been implemented and presented in this study.