Yazar "Emekci, Fatih" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    Dividing secrets to secure data outsourcing
    (Elsevier Science Inc, 2014) Emekci, Fatih; Methwally, Ahmed; Agrawal, Divyakant; El Abbadi, Amr
    Data outsourcing or database as a service is a new paradigm for data management. The third party service provider hosts databases as a service. These parties provide efficient and cheap data management by obviating the need to purchase expensive hardware and software, deal with software upgrades and hire professionals for administrative and maintenance tasks. However, due to recent governmental legislations, competition among companies and database thefts, companies cannot use database service providers directly. They need secure and privacy preserving data management techniques to be able to use them in practice. Since data is remotely stored in a privacy preserving manner, there are efficiency related problems such as poor query response time. We propose a new framework that provides efficient and scalable query response times by reducing the computation and communication costs. Furthermore, the proposed technique uses several service providers to guarantee the availability of the services while detecting the dishonest or faulty service providers without introducing additional overhead on the query response time. The evaluations demonstrate that our data outsourcing framework is scalable and practical. (C) 2013 Elsevier Inc. All rights reserved.
  • Küçük Resim Yok
    Öğe
    Name spell-check framework for social networks
    (Tubitak Scientific & Technological Research Council Turkey, 2016) Yildiz, Burak; Emekci, Fatih
    The problem of finding similar strings is very important in most real life applications including spell-checking, data cleaning, next generation sequencing, and alignment. In order to query and manage string data online, scalable algorithms and frameworks are essential. Scalable frameworks and algorithms have been introduced in the past few years. However, these frameworks mainly deal with caching and querying structured data. They do not deal with fuzzy queries, where we need to search for an approximate string. In this paper, we propose an edit distance aware filtering algorithm for all kinds of approximate string search problems. We also propose a novel name spell-check engine mainly for social networks. Our experiments show that our edit distance aware filtering mechanism alone improves the query processing time and throughput by almost 30%. Additionally, our name spell-check engine improved the name spell-check response time and throughput almost 10 times by using our filtering scheme and some domain specific observations.
  • Küçük Resim Yok
    Öğe
    Privacy Preserving Secure Online Advertising
    (Elsevier Science Bv, 2015) Oger, Mustafa; Olmez, Isa; Inci, Erinc; Kucukbay, Serkan; Emekci, Fatih
    Online advertisement has generated 48.09 billion dollars in 2013 (IAB, 2014) in US sponsoring most of the free services and content on the Internet today. However, problems stemming from the rivalry to get a share of this high revenue can disrupt the system jeopardizing online privacy and security. There have been a lot of malicious cases where fraudsters have deployed methods such as malware or phishing to get a share from the market thus compromising the security of end users and thus harming the end users. In addition to this, tracking methods deployed to increase the targeting accuracy of the advertisements reveals a lot of private information about the users without any permission from them. The rules around collection and usage of personal data is ambiguous and can be harmful in various ways such when it is shared with third parties. In response to these threats, end users have been inclining to use ad-avoidance tools which disable user tracking or block ad requests completely. This situation harms the Internet ecosystem which heavily relies on online advertising. To remedy this problem, Comodo is developing secure and privacy-friendly advertising solutions around its trusted ads concept using methods such as anonymous user tracking, safe ad content delivery and continuous ad inspection (malware/virus scanning). These methods guarantee secure advertisements delivery to end users, while preserving effectiveness of ad delivery. The beta versions of the products have been deployed and tested in real life for 6 months. Based on our observations, users are concerned about their privacy; but they also want customized content according to their needs and make money from ads as long as it is not harmful to them. Based on this data, we are willing to expand our system and make it one of the major ad-delivery networks. (C) 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).