دكتوراه
العلوم والتقنية
University of York
| مجال التميز | بحثي ودراسي |
|
البحوث المنشورة |
|
| البحث (1): | |
| عنوان البحث: | A Survey of Attacks Against Twitter Spam Detectors
in an Adversarial Environment |
| رابط إلى البحث: | https://www.mdpi.com/2218-6581/8/3/50 |
| تاريخ النشر: | 04/07/2019 |
| موجز عن البحث: | Online Social Networks (OSNs), such as Facebook and Twitter, have become a very important part of many people’s daily lives. Unfortunately, the high popularity of these platforms makes them very attractive to spammers. Machine learning (ML) techniques have been widely used as a tool to address many cybersecurity application problems (such as spam and malware detection). However, most of the proposed approaches do not consider the presence of adversaries that target the defense mechanism itself. Adversaries can launch sophisticated attacks to undermine deployed spam detectors either during training or the prediction (test) phase. Not considering these adversarial activities at the design stage makes OSNs’ spam detectors vulnerable to a range of adversarial attacks. Thus, this paper surveys the attacks against Twitter spam detectors in an adversarial environment, and a general taxonomy of potential adversarial attacks is presented using common frameworks from the literature. Examples of adversarial activities on Twitter that were discovered after observing Arabic trending hashtags are discussed in detail. A new type of spam tweet (adversarial spam tweet), which can be used to undermine a deployed classifier, is examined. In addition, possible countermeasures that could increase the robustness of Twitter spam detectors to such attacks are investigated |
|
المؤتمرات العلمية |
|
| المؤتمر (1): | |
| عنوان المؤتمر: | Conference for Truth and Trust Online 2019 |
| تاريخ الإنعقاد: | 24/10/2019 |
| مكان الإنعقاد: | UK |
| طبيعة المشاركة: | poster |
| عنوان المشاركة: | An Approach for Detecting Image Spam in OSNs |
| ملخص المشاركة: | In recent years, the number of images uploaded into Online Social Networks (OSNs), such as Facebook and Twitter has been growing, which presents challenges to Machine Learning-based spam detector. Most current detection models use text-based, statistic info-based and graph-based features can easily be fooled by image-based spam. These approaches do not have the ability to recognize text embedded in images. Adversaries take advantage of this issue to launch more sophisticated attacks, such as evasion attacks. Thus, this paper proposes an adversary-aware model for detecting spam images in OSNs. The proposed model adopted EAST (an Efficient and Accurate Scene Text Detector) and CRNN (Convolutional Recurrent Neural Network) models for text detection/ recognition tasks. After the text extraction step, a blacklist and white-list with Human-in-the-loop approach is applied for text classification task. Although the technique used is simple, it is adaptable and robust against adversarial text attacks. |
| المؤتمر (2): | |
| عنوان المؤتمر: | 2019 International Conference on Document Analysis and Recognition Workshops (ICDARW) |
| تاريخ الإنعقاد: | 10/10/2019 |
| مكان الإنعقاد: | Australia |
| طبيعة المشاركة: | workshop |
| عنوان المشاركة: | Detecting Spam Images with Embedded Arabic Text in Twitter |
| ملخص المشاركة: | Detecting image-based spam in Online Social Networks (OSNs), such as Facebook and Twitter, is an ongoing problem. Spam is prevalent in all forms of online communication (such as email and the web) However, researchers’ and practitioners’ attention has increasingly shifted to spam in OSNs, due to the growing number of spammers and the possible negative effects on users. There are different types of spam messages that can be found in OSNs. Spam images, which are images embedded with malicious text, are one of the most difficult types of spam to tackle. Processing images overwhelms classifiers and affects detection performance. Consequently, spammers take advantage of this issue to launch more sophisticated attacks, such as evasion attacks. After observing some Arabic trending hashtags and topics in Twitter, a substantial amount of image-based spam was found. Thus, this paper proposes an approach for detecting image-based spam with Arabic text in Twitter through using Deep Learning (DL) techniques. In this paper, an Efficient and Accurate Scene Text Detector (EAST) and Convolutional Recurrent Neural Network (CRNN) models were used for text detection and text recognition. After the text extraction step, a blacklist and whitelist approach was applied for classifying text as either spam or non-spam. The proposed text classification technique is adaptable and robust against some text classification attacks. |
