Majed A. AlRubaian

A Performance Evaluation of Machine Learning-Based Streaming Spam Tweets Detection

The popularity of Twitter attracts more and more spammers. Spammers send unwanted tweets to Twitter users to promote websites or services, which are harmful to normal users. In order to stop spammers, researchers have proposed a number of mechanisms. The focus of recent works is on the application of machine learning techniques into Twitter spam detection. However, tweets are retrieved in a streaming way, and Twitter provides the Streaming API for developers and researchers to access public tweets in real time. There lacks a performance evaluation of existing machine learning-based streaming spam detection methods. In this paper, we bridged the gap by carrying out a performance evaluation, which was from three different aspects of data, feature, and model. A big ground-truth of over 600 million public tweets was created by using a commercial URL-based security tool. For real-time spam detection, we further extracted 12 lightweight features for tweet representation. Spam detection was then transformed to a binary classification problem in the feature space and can be solved by conventional machine learning algorithms. We evaluated the impact of different factors to the spam detection performance, which included spam to nonspam ratio, feature discretization, training data size, data sampling, time-related data, and machine learning algorithms. The results show the streaming spam tweet detection is still a big challenge and a robust detection technique should take into account the three aspects of data, feature, and model.

Facial Expression Recognition Utilizing Local Direction-Based Robust Features and Deep Belief Network

Emotional health plays very vital role to improve people’s quality of lives, especially for the elderly. Negative emotional states can lead to social or mental health problems. To cope with emotional health problems caused by negative emotions in daily life, we propose efficient facial expression recognition system to contribute in emotional healthcare system. Thus, facial expressions play a key role in our daily communications, and recent years have witnessed a great amount of research works for reliable facial expressions recognition (FER) systems. Therefore, facial expression evaluation or analysis from video information is very challenging and its accuracy depends on the extraction of robust features. In this paper, a unique feature extraction method is presented to extract distinguished features from the human face. For person independent expression recognition, depth video data is used as input to the system where in each frame, pixel intensities are distributed based on the distances to the camera. A novel robust feature extraction process is applied in this work which is named as local directional position pattern (LDPP). In LDPP, after extracting local directional strengths for each pixel such as applied in typical local directional pattern (LDP), top directional strength positions are considered in binary along with their strength sign bits. Considering top directional strength positions with strength signs in LDPP can differentiate edge pixels with bright as well as dark regions on their opposite sides by generating different patterns whereas typical LDP only considers directions representing the top strengths irrespective of their signs as well as position orders (i.e., directions with top strengths represent 1 and rest of them 0), which can generate the same patterns in this regard sometimes. Hence, LDP fails to distinguish edge pixels with opposite bright and dark regions in some cases which can be overcome by LDPP. Moreover, the LDPP capabilities are extended through principal component analysis (PCA) and generalized discriminant analysis (GDA) for better face characteristic illustration in expression. The proposed features are finally applied with deep belief network (DBN) for expression training and recognition.

Defending unknown attacks on cyber-physical systems by semi-supervised approach and available unlabeled data

Abstract Cyber-physical systems (CPS) are used increasingly in modern industrial systems. These systems currently encounter a significant threat of malicious activities created by malicious software intent on exploiting the fact that the software of such industrial systems is integrated with hardware and network systems. Malicious codes dynamically and continuously change their internal structure and attack patterns using obfuscation techniques, such as polymorphism and metamorphism, in order to bypass and hide from conventional malware detection engines. This requires continuously updating the database of the malware detection engine, which requires periodic effort from manual experts. This could limit the real-time protection of CPS. In addition, this also makes preserving the availability and integrity of the services provided by CPS against malicious code challenging because there is a demand for the development of specialized malware detection techniques for CPS. In this paper, we propose a semi-supervised approach that automatically integrates the knowledge about unknown malware from already available and cheap unlabeled data into the detection system. The novelty of the proposed approach is that it does not require expert effort to update the database of the detection engine. Instead, the dynamic changes in malware attack patterns are extracted by unsupervised clustering from already available unlabeled data. Then the extracted geometric information about the intrinsic attack characteristics of the clusters is integrated into the classification systems of the detection engine, which updates the detection system automatically. The proposed approach uses global K-means clustering with term-frequency (TF), inverse document frequency (IDF), and cosine similarity as a distance measure for extracting the cluster information and adding it to a support vector machine (SVM) classification system. The proposed approach has been tested extensively on a real malware data set for both static and dynamic malware features. The experiment results show that the proposed semi-supervised approach achieves higher accuracy over the existing supervised approaches for all classifiers. We note that the static feature-based semi-supervised approach can improve detection accuracy significantly. While applying the proposed semi-supervised approach with the run-time characteristics of dynamic feature analysis, the combined effect of dynamic analysis and the proposed approach further increases the detection accuracy of all classifiers by up to a 100% for the SVM and the random forest classifiers, thus exceeding the existing supervised approaches with similar features.

Sybil Defense Techniques in Online Social Networks: A Survey

The problem of malicious activities in online social networks, such as Sybil attacks and malevolent use of fake identities, can severely affect the social activities in which users engage while online. For example, this problem can affect content publishing, creation of friendships, messaging, profile browsing, and commenting. Moreover, fake identities are often created to disseminate spam, use the private information of other users, commit fraud, and so on. A malicious person can generate numerous fake accounts for these purposes to reach a large number of trustworthy users. Thus, these types of malicious accounts must be detected and deactivated as quickly as possible. However, this objective is challenging, because a fake account can exhibit trustworthy behaviors and have a type of name that will prevent it from being detected by the security system. In this paper, we provide a comprehensive survey of literature from 2006 to 2016 on Sybil attacks in online social networks and use of social networks as a tool to analyze and prevent these attack types. We first review existing Sybil attack definitions, including those in the context of online social networks. We then discuss a new taxonomy of Sybil attack defense schemes and methodologies. Finally, we compare the literature and identify areas for further research in Sybil attacks in online social networks.

Quality of service aware cloud resource provisioning for social multimedia services and applications

The increasing number of next-generation multimedia services and social media applications in cloud computing put additional challenges in efficient resource provisioning that targets to minimize under or over utilization of resources as well as to increase user satisfaction. Most of the works in the literature focused either on resource estimation and scheduling approaches or energy consumption for executing social media data processing applications. However, they do not consider energy consumption cost for communication devices and network appliances and schedule Virtual Machines (VMs) based on centralized job placement approach. In this paper, we develop a Quality of Service (QoS) aware cloud resource management system that decreases energy consumption and increases resource utilization by diverse multimedia social applications. In order to minimize the VM creation time we allow recycling of VM resources for user request with similar resource requirements. We have developed two distributed and localized resource management algorithms based on energy conservation, and requirements and availability of resources. The results of simulation experiments depict that the resource scheduling system greatly reduces the amount of energy consumption while maintaining the QoS of social multimedia applications.

Maximizing quality of experience through context-aware mobile application scheduling in cloudlet infrastructure

Application software execution requests, from mobile devices to cloud service providers, are often heterogeneous in terms of device, network, and application runtime contexts. These heterogeneous contexts include the remaining battery level of a mobile device, network signal strength it receives and quality‐of‐service (QoS) requirement of an application software submitted from that device. Scheduling such application software execution requests (from many mobile devices) on competent virtual machines to enhance user quality of experience (QoE) is a multi‐constrained optimization problem. However, existing solutions in the literature either address utility maximization problem for service providers or optimize the application QoS levels, bypassing device‐level and network‐level contextual information. In this paper, a multi‐objective nonlinear programming solution to the context‐aware application software scheduling problem has been developed, namely, QoE and context‐aware scheduling (QCASH) method, which minimizes the application execution times (i.e., maximizes the QoE) and maximizes the application execution success rate. To the best of our knowledge, QCASH is the first work in this domain that inscribes the optimal scheduling problem for mobile application software execution requests with three‐dimensional context parameters. In QCASH, the context priority of each application is measured by applying min–max normalization and multiple linear regression models on three context parameters—battery level, network signal strength, and application QoS. Experimental results, found from simulation runs on CloudSim toolkit, demonstrate that the QCASH outperforms the state‐of‐the‐art works well across the success rate, waiting time, and QoE. Copyright

Quality of Service Aware Reliable Task Scheduling in Vehicular Cloud Computing

Vehicular Cloud Computing (VCC) facilitates real-time execution of many emerging user and intelligent transportation system (ITS) applications by exploiting under-utilized on-board computing resources available in nearby vehicles. These applications have heterogeneous time criticality, i.e., they demand different Quality-of-Service levels. In addition to that, mobility of the vehicles makes the problem of scheduling different application tasks on the vehicular computing resources a challenging one. In this article, we have formulated the task scheduling problem as a mixed integer linear program (MILP) optimization that increases the computation reliability even as reducing the job execution delay. Vehicular on-board units (OBUs), manufactured by different vendors, have different architecture and computing capabilities. We have exploited MapReduce computation model to address the problem of resource heterogeneity and to support computation parallelization. Performance of the proposed solution is evaluated in network simulator version 3 (ns-3) by running MapReduce applications in urban road environment and the results are compared with the state-of-the-art works. The results show that significant performance improvements in terms of reliability and job execution time can be achieved by the proposed task scheduling model.

Oily Residuals Security Threat on Smart Phones

One of the famous and attractive features of a mobile device is the touch screen, especially in smart phones, where the same physical space could be used for various functions in different modes. Smudge or the remaining mark of the figure after touching the screen is very dangerous and could be used by hackers and malware applications to gain sensitive information. In this paper we will study the problem of smudge attacks on smart-phone touch screens and their impacts on e-commerce applications. The main focus will be on how such smudge attacks could be used for fraud and identity spoofing. Finally, we will try to propose some solutions for this attack and discuss the feasibility of their implementations.

A Multistage Credibility Analysis Model for Microblogs

Currently, microblogs such as the well-known social network Twitter are one of the most important sources of information in an era of information overload, restiveness and uncertainty. Consequently, developing models to verify information from Twitter has become both a challenging and necessary task. In this paper, we propose a novel multi-stage credibility analysis framework to identify implausible content in Twitter in order to prevent the proliferation of fake or malicious information. We used Naive Bayes classifier and it is enhanced by considering the relative importance of the used features to improve the classification accuracy. We examine the classifier with 1000 unique tweets along with 700 account. The result quite motivating with accuracy 90.3%, 86.24% Precision and 98.8% recall.

Mining Regularities in Body Sensor Network Data

The recent emergence of body sensor networks BSNs has made it easy to continuously collect and process various health-oriented data related to temporal, spatial and vital sign monitoring of patient. As such, discovering or mining interesting knowledge from the BSN data stream is becoming an important issue to promote and assist important decision making in healthcare. In this paper, we focus on mining the inherent regularity of different parameter readings obtained from different body sensors related to vital sign data of a patent for the purpose of following up health condition to prevent some kinds of chronic diseases. Specifically we design and develop an efficient and scalable regular pattern mining technique that can mine the complete set of periodically/regularly occurring patterns in BSN data stream based on a user-specified periodicity/regularity threshold for the data and the subject. Various experiments were carried on both real and synthetic data to validate the efficiency of the proposed regular pattern mining technique as compared to state-of-the-art approaches.