Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to main content
The rise of fake news has been the subject of several research studies in the last decade. This is due to the increasing number of Internet users and the simplicity in posting news over platforms and websites. Hence, researchers have been... more
The rise of fake news has been the subject of several research studies in the last decade. This is due to the increasing number of Internet users and the simplicity in posting news over platforms and websites. Hence, researchers have been developing machine learning (ML) models to detect fake contents and warn readers. However, there is a limited number of Arabic fake news datasets in terms of articles and news sources. This paper aims at introducing the first large Arabic fake news corpus which consists of 606912 articles collected from 134 Arabic online news sources. An Arabic fact-check platform is used to annotate news sources as credible, not-credible, and undecided. Moreover, different ML algorithms are used for the detection task. Experiments show that deep learning models perform better than traditional ML models. Models training showed underfitting and overfitting problems which indicate that the corpus is noisy and challenging.
The rise of fake news has been the subject of several research studies in the last decade. This is due to the increasing number of Internet users and the simplicity in posting news over platforms and websites. Hence, researchers have been... more
The rise of fake news has been the subject of several research studies in the last decade. This is due to the increasing number of Internet users and the simplicity in posting news over platforms and websites. Hence, researchers have been developing machine learning (ML) models to detect fake contents and warn readers. However, there is a limited number of Arabic fake news datasets in terms of articles and news sources. This paper aims at introducing the first large Arabic fake news corpus which consists of 606912 articles collected from 134 Arabic online news sources. An Arabic fact-check platform is used to annotate news sources as credible, not-credible, and undecided. Moreover, different ML algorithms are used for the detection task. Experiments show that deep learning models perform better than traditional ML models. Models training showed underfitting and overfitting problems which indicate that the corpus is noisy and challenging.
ABSTRACT The design of an effective last-level cache (LLC) is crucial to the overall processor performance and, consequently, continues to be the center of substantial research. Unfortunately, LLCs in modern high-performance processors... more
ABSTRACT The design of an effective last-level cache (LLC) is crucial to the overall processor performance and, consequently, continues to be the center of substantial research. Unfortunately, LLCs in modern high-performance processors are not used efficiently. One major problem suffered by LLCs is their low hit rates caused by the large fraction of cache blocks that do not get re-accessed after being brought into the LLC following a cache miss. These blocks do not contribute any cache hits and usually induce cache pollution and thrashing. Cache bypassing presents an effective solution to this problem. Cache blocks that are predicted not to be accessed while residing in the cache are not inserted into the LLC following a miss, instead they bypass the LLC and are only inserted in the higher cache levels. This paper presents a simple, low-hardware overhead, yet effective, cache bypassing algorithm that dynamically chooses which blocks to insert into the LLC and which to bypass it following a miss based on past access/bypass patterns. Our proposed algorithm is thoroughly evaluated using a detailed simulation environment where its effectiveness, performance-improvement capabilities, and robustness are demonstrated. Moreover, it is shown to outperform the state-of-the-art cache bypassing algorithm in both a uniprocessor and a multi-core processor settings.
The first large Arabic fake news corpus consists of 606912 articles collected from 134 Arabic online news sources. An Arabic fact-check platform is used to annotate news sources as credible, not-credible, and undecided. The details of the... more
The first large Arabic fake news corpus consists of 606912 articles collected from 134 Arabic online news sources. An Arabic fact-check platform is used to annotate news sources as credible, not-credible, and undecided. The details of the selected news sources are stored in a single JSON file. Each online news source has a key and value. The key is the source name and the value is a dictionary that contains the source label and two lists. The first list has the RSS website links and the second one has the website links of the local news pages. The news articles in the Arabic language are stored in 134 JSON files. Each file corresponds to each source and contains the details of each news article which includes the headline, news content, publish date, and the article link.
Manual video inspection, searching, and analyzing is exhausting and inefficient. This paper presents an intelligent system to search surveillance video contents using deep learning. The proposed system reduced the amount of work that is... more
Manual video inspection, searching, and analyzing is exhausting and inefficient. This paper presents an intelligent system to search surveillance video contents using deep learning. The proposed system reduced the amount of work that is needed to perform video searching and improved the speed and accuracy. A pre-trained VGG-16 CNNs model is used for dataset training. In addition, key frames of videos were extracted in order to save space, reduce the amount of work, and reduce the execution time. The extracted key frames were processed using the sobel operator edge detector and the max-pooling in order to eliminate redundancy. This increases compaction and avoids similarities between extracted frames. A text file, that contains key frame index, time of occurrence, and the classification of the VGG-16 model, is produced. The text file enables humans to easily search for objects of interest. VIRAT and IVY LAB datasets were used in the experiments. In addition, 128 different classes wer...
Fast Flux service networks (FFSNs) are used by adversaries to achieve a high resilient technique for their malicious servers while keeping them hidden from direct access. In this technique, a large number of botnet machines, that are... more
Fast Flux service networks (FFSNs) are used by adversaries to achieve a high resilient technique for their malicious servers while keeping them hidden from direct access. In this technique, a large number of botnet machines, that are known as flux agents, work as proxies to relay the traffic between end users and a malicious mothership server which is controlled by an adversary. Various mechanisms have been proposed for detecting FFSNs. Such mechanisms depend on collecting a large amount of DNS traffic traces and require a considerable amount of time to identify fast flux domains. In this paper, we propose an efficient AI-based online fast flux detection system that performs highly accurate and extremely fast detection of fast flux domains. The proposed system, called PASSVM, is based on features that are associated with DNS response messages of a given domain name. The approach relies on features that are stored in two local databases, in addition to features that are extracted fro...
Smart home management systems are essential for smart buildings and cities. The heterogeneity of smart devices in terms of performance, heating behaviors, and energy consumption can result in different patterns of smart home behaviors and... more
Smart home management systems are essential for smart buildings and cities. The heterogeneity of smart devices in terms of performance, heating behaviors, and energy consumption can result in different patterns of smart home behaviors and measurements. This paper proposes a modeling and simulation approach using Discrete Event System Specification (DEVS) formalism to simulate the behaviors in terms of power usage and cost in smart home devices under different scenarios and settings. Smart devices are divided into two categories: monitoring devices and control devices. Monitoring devices include sensors that capture data of climate, energy, electrical power usage, performance, and occupants' behavior. Control devices are used to send signals for setting and remote controlling of the different devices in the smart home. The result of this work enables designers to provide key solutions such as automatic control and monitoring of the power consumption, electrical load, and heating ...
The communication networks of low-resources applications require implementing cryptographic protocols and operations with less computational and architectural complexities. In this paper, an efficient method for high speed calculations of... more
The communication networks of low-resources applications require implementing cryptographic protocols and operations with less computational and architectural complexities. In this paper, an efficient method for high speed calculations of square (SQR) root is proposed over Galois Fields GF (2m). The method is based on using the results of certain pre-computations, and transforming the SQR root calculations into a system of linear equations. The computational complexity of our proposed method for computing the SQR root in GF (2m) is O(m) which is significantly better than existing methods such as Tonelli-Shanks and Cipolla. Our proposed method was implemented using different types of multipliers over several polynomial degrees. Software and hardware implementations were developed in NTL-C++ and VHDL, respectively. Our software experimental results show up to 38 times faster than Doliskani & Schost method. Moreover, our method is 840 times faster than Tonelli-Shanks method. In terms o...
In this paper, we investigate the joint optimization of different smart grid components. Unlike most of previous works that optimize only one objective function, this paper proposes a multi-objective optimization solution that includes... more
In this paper, we investigate the joint optimization of different smart grid components. Unlike most of previous works that optimize only one objective function, this paper proposes a multi-objective optimization solution that includes the energy cost, users' comfort level, and the lifetime of storage devices. Our solution is based on the Multi-Objective Evolutionary Algorithm (MOEA) framework. Experimental results, based on Reference Energy Disaggregation Data Set (REDD) representing real power demand from different houses, show that our solution provides energy saving by up to 46% while keeping the comfort level above 70% and the lifetime of storage devices above 8 years by reducing the number of charging/discharging cycles.
Pattern matching is the most time consuming task in many cybersecurity, bioinformatics and computational biological applications. Speeding up the pattern matching task is an essential step for the success of the aforementioned... more
Pattern matching is the most time consuming task in many cybersecurity, bioinformatics and computational biological applications. Speeding up the pattern matching task is an essential step for the success of the aforementioned applications. Wu–Manber algorithm is one of the fastest and most widely used algorithms for multi-pattern matching. Many researchers focused on improving the performance of Wu–Manber algorithm and this work presents a novel attempt parallelize Wu–Manber and make it suitable for multi-core machines. This paper uses Kahn processing network (KPN) model to effectively parallelize data and functional tasks. KPN suggests a parallel programming model that can be utilized in today’s multi-core machines. Hence, we use the KPN model to tailor the execution of Wu–Manber algorithm by breaking down the complexity of data sharing and task processing. The data parallelization is implemented using concurrent executions of multiple KPNs. In addition, task parallelization is achieved within each executing KPN. A single KPN consists of two threads, a producer thread and a consumer thread. The proposed KPN-based parallelization achieves up to 4× speedup over the serial implementation of the algorithm. Finally, the algorithm performance scales well with increasing workloads and the speedup up remains almost constant with increasing number of attack signatures.
The ability to manage and exploit geographically distributed systems of service providers is rather limited in today engineering solutions. Existing techniques suffer from three main problems: first, current techniques cannot provide... more
The ability to manage and exploit geographically distributed systems of service providers is rather limited in today engineering solutions. Existing techniques suffer from three main problems: first, current techniques cannot provide brokering in managing loosely coupled service providers. Second, the engineering design of existing management tools does not provide enough expressive capabilities for varying user behaviors or when different domains are encountered. Third, lack of interaction between different requestors and providers yields inefficient and very costly agreements. In this thesis, we will present an automated Domain-Independent Marketplace architecture that allows user agents to interact with provider agents using two simple and yet powerful negotiation protocols which define the rules of interactions in multi-agent environments. Having a trusted third party marketplace supports privacy and transparency among collaborative agents and service providers. Service provider...
Competitive markets in supply chains choose not to share their inventory, backlog, and revenue costs and hence global information is not available. In this paper, we propose a new framework for supply chain management based on trusted... more
Competitive markets in supply chains choose not to share their inventory, backlog, and revenue costs and hence global information is not available. In this paper, we propose a new framework for supply chain management based on trusted mediator agents. A mediator agent places an order on behalf of its customer to a corresponding supplier. The agents use local information and apply adaptive heuristic rules in order to enhance the performance of the entire supply chain. We have evaluated our framework through conducting extensive experiments in an agent-based modeling and simulation environment. The results show a consistent improvement in all the cases that were considered in the literature. We show that local information can in fact lead artificial mediator agents to discover effective ordering strategies.
In agent-based simulation, emergent equilibrium describes the macroscopic steady states of agents' interactions. While the state of individual agents might be changing, the collective behavior pattern remains the same in macroscopic... more
In agent-based simulation, emergent equilibrium describes the macroscopic steady states of agents' interactions. While the state of individual agents might be changing, the collective behavior pattern remains the same in macroscopic equilibrium states. Traditionally, these emergent equilibriums are calculated using Monte Carlo methods. However, these methods require thousands of repeated simulation runs, which are extremely time-consuming. In this paper, we propose a novel three-layer framework to efficiently compute emergent equilibriums. The framework consists of a macro-level pseudo-arclength equilibrium solver (PAES), a micro-level simulator (MLS) and a macro-micro bridge (MMB). It can adaptively explore parameter space and recursively compute equilibrium states using the predictor-corrector scheme. We apply the framework to the popular opinion dynamics and labour market models. The experimental results show that our framework outperformed Monte Carlo experiments in terms of...
String matching algorithms are computationally intensive operations in computer science. The algorithms fi nd the occurrences of one or more strings patterns in a larger string or text. String matching algorithms are important for network... more
String matching algorithms are computationally intensive operations in computer science. The algorithms fi nd the occurrences of one or more strings patterns in a larger string or text. String matching algorithms are important for network security, biomedical applications, Web search, and social networks. Nowadays, the high network speeds and large storage capacity put a high requirement on string matching methods to perform the task in a short time. Traditionally, Aho-Corasick algorithm, which is used to fi nd the string matches, is executed sequentially. In this paper, a new multi-threaded and interleaving approach of Aho-Corasick using graphics processing units (GPUs) is designed and implemented to achieve high-speed string matching. Compute Unified Device Architecture (CUDA) programming language is used to implement the proposed parallel version. Experimental results show that our approach achieves more than 5X speedup over the sequential and other para llel implementations. Hen...
Manual video inspection, searching, and analyzing is exhausting and inefficient. This paper presents an intelligent system to search surveillance video contents using deep learning. The proposed system reduced the amount of work that is... more
Manual video inspection, searching, and analyzing is exhausting and inefficient. This paper presents an intelligent system to search surveillance video contents using deep learning. The proposed system reduced the amount of work that is needed to perform video searching and improved the speed and accuracy. A pre-trained VGG-16 CNNs model is used for dataset training. In addition, key frames of videos were extracted in order to save space, reduce the amount of work, and reduce the execution time. The extracted key frames were processed using the sobel operator edge detector and the max-pooling in order to eliminate redundancy. This increases compaction and avoids similarities between extracted frames. A text file, that contains key frame index, time of occurrence, and the classification of the VGG-16 model is produced. The text file enables humans to easily search for objects of interest. VIRAT and IVY LAB datasets were used in the experiments. In addition, 128 different classes were...
The rapid increase in wired Internet speed and the constant growth in the number of attacks make network protection a challenge. Intrusion detection systems (IDSs) play a crucial role in discovering suspicious activities and also in... more
The rapid increase in wired Internet speed and the constant growth in the number of attacks make network protection a challenge. Intrusion detection systems (IDSs) play a crucial role in discovering suspicious activities and also in preventing their harmful impact. Existing signature-based IDSs have significant overheads in terms of execution time and memory usage mainly due to the pattern matching operation. Therefore, there is a need to design an efficient system to reduce overhead. This research intends to accelerate the pattern matching operation through parallelizing a matching algorithm on a multi-core CPU. In this paper, we parallelize a bit-vector algorithm, Myers algorithm, on a multi-core CPU under the MapReduce framework. On average, we achieve four times speedup using our multi-core implementations when compared to the serial version. Additionally, we use two implementations of MapReduce to parallelize the Myers algorithm using Phoenix++ and MAPCG. Our MapReduce parallel...
Alexis Tourapis Toshiyuki Amano Huiyu Zhou Antonio Ortega John Barron Bedrich Benes Berlin Chen Francois Berry Mingli Song Clinton Fookes Carlos Vazquez Chi-Fa Chen Changbo Hu Cha Zhang Gene Cheung Chin-Hun Teng Chunhua Shen Rong Chu Chuo... more
Alexis Tourapis Toshiyuki Amano Huiyu Zhou Antonio Ortega John Barron Bedrich Benes Berlin Chen Francois Berry Mingli Song Clinton Fookes Carlos Vazquez Chi-Fa Chen Changbo Hu Cha Zhang Gene Cheung Chin-Hun Teng Chunhua Shen Rong Chu Chuo Hao Yeo ...
ABSTRACT The use of ontologies has been applied to many scientific disciplines to systematically enhance data and resources management. This paper presents a new modeling approach to manage resources in distributed computing environments.... more
ABSTRACT The use of ontologies has been applied to many scientific disciplines to systematically enhance data and resources management. This paper presents a new modeling approach to manage resources in distributed computing environments. Information about distributed nodes is arranged into an ontology that is understandable by different nodes inside the cluster. In designing the ontology, we use System Entity Structure formalism (SES). The distributed environment specifications along with the SES-based ontology generate a Pruned Entity Structure (PES) that describes the grid. Grid simulators can interact with the PES ontology to extract information that is needed for their run. In order to simplify the process of describing a distributed environment, we developed a user friendly interface that allows users to build the distributed environment by providing the computing sites and their specifications. In addition, the users draw the network topology that connects the different sites. The information is mapped into PES automatically. We validate our modeling and simulation methodology using the OptorSim simulator that is developed by the European Organization for Nuclear Research (CERN).
Research Interests:

And 16 more