Dima Suleiman

SMS Spam Detection using H2O Framework

Abstract SMS spams are one of the concerns and many people do not like to receive them since they are annoying. Many SMS spam detection methods already exist and different classifiers were used, such classifiers depended on Support Vector machine, Naive Bays and many other machine learning algorithms. In this paper, new classifier is proposed which depends mainly on using H2O as platform to make comparisons between different machine learning algorithms. Moreover, Machine learning algorithms that are used for comparisons are random forest, deep learning and naive bays. In addition to using deep learning and random forest as classifiers, they are also used to determine the most important features that can be used as input to random forest, deep learning and naive bays classifiers. Experimental results show that the most significant features that can affect the detection of SMS spam are the number of digits and existing of URL in SMS text. The dataset that is used in experiment is the one proposed by UCI Machine Learning Repositories. Therefore, experiments show that the faster algorithm that achieves high performance is naive bays with runtime 0.6 seconds, however after comparing it with deep learning and random forest it has the lowest precision, recall, f-measure and accuracy. On the other hand, random forest is the best in term of accuracy with 50 trees and 20 maximum depths, where precision, recall, f-measure and accuracy are 96%, 86%, 91% and 0.977% respectively; nevertheless the runtime is high 30.28 seconds.

Bag-of-concept based keyword extraction from Arabic documents

This paper proposes a new keyword extraction method that uses bag-of-concept to extract keywords from Arabic text. The proposed algorithm utilizes semantic vector space model instead of traditional vector space model to group words into classes. The new method built word-context matrix where the synonym words will be grouped into the same class. The evaluation of new approach was conducted using dataset which consists of three documents and compared with Keyword Extraction from Arabic Documents using Term Equivalence Classes method; experimental results showed that the proposed method provides significant results.

Statistical Arabic Name Entity Recognition Approaches: A Survey

Abstract With the increase of Arabic textual information via internet websites and services, a tools for processing Arabic text information are needed to extract knowledge from them. Name Entity recognition aims to extract name entities such as: person names, locations and organizations from a given text. Name Entity recognition approaches were classified into two main approaches: rule-based approach and statistical approach. Although the literature on Name Entity recognition is quit extensive, few researches to extract Name Entities in the Arabic language could be found. This paper provides a comprehensive survey about statistical approaches of Arabic Name Entity extraction.

An Empirical Evaluation of Intelligent Machine Learning Algorithms under Big Data Processing Systems

Abstract The rapid increase in the magnitude of data produced by industries that need to be processed using Machine Learning algorithms to generate business intelligence has created a dilemma for data scientists. This is due to the fact that traditional machine learning platforms such as Weka and R are not designed to handle data with such Volume, Velocity and Variety. Several machine learning algorithms and associated toolkits have been built specifically to work with big data; however, their performance is yet to be evaluated to allow researchers to get the most of these platforms. In this paper, the authors intend to provide an empirical evaluation of two emerging machine learning platforms under big data processing systems namely, H2O and Sparkling Water, by performing an experimental comparison between the two platforms in terms of performance over several generalization error metrics and model training time using the Santander Bank Dataset. Up to the authors’ knowledge, this is the first time such a study is conducted. The evaluation results showed that the H2O platform has significantly outperformed the Sparkling Water platform in terms of model training time almost by fifty percent, while achieving convergent results.

The Use of Hidden Markov Model in Natural ARABIC Language Processing: a survey

Abstract Hidden Markov Model is an empirical tool that can be used in many applications related to natural language processing. In this paper a comparative study was conducted between different applications in natural Arabic language processing that uses Hidden Markov Model such as morphological analysis, part of speech tagging, text classification, and name entity recognition. Comparative results showed that HMM can be used in different layers of natural language processing, but mainly in pre-processing phase such as: part of speech tagging, morphological analysis and syntactic structure; however in high level applications text classification their use is limited to certain number of researches.

An experimental Software Defined Security controller for Software Defined Network

Software Defined Networking is an emerging technology that permits computer network infrastructure to be scaled dynamically as needed while enhancing the manageability of the various network devices in heterogeneous environment as opposed to classical networking. These capabilities emerge from the separation of the data plane from the control plane; thus, allowing the network devices to be programmatically managed and controlled. Similarly, the concept of Software Defined Security allows security solutions such as Firewalls and Intrusion Detection Systems to be dynamically implemented, controlled and managed using programmable interfaces. In this research, the authors propose an experimental software defined security controller based on the Open vSwitch Controller to detect and prevent IP and MAC spoofing attacks on the network. The proposed controller is simulated using Mininet. The simulation results confirm that the proposed controller is capable to detect and prevent the aforementioned attacks with high precision.

Test Case Reduction Techniques - Survey

Regression testing is considered to be the most expensive phase in software testing. Therefore, regression testing reduction eliminates the redundant test cases in the regression testing suite and saves cost of this phase. In order to validate the correctness of the new version software project that resulted from maintenance phase, Regression testing reruns the regression testing suite to ensure that the new version. Several techniques are used to deal with the problem of regression testing reduction. This research is going to classify these techniques regression testing reduction problem.