Claim Missing Document
Check
Articles

Found 6 Documents
Search

A comparative study of machine learning tools for detecting Trojan horse infections in cloud computing environments Kanaker, Hasan; Tarawneh, Monther; Karim, Nader Abdel; Alsaaidah, Adeeb; Abuhamdeh, Maher; Qtaish, Osama; Alhroob, Essam; Alhalhouli, Zaid
International Journal of Electrical and Computer Engineering (IJECE) Vol 14, No 6: December 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijece.v14i6.pp6642-6655

Abstract

Cloud computing offers several advantages, including cost savings and easy access to resources, it is also could be vulnerable to serious security attacks such as cloud Trojan horse infection attacks. To address this issue, machine learning is a promising approach for detecting these threats. Thus, different machine learning tools and models have been employed to detect Trojan horse infection such as Weka and Python Colab. This study aims to compare the performance of Weka and Python Colab, as popular tools for building machine learning models. This study evaluates the recall, accuracy, and F1-score of machine learning models built with Weka and Python Colab and compares their computational resources required employing several machine learning algorithms. The dataset collected and analyzed using dynamic analysis of Trojan horse infection in control lab environment. The findings of this study can help determine the decision about which tool to use to detect Trojan horse infections and provide insights into the strengths and limitations of Weka and Python Colab for building machine-learning models in general.
Enhancing Spam Detection Using Hybrid of Harris Hawks and Firefly Optimization Algorithms Abualhaj, Mosleh M.; Shambour, Qusai Y.; Alsaaidah, Adeeb; Abu-Shareha, Ahmad; Al-Khatib, Sumaya; Hiari, Mohammad O.
Journal of Applied Data Sciences Vol 5, No 3: SEPTEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i3.279

Abstract

The emergence of the modern Internet has presented numerous opportunities for attackers to profit illegally by distributing spam mail. Spam refers to irrelevant or inappropriate messages that are sent on the Internet to numerous recipients. Many researchers use many classification methods in machine learning to filter spam messages. However, more research is still needed to assess using metaheuristic optimization algorithms to classify spam emails in feature selection. In this paper, we endorse fighting spam emails by employing a union of Firefly Optimization Algorithm (FOA) and Harris Hawks Optimization (HHO) algorithms to classify spam emails, along with one of the most well-known and efficient methods in this area, the Random Forest (RF) classifier. In this process, the experimental studies on the ISCX-URL2016 spam dataset yield promising results. For instance, the union of HHO and FOA, along with using an RF classifier, achieved an accuracy of 99.83% in detecting spam emails.
Comparative analysis of whale and Harris Hawks optimization for feature selection in intrusion detection Abualhaj, Mosleh M.; Hiari, Mohammad O.; Alsaaidah, Adeeb; Al-Zyoud, Mahran M.
Indonesian Journal of Electrical Engineering and Computer Science Vol 37, No 1: January 2025
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v37.i1.pp179-185

Abstract

This research paper explores the efficacy of two nature-inspired optimization algorithms, the whale optimization algorithm (WOA) and Harris Hawks optimization (HHO), for feature selection in the context of intrusion detection and prevention systems (IDPS). Leveraging the NSL-KDD dataset as a benchmark, our study employs Python for implementation and uses decision tree (DT) as the classification model. The objective is to assess the impact of the HHO and WOA optimization techniques on the performance of IDPS through feature selection. The WOA and HHO techniques were able to lessen the features from 40 to 16 and 13, respectively. Results indicate that DT integrated with HHO achieves an impressive accuracy of 97.59%, outperforming the WOA-enhanced model, which attains an accuracy of 97.5%. This study contributes valuable insights into the comparative effectiveness of WOA and HHO optimization algorithms in enhancing the accuracy of IDPSs, shedding light on their potential applications in the realm of cybersecurity.
Exploring the dynamics of providing cognition using a computational model of cognitive insomnia Rateb, Roqia; M. Abualhaj, Mosleh; Alsaaidah, Adeeb; A. Alsharaiah, Mohammad; Shorman, Amaal; Jaber Thalji, Nisrean
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 14, No 1: February 2025
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v14.i1.pp92-101

Abstract

Insomnia is a common sleep-related neuropsychological disorder that can lead to a range of problems, including cognitive deficits, emotional distress, negative thoughts, and a sense of insufficient sleep. This study proposes a providing computational dynamic cognitive model (PCDCM) insight into providing cognitive mechanisms of insomnia and consequent cognitive deficits. Since the support providing is significantly dynamic and it includes substantial changes as demanding condition happen. From this perspective the underlying model covers integrating of both coping strategies, provision preferences and adaptation concepts. The model was found to produce realistic behavior that could clarify conditions for providing support to handle insomnia individuals, which was done by employing simulation experiments under various negative events, personality resources, altruistic attitude and personality attributes. Simulation results show that, a person with bonadaptation and either problem focused or emotion focused coping can provide different social support based on his personality resources, personality attributes, and knowledge level, whereas a person with maladaptation regardless the coping strategies cannot provide any type of social support. Moreover, person with close tie tends to provide instrumental, emotional, and companionship support than from weak tie. Finally, a mathematical analysis was used to examine the possible equilibria of the model. 
ARP Spoofing Attack Detection Model in IoT Network using Machine Learning: Complexity vs. Accuracy Alsaaidah, Adeeb; Almomani, Omar; Abu-Shareha, Ahmad Adel; Abualhaj, Mosleh M; Achuthan, Anusha
Journal of Applied Data Sciences Vol 5, No 4: DECEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i4.374

Abstract

Spoofing attacks targeting the address resolution protocol, or the so-called ARP, are common cyber-attacks in IoT environments. In such an attack, the attacker sends a fake message over a local area network to spoof the users and interfere with the communication transferred from and into these users. As such, to detect such attacks, there is a need to check the network gateways and routers continuously to capture and analyze the transmitted traffic. However, there are three major problems with such traffic data: 1) there are substantial irrelevant data to the ARP attacks, 2) there are massive patterns in the way by which the spoof can be implemented, and 3) there is a need for fast processing of such data to reduce any delay resulting from the processing stage. Accordingly, this paper proposes a detection approach using supervised machine learning algorithms. The focus of this paper is to show the tradeoff between speed and accuracy to offer various solutions based on the demanded quality. Various algorithms were tested to find a solution that balanced time requirements and accuracy. As such, the results using all features and with various feature selection techniques were reported. Besides, the results using simple classifiers and ensemble learning algorithms were also reported. The proposed approach is evaluated on an IoT network intrusion dataset (IoTID20) collected from different IoT devices. The results showed that the highest accuracy is obtained using the RF classifier with a subset of features produced by the wrapper technique. In such a case, the accuracy obtained was 99.74%, with running time equal to 305 milliseconds. However, If time is more critical for a given application, then DT can be used with the whole feature set. In such a case, the accuracy was 99.41%, with running time equal to 11  milliseconds.
Spam Feature Selection Using Firefly Metaheuristic Algorithm Abualhaj, Mosleh M; Hiari, Mohammad O; Alsaaidah, Adeeb; Al-Zyoud, Mahran; Al-Khatib, Sumaya
Journal of Applied Data Sciences Vol 5, No 4: DECEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i4.336

Abstract

This paper presents a novel method for improving spam detection by utilizing the Firefly Algorithm (FA) for feature selection. The FA, a bio-inspired metaheuristic optimization algorithm, is applied to identify the most relevant features from the ISCX-URL2016 dataset, which contains 72 features. By balancing exploration (searching for new solutions) and exploitation (focusing on the best solutions), FA is able to effectively reduce the feature space from 72 to 31 features. This reduction improves model efficiency without sacrificing performance, as only the most impactful features are retained for the classification task. The selected features were then used to train three machine learning classifiers: Decision Tree (DT), Gradient Boost Tree (GBT), and Naive Bayes (NB). Each classifier's performance was evaluated based on accuracy, with DT achieving the highest accuracy of 99.81%, GBT achieving 99.70%, and NB scoring 90.33%. The superior performance of the DT algorithm is attributed to its ability to handle non-linear relationships and high-dimensional data, making it particularly well-suited for the FA-selected features. This combination of FA for feature selection and DT for classification demonstrates significant improvements in spam detection performance, highlighting the importance of selecting the most relevant features. The results show that by reducing the dimensionality of the dataset, the FA algorithm not only accelerates the classification process but also enhances detection accuracy.