Wednesday, November 30, 2022


 
 
IMPROVING E-COMMERCE APPLICATION THROUGH SENSE OF AGENCY OF A CALIBRATED INTERACTIVE VR APPLICATION
Nurul Aiman Abdul Rahim, Mohd Adili Norasikin & Zulisman Maksom
 
Shahrizan Jamaludin, Ahmad Faisal Mohamad Ayob, Syamimi Mohd Norzeli & Saiful Bahri Mohamed
 
Friday Zinzendoff Okwonu, Nor Aishah Ahad, Nicholas Oluwole Ogini, Innocent Ejiro Okoloko & Wan Zakiyatussariroh Wan Husin
 

 
A MACHINE LEARNING CLASSIFICATION APPROACH TO DETECT TLS-BASED MALWARE USING ENTROPY-BASED FLOW SET FEATURES
1*Kinan Keshkeh, 2Aman Jantan & 3Kamal Alieyan
1&2School of Computer Sciences, Universiti Sains Malaysia, Malaysia
3Faculty of Computer Sciences and Informatics, Amman Arab University, Amman, Jordan
*kinan.keshkeh@student.usm.my; aman@usm.my; k.alieyan@aau.edu.jo
*Corresponding author
 
 
ABSTRACT
 
Transport Layer Security (TLS) based malware is one of the most hazardous malware types, as it relies on encryption to conceal connections. Due to the complexity of TLS traffic decryption, several anomaly-based detection studies have been conducted to detect TLS-based malware using different features and machine learning (ML) algorithms. However, most of these studies utilized flow features with no feature transformation or relied on inefficient flow feature transformations like frequency-based periodicity analysis and outliers percentage. This paper introduces TLSMalDetect, a TLS-based malware detection approach that integrates periodicity-independent entropy-based flow set (EFS) features generated by a flow feature transformation technique to solve flow feature utilization issues in related research. EFS features effectiveness was evaluated in two ways: (1) by comparing them to the corresponding outliers percentage and flow features using four feature importance methods, and (2) by analyzing classification performance with and without EFS features. Moreover, new Transmission Control Protocol features not explored in literature were incorporated into TLSMalDetect, and their contribution was assessed. This study’s results proved EFS features of the number of packets sent and received were superior to related outliers percentage and flow features and could remarkably increase the performance up to ~42% in the case of Support Vector Machine accuracy. Furthermore, using the basic features, TLSMalDetect achieved the highest accuracy of 93.69% by Naïve Bayes (NB) among the ML algorithms applied. Also, from a comparison view, TLSMalDetect’s Random Forest precision of 98.99% and NB recall of 92.91% exceeded the best relevant findings of previous studies. These comparative results demonstrated the TLSMalDetect’s ability to detect more malware flows out of total malicious flows than existing works. It could also generate more actual alerts from overall alerts than earlier research.
 
Keywords: Malware detection, machine learning, TLS, entropy, flow features.
 
btn fulltext
 

 
IMPROVING E-COMMERCE APPLICATION THROUGH SENSE OF AGENCY OF A CALIBRATED INTERACTIVE VR APPLICATION
*1Nurul Aiman Abdul Rahim, 2Mohd Adili Norasikin & 3Zulisman Maksom
Pervasive Computing & Educational Technology, Fakulti Teknologi Maklumat dan Komunikasi, 
Universiti Teknikal Malaysia Melaka
*p032020001@student.utem.edu.my; adili@utem.edu.my; zulisman@utem.edu.my
*Corresponding author
 
 
ABSTRACT
 
Virtual Reality (VR) technologies create and control different virtual world instead of the actual environment, and this contributes to the feeling of control known as the sense of agency (SoA). The SoA exists from the contrast between the expected sensory consequence of one’s action from efference copy and the real sensory effects. However, the size representation of objects differs between the physical and virtual world due to certain technical limitations, such as the VR application’s virtual hand not reflecting the user’s actual hand size. A limitation that will incur low quality of perception and SoA for digital application. Here, we proposed a proof-of-concept of an interactive e-commerce application that incorporates VR capability and size calibration mechanism. The mechanism uses a calibration method based on the reciprocal scale factor from the virtual object to its real counterpart. The study of the SoA focusing on user perception and interaction was done. The proposed method was tested on twenty two participants − who are also online shopping users. Nearly half of the participants (45%) buy online products frequently, at least one transaction per day. The outcome indicates that our proposed method improves 47% of user perception and interaction compared to the conventional e-commerce application with its static texts and images. Our proposed method is rudimentary yet effective and can be easily implemented in any digital field.
 
Keywords: Virtual environment, sense of agency, virtual hand, online shopping.
 
btn fulltext
 

 
RECENT TRENDS OF MACHINE LEARNING PREDICTIONS USING OPEN DATA: A SYSTEMATIC REVIEW
*1Norismiza Ismail & 2Umi Kalsom Yusof
1&2School of Computer Sciences, Universiti Sains Malaysia, Malaysia
1Digital Management and Development Centre, Universiti Malaysia Perlis, Malaysia
 
 
ABSTRACT
 
Machine learning (ML) prediction determinants based on open data (OD) are investigated in this work, which is accomplished by examining current research trends over ten years. Currently, OD is commonly regarded as the most crucial trend for users to improve their ability to make decisions, particularly to the exponential expansion of social networking sites (SNSs) and open government data (OGD).The purpose of this study was to examine if there was an increase in the usage of OD in ML prediction techniques by conducting a systematic literature review (SLR) of the results of the trends. The papers published in major online scientific databases between 2011 and 2020, including ScienceDirect, Scopus, IEEE Xplore, ACM, and Springer, were identified and analysed. After various selection and Springer, were identified and analysed. After various selection processes, according to SLR based on precise inclusion and exclusion criteria, a total of 302 articles were located. However, only 81 of them were included. The findings were presented and plotted based on the research questions (RQs). In conclusion, this research could be beneficial to organisations, practitioners, and researchers by providing information on current trends in the implementation of ML prediction using OD setting by mapping studies based on the RQs designed, the most recent growth, and the necessity for future research based on the findings.
 
Keywords: Machine learning, open data, prediction, systematic literature review.
 
btn fulltext
 

 
IMAGE-BASED OIL PALM LEAVES DISEASE DETECTION USING CONVOLUTIONAL NEURAL NETWORK
1Jia Heng Ong, *2Pauline Ong & 3Woon Kiow Lee
Faculty of Mechanical and Manufacturing Engineering, University Tun Hussein Onn Malaysia, Malaysia
ongjiaheng0612@gmail.com; *ongp@uthm.edu.my; wklee@uthm.edu.my
*Corresponding author
 
 
ABSTRACT
 
Over the years, numerous studies have been conducted on the integration of computer vision and machine learning in plant disease detection. However, these conventional machine learning methods often require the contour segmentation of the infected region from the entire leaf region and the manual extraction of different discriminative features before the classification models can be developed. In this study, deep learning models, specifically, the AlexNet convolutional neural network (CNN) and the combination of AlexNet and support vector machine (AlexNet-SVM), which overcome the limitation of handcrafting of feature representation were implemented for oil palm leaf disease identification. The images of healthy and infected leaf samples were collected, resized, and renamed before the model training. These images were directly used to fit the classification models, without the need for segmentation and feature extraction as in models, without the need for segmentation and feature extraction as in the conventional machine learning methods. The optimal architecture of AlexNet CNN and AlexNet-SVM models were then determined and subsequently applied for the oil palm leaf disease identification.Comparative studies showed that the overall performance of the AlexNet CNN model outperformed AlexNet-SVM-based classifier.
 
Keywords: AlexNet, convolutional neural network, leaf disease, oilpalm, support vector machine.
 
btn fulltext
 

 
ADAPTIVE INITIAL CONTOUR AND PARTLY-NORMALIZATION ALGORITHM FOR IRIS SEGMENTATION OF BLURRY IRIS IMAGES
*1Shahrizan Jamaludin, 2Ahmad Faisal Mohamad Ayob, 3Syamimi Mohd Norzeli & 4Saiful Bahri Mohamed
1&2Faculty of Ocean Engineering Technology and Informatics, Universiti Malaysia Terengganu, Malaysia
3&4Faculty of Innovative Design and Technology, Universiti Sultan Zainal Abidin, Malaysia
*shahrizanj@umt.edu.my; ahmad.faisal@umt.edu.my; syamiminorzeli@unisza.edu.my; saifulbahri@unisza.edu.my
*Corresponding author
 
 
ABSTRACT
 
Iris segmentation is a process to isolate the accurate iris region from the eye image for iris recognition. Iris segmentation on non-ideal and noisy iris images is accurate with active contour. Nevertheless, it is currently unclear on how active contour responds to blurry iris images or motion blur, which presents a significant obstacle in iris segmentation. Investigation on blurry iris images, especially on the initial contour position, is rarely published and must be clarified. Moreover, evolution or convergence speed remains a significant challenge for active contour as it segments the precise iris boundary. Therefore, this study carried out experiments to achieve an efficient iris segmentation algorithm in terms of accuracy and fast execution, according to the aforementioned concerns. In addition, initial contour was explored to clarify its position. In order to accomplish these goals, the Wiener filter and morphological closing were used for preprocessing and reflection removal. Next, the adaptive initial contour (AIC), δ, and stopping function were integrated to create the adaptive Chan-Vese active contour (ACVAC) algorithm. Finally, the partly -normalization method for normalization and feature extraction was designed by selecting the most prominent iris features. The findings revealed that the algorithm outperformed the other active contour-based approaches in computational time and segmentation accuracy. It proved that in blurry iris images, the accurate initial contour position could be established. This algorithm is significant to solve inaccurate segmentation on blurry iris images.
 
Keywords: Iris segmentation, adaptive initial contour, adaptive Chan-Vese active contour, partly-normalization, segmentation accuracy.
 
btn fulltext
 
 
COMPARATIVE PERFORMANCE EVALUATION OF EFFICIENCY FOR HIGH DIMENSIONAL CLASSIFICATION METHODS
1Friday Zinzendoff Okwonu, *2Nor Aishah Ahad, 3Nicholas Oluwole Ogini, 4Innocent Ejiro Okoloko & 5Wan Zakiyatussariroh Wan Husin
1Department of Mathematics, Faculty of Science, Delta State University, Nigeria
2School of Quantitative Sciences, College of Arts and Sciences, Universiti Utara Malaysia, Malaysia
3Department of Computer Science, Delta State University, Nigeria
4Faculty of Computing, Dennis Osadebay University, Nigeria
5Faculty of Computer and Mathematical Science, Universiti Teknologi MARA, Kelantan Branch, Malaysia
okwonufz@delsu.edu.ng; *aishah@uum.edu.my; ogini@delsu.edu.ng; okoloko@ieee.org;  wanzh@uitm.edu.my
*Corresponding author
 
 
ABSTRACT
 
This paper aimed to determine the efficiency of classifiers for high-dimensional classification methods. It also investigated whether an extreme minimum misclassification rate translates into robust efficiency. To ensure an acceptable procedure, a benchmark evaluation threshold (BETH) was proposed as a metric to analyze the comparative performance for high-dimensional classification methods. A simplified performance metric was derived to show the efficiency of different classification methods. To achieve the objectives, the existing probability of correct classification (PCC) or classification accuracy reported in five different articles was used to generate the BETH value. Then, a comparative analysis was performed between the application of BETH value and the well-established PCC value ,derived from the confusion matrix. The analysis indicated that the BETH procedure had a minimum misclassification rate, unlike the Optimal method. The results also revealed that as the PCC inclined toward unity value, the misclassification rate between the two methods (BETH and PCC) became extremely irrelevant. The study revealed that the BETH method was invariant to the performance established by the classifiers using the PCC criterion but demonstrated more relevant aspects of robustness and minimum misclassification rate as compared to the PCC method. In addition, the comparative analysis affirmed that the BETH method exhibited more robust efficiency than the Optimal method. The study concluded that a minimum misclassification rate yields robust performance efficiency.
 
Keywords: Classification, confusion matrix, efficiency, high dimensional data, robustness.
 
btn fulltext
 

Universiti Utara Malaysia Press
 Universiti Utara Malaysia, 06010 UUM Sintok
Kedah Darul Aman, MALAYSIA
Phone: +604-928 4816, Fax : +604-928 4792

Creative Commons License

All articles published in Journal of Information and Communication Technology (JICT) are licensed under a Creative Commons Attribution 4.0 International License.