Monday, August 03, 2020

These articles have been peer-reviewed and accepted for publication in JICT, but are pending final changes, are not yet published and may not appear here in their final order of publication until they are assigned to issues. Therefore, the content conforms to our standards but the presentation (e.g. typesetting and proof-reading) is not necessarily up to the JICT standard. Additionally, titles, authors, abstracts and keywords may change before publication.
Shamini Raja Kumaran, Mohd Shahizan Othman & Lizawati Mi Yusuf
Faculty Engineering, School of Computing, Universiti Teknologi Malaysia;;
Missing values were a huge constraint in microarray technologies toward improving and identifying the disease-causing genes. Estimating missing values is an undeniable scenario faced by the experts and the imputation methods are an effective way to impute the proper values to proceed with the next processes in microarray technology. Missing value imputation approaches may increase the classification accuracy. Although these approaches might predict the values, the accuracy rates prove its abilities of these approaches to identify the missing values in the gene expression data. In this article, a novel approach, optimized hybrid fuzzy c-Means and majority vote (opt-FCMMV), is proposed to identify the missing values in the data. Using the majority vote (MV) and optimization through particle swarm optimization (PSO), this approach predicts the missing values in the data to form more informative and solid data. In order to verify the effectiveness of opt-FCMMV, several experiments were carried out on two publicly available microarray datasets (ovary and lung cancer samples) under three missing value mechanisms with five different percentage values in the biomedical domain using Support Vector Machine (SVM) classifier. The experimental results showed that the proposed approach functioned efficiently by showcasing highest accuracy rates compared to no imputations, FCM and FCMMV. For exemplary, the accuracy rates for Ovary data with 5% missing values were no imputation (64.0%), FCM (81.8%), FCMMV (90.0%) and opt-FCMMV (93.7%). While for future work, other metaheuristic algorithms can be used to solve missing values via optimization to improve the performance of data.
Keywords: fuzzy c-Means, majority vote, missing values, microarray data, data optimization.

1Yoanes Bandung & 1Joshua Tanuraharja
1School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Indonesia;
QoS provisioning for real-time multimedia applications is largely determined by network’s available bandwidth. Until now there is no standard method for estimating bandwidth on wireless networks. Therefore, in this study a mathematical model called Modified Passive Available Bandwidth Estimation (MPABE) was developed to estimate the available bandwidth passively on a Distributed Coordination Function (DCF) wireless network on the IEEE 802.11 protocol. The mathematical model developed was a modification of the three existing mathematical models, namely Available Bandwidth Estimation (ABE), Cognitive Passive Estimation of Available Bandwidth V2 (cPEAB-V2), and Passive Available Bandwidth Estimation (PABE). The proposed mathematical model gives emphasis on what will be faced to estimate available bandwidth and will help in the coming up of the strategies to estimate available bandwidth on the IEEE 802.11. The developed mathematical model consisted of idle-period synchronization between sender and receiver, the overhead probability occurring in the Medium Access Control (MAC) layer, as well as the successful packet transmission probability. A successful packet transmission is influenced by three variables, the packet collision probability caused by a number of neighboring nodes, the packet collision probability caused by traffic from the hidden nodes, and the packet error probability. The proposed mathematical model was tested by comparing it with other relevant mathematical models. The performance of the four mathematical models was compared with the actual bandwidth. Using a series of experiments that have been done, it is found that the proposed mathematical model is approximately 26% more accurate than the ABE, 36% more accurate than the cPEAB-V2, and 32% more accurate than the PABE.
Keywords: Available bandwidth estimation, distributed coordination function, IEEE 802.11, hidden nodes.

1Noor Huzaimi@Karimah Mohd Noor, 2Shahrul Azman Mohd Noah & 2Mohd Juzaiddin Ab Aziz
1Faculty of Computing, Universiti Malaysia Pahang, Malaysia
2Faculty of Information Science & Technology, Universiti Kebangsaan Malaysia, Malaysia; shahrul,
Anaphor candidate determination is an important process in Anaphora Resolution (AR) systems. There are several types of anaphor, one of which is pronominal anaphor. Pronominal anaphor is an anaphor that involves pronoun. In some of the cases, certain pronouns can be used without referring to any situation or entities in a text, and this phenomenon is known as pleonastic. In the case of the Malay language, it usually occurs for the pronoun nya. The pleonastic that exists in every text causes a severe problem to the anaphora resolution systems. The process to determine the pleonastic nya is not the same as identifying pleonastic ‘it’ in the English language, where the syntactic pattern could not be used because the structure of nya comes at the end of a word. As an alternative, the semantic classes are used to identify the pleonastic itself and the anaphoric nya. In this paper, the automatic semantic tag is used to determine the type of nya, which at the same time can determine nya as an anaphor candidate. The new algorithms and MalayAR architecture are proposed. The results of the F-measure showed the detection of clitic nya as a separate word achieved a perfect 100% result. In comparison, the clitic nya as a pleonastic achieved 88%, clitic nya referring to humans achieved 94%, and clitic nya for referring to non-humans achieved 63%. The results showed that the proposed algorithms are acceptable to solve the issue of the clitic nya as pleonastic, human referral as well as non-human referral.
Keywords: Anaphora resolution, natural language processing, Malay anaphora resolution, anaphor candidate determination.

1,2Mohammad Raquibul Hossain & 1Mohd Tahir Ismail
1School of Mathematical Sciences, Universiti Sains Malaysia, Malaysia.
2Department of Applied Mathematics, Noakhali Science and Technology University, Bangladesh.;
Forecasting is a challenging task as time-series data exhibit many features that cannot be captured by a single model. Thus, many researchers have proposed some hybrid model in order to accommodate these features to improve forecasting results. This work proposes a hybrid method between Empirical Mode Decomposition (EMD) and Theta method by considering better forecasting potentiality. Both EMD and Theta are efficient methods in their own ground of tasks for decomposition and forecasting, respectively. Combining them to get better synergic outcome deserves consideration. EMD decomposed training data from each of the five FTSE 100 index (Financial Times Stock Exchange 100 Index) companies’ stock price time series data into Intrinsic Mode Functions (IMF) and a residue. Then the Theta method forecasted each decomposed subseries. Considering different forecast horizons, the effectiveness of this hybridization was evaluated through values of some conventional error measures found for test data and forecast data which is obtained by adding forecast results for all component counterparts extracted from EMD process. This study found that the proposed method produced better forecast accuracy than the other three classic methods and the hybrid EMD-ARIMA models.
Keywords: Forecasting Stock Price, Empirical Mode Decomposition (EMD), Intrinsic Mode Functions (IMF), Theta Method, Time Series.

Ashwindran Naidu Sanderasagran, Azizuddin Abd Aziz & Daing Mohamad Nafiz Daing Idris
Faculty of Mechanical and Automotive Engineering Technology, Universiti Malaysia Pahang, Malaysia.; azizuddin,
The behavior of fluid flow is a complex paradigm for cognitive interpretation and visualization. Engineers need to visualize the behavior mechanics of flow field response in order to enhance the cognitive ability in problem solving. Therefore, mixed reality related technology is the solution for enhanced virtual interactive learning environment. However, there are limited AR platforms on fluid flow interactive learning. Hence, an interactive education app is proposed for students and engineers to interact and understand the complex flow behavior pattern subjected to elementary geometry body relative to external flow. This paper presents the technical development of a real-time flow response visualization augmented reality app for computational fluid dynamics application. It is developed with the assistance of several applications such as Unity, Vuforia, and Android IDE. Particle system modules available in Unity engine are used to create a 2D flow stream domain. The flow visualization and interaction are limited to 2D and the numerical fluid continuum response was not analyzed. The physical flow response pattern of three simple geometry bodies is validated against ANSYS simulated results based on visual empirical observation. The particle size and number of particles emitted are adjusted in order to emulate the physical representation of fluid flow. Color contour is set to change according to fluid velocity. Visual validation indicates trivial dissimilarities between FLUENT generated results and flow response exhibited by the proposed augmented reality app.
Keywords: Augmented reality, computational fluid dynamics, image target, vuforia, unity engine, particle system.

1Zahra Bokaee Nezhad & 2Mohammad Ali Deihimi
1Department of Computer Engineering, Zand University, Iran
2Department of Electronics Engineering, Bahonar University, Iran;
Sarcasm is a form of communication where the individual states the opposite of what is implied. Hence, detecting a sarcastic tone is somewhat complicated due to its ambiguous nature. On the other hand, identification of sarcasm is vital to various Natural Language Processing (NLP) tasks such as sentiment analysis and text summarization. However, research on sarcasm detection in Persian is very limited. Therefore, we investigate sarcasm detection technique on Persian tweets by combining deep learning-based and machine learning-based approaches. We propose four sets of features that cover different types of sarcasm. They are deep polarity feature, sentiment feature, part of speech feature and punctuation feature. We use these features to classify tweets as sarcastic and non-sarcastic. In this study, the deep polarity feature is proposed by conducting sentiment analysis using deep neural network architecture. In addition, to extract sentiment feature we decide to create a Persian sentiment dictionary which consists of four sentiment categories. We also provide a new Persian proverb dictionary using in the preparation step to enhance the accuracy of the proposed model. The performance of our model is analyzed using several standard machine learning algorithms. The results of the experiment show our method outperforms the baseline method and reaches an accuracy of 80.82%. We also study the importance of each set of proposed features and evaluate its added value to the classification.
Keywords: Sarcasm Detection, Natural Language Processing, Machine Learning, Sentiment Analysis, Classification.

1Norliza Katuk, 1Ku Ruhana Ku-Mahamud, 1Nur Haryani Zakaria & 1,2Ayad Mohammed Jabbar
1School of Computing, Universiti Utara Malaysia, Malaysia
2College of Arts and Sciences, Shatt Al-Arab University, Basra, Iraq
{k.norliza, ruhana, haryani},
Citations have been an acceptable journal performance metric which has been used by many indexing databases for inclusion and discontinuation of journals in their list. Therefore, editorial teams must maintain their journal performance by increasing the article citations for continuous content indexing in the databases. With this aim in hand, this study intends to assist the editorial team of the Journal of Information and Communication Technology (JICT) in increasing the performance and impact of the journal. Currently, the journal suffered from low citations count, which may jeopardise its sustainability. Past studies in library science suggested a positive correlation between keywords and citations. Thus, keyword and topic analyses could be a solution to address the issue of the journal citation. This article describes a scientometric analysis of emerging topics in general computer science, the Scopus subject area for which JICT is indexed. This study extracted bibliometric data of the top 10% journals in the subject area to create a dataset of 5546 articles. The results of the study suggested ten emerging topics in computer science that can be considered by the journal editorial team in selecting the article and a list of highly-used keywords in articles published in 2019 and 2020 (as of 15 April 2020). The outcome of this study might be considered by JICT editorial team other journals in general computer science that suffers from a similar issue.
Keywords: Scientometrics, Scientometric Analysis, Bibliometrics, Citation Analysis, Research Trends.

1Hock Hung Chieng, 1Noorhaniza Wahid & 2Pauline Ong
1Faculty of Information Technology and Computer Science, Universiti Tun Hussein Onn Malaysia, Malaysia
2Faculty of Mechanical and Manufacturing Engineering, Universiti Tun Hussein Onn Malaysia, Malaysia;;
Activation function is a key component in deep learning that performs non-linear mappings between the inputs and outputs. Rectified Linear Unit (ReLU) has been the most popular activation function across the deep learning community. However, ReLU contains several shortcomings that can result in inefficient training of the deep neural networks, these are: 1) the negative cancellation property of ReLU tends to treat negative inputs as unimportant information for the learning, hence resulting in a performance degradation; 2) the inherent predefined nature of ReLU is unlikely to promote additional flexibility, expressivity, and robustness to the networks; 3) the mean activation of ReLU is highly positive and leading to bias shift effect in network layers; 4) the multilinear structure of ReLU restricts the non-linear approximation power of the networks. To tackle these shortcomings, this paper introduces a Parametric Flatten-T Swish (PFTS) as an alternative to ReLU. By taking ReLU as a baseline method, the experiments shown PFTS improved classification accuracy on SVHN dataset by 0.31%, 0.98%, 2.16%, 17.72%, 1.35%, 0.97%, 39.99% and 71.83% on DNN-3A, DNN-3B, DNN-4, DNN-5A, DNN-5B, DNN-5C, DNN-6 and DNN-7, respectively. Besides, PFTS also achieved the highest mean rank among the comparison methods. The proposed PFTS manifests higher non-linear approximation power during training and thereby improve the predictive performance of the networks.
Keywords: activation function, deep learning, Flatten-T Swish, non-linearity, ReLU.

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.