ICCSE 2018THE THIRD INTERNATIONAL

CONFERENCE ON COMPUTATIONALSCIENCE AND ENGINEERING

BOOK OF ABSTRACTSOrganized By

Knowledge Technology Research Unit,Universiti Malaysia Sabah, Malaysia, and

Borneo Wildlife Alliance, Kota Kinabalu, Sabah, MalaysiaSupported by

Sabah Tourism Board, HILTI Corporation and Scientific &Technical Products Sdn Bhd

ORGANIZING COMMITTEEGeneral Chair

Assoc. Professor Dr. Rayner Alfred (Universiti Malaysia Sabah, Malaysia)

Conference Co-ChairsAssoc. Professor Ag. Asri Ag. Ibrahim (Universiti Malaysia Sabah, Malaysia)

Dr. Joe Henry Obit (Universiti Malaysia Sabah, Malaysia)Dr. Raymond Alfred (Borneo Wildlife Alliance)

Local Arrangements ChairsBen Romero Didimus (Borneo Conservation Trust, Malaysia)

Treasury ChairAssoc. Professor Dr. Stephen Laison Sondoh Jr (Universiti Malaysia Sabah, Malaysia)

Publication ChairDr. Ashish Doshi (University of South Australia, Australia)

Publicity ChairAssoc. Professor Dr. I-Hsien Ting (National University of Kaohsiung, Taiwan)

Programme Committee MembersEmeritus Prof. Dr. Lorna Uden (Staffordshire University, United Kingdom)

Emeritus Prof. Dr. Ian Witten (University of Waikato, New Zealand)Professor Dr. Ali Chekima (Universiti Malaysia Sabah, Malaysia)

Professor Dr. Amir A. Shafie (International Islamic University Malaysia)Professor Dr. Azuraliza Abu Bakar (Universiti Kebangsaan Malaysia, Malaysia)

Professor Dr. Daniela Zaharie (West University of Timisoara, Romania)Professor Dr. Dana Petcu (West University of Timisoara, Romania)

Professor Dr. Limsoon wong (National University of Singapore, Singapore)Professor Dr. Wai Kiang (Albert) Yeap (Auckland University of Technology, New Zealand)

Professor Dr. Jooyoung Lee (Korea Institute for Advanced Study, South Korea)Professor Dr. Fatimah Dato Ahmad (National Defence University of Malaysia)

Professor Dr. Gabriela Czibula (Babes-Bolyai University, Romania)Professor Dr. Goutam Chakraborty (Iwate Prefectural University, Japan)

Professor Dr. Hiroyuki Iida (Japan Institute of Science and Technology, Japan)Professor Dr. Jharna Majumdar (Nitte Meenakshi Institute of Technology, India)Professor Dr. Leon Shyue-Liang Wang (National University of Kaohsiung, Taiwan)

Professor Dr. Shobha B.N (S.J.C. Institute of Technology, India)Professor Dr. Masri Ayob (Universiti Kebangsaan Malaysia, Malaysia)

ii

Professor Dr. Amir A. Shafie (International Islamic University Malaysia)Professor Dr. Fatimah Dato Ahmad (Universiti Pertahanan Nasional Malaysia)

Professor Dr. Norazah Mohd Suki (Universiti Utara Malaysia, Malaysia)Professor Dr. Yong Jin Park (Universiti Utara Malaysia, Malaysia)

Assoc. Professor Dr. Do Phuc (University of Information Technology, Vietnam)Assoc. Professor Dr. Edmund Ng Giap Weng (Universiti Malaysia Sabah, Malaysia)

Assoc. Professor Dr. Farrah Wong (Universiti Malaysia Sabah, Malaysia)Assoc. Professor Dr. I-Hsien Ting (National University of Kaohsiung, Taiwan)Assoc. Professor Dr. Jane Labadin (Universiti Malaysia Sarawak, Malaysia)Assoc. Professor Dr. Jason Teo Tze Wi (Universiti Malaysia Sabah, Malaysia)

Assoc. Professor Dr. Kashif Nisar (Universiti Malaysia Sabah, Malaysia)Assoc. Professor Dr. Mohd Hanafi Ahmad Hizaji (Universiti Malaysia Sabah, Malaysia)Assoc. Professor Dr. Nhuyen Minh Le (Japan Institute of Science and Technology, Japan)

Assoc. Professor Dr. Norbayah Mohd Suki (Universiti Utara Malaysia, Malaysia)Assoc. Professor Dr. Ryo Maezono (Japan Institute of Science and Technology, Japan)Assoc. Professor Dr. Stephane Bressan (National University of Singapore, Singapore)

Assoc. Professor Dr. Syed Nasirin Bin Syed Zainol (Universiti Malaysia Sabah, Malaysia)Assoc. Professor Dr. Yuto Lim (Japan Institute of Science and Technology, Japan)

Dr. Ali Farzamnia (Universiti Malaysia Sabah, Malaysia)Dr. Aslina Baharum (Universiti Malaysia Sabah, Malaysia)

Dr. Carolyn Bt. Salimun @ Jackson (Universiti Malaysia Sabah, Malaysia)Dr. Cathy Hsing-Tzu Lin (National University of Kaohsiung, Taiwan)

Dr. Chin Kim On (Universiti Malaysia Sabah, Malaysia)Dr. Chin Pei Yee (Universiti Malaysia Sabah, Malaysia)

Dr. Chung Seng Kheau (Universiti Malaysia Sabah, Malaysia)Dr. Choo Yun Huoy (Technical University of Malaysia Malacca)Dr. Cristian Mihaescu (Universitatea din Craiova, Romania)Dr. Goh Say Leng (Universiti Malaysia Sabah, Malaysia)

Dr. Haviluddin Haviluddin (Mulawarman University, Indonesia)Dr. Jane Lai Po Hung (Universiti Malaysia Sabah, Malaysia)Dr. Joe Henry Obit (Universiti Malaysia Sabah, Malaysia)Dr. Khalifa Chekima (Universiti Malaysia Sabah, Malaysia)

Dr. Khairul Anuar Mohamad (Universiti Malaysia Sabah, Malaysia)Dr. Lau Hui Keng (Universiti Malaysia Sabah, Malaysia)Dr. Leau Yu Beng (Universiti Malaysia Sabah, Malaysia)

Dr. Liawas Barukang (Universiti Malaysia Sabah, Malaysia)Dr. Muralindran Mariappan (Universiti Malaysia Sabah, Malaysia)

Dr. Muzaffar Hamzah (Universiti Malaysia Sabah, Malaysia)Dr. Norazlina Khamis (Universiti Malaysia Sabah, Malaysia)

Dr. Patricia Anthony (Lincoln University New Zealand, New Zealand)Dr. Puteri Nor Ellyza Nohuddin (Universiti Kebangsaan Malaysia)

iii

Dr. Renee Chin Ka Yin (Universiti Malaysia Sabah, Malaysia)Dr. Samry Mohd. Shamrie bin Sainin (Universiti Utara Malaysia, Malaysia)

Dr. Suresh Manandhar (University of York, United Kingdom)Dr. Suaini Binti Sura (Universiti Malaysia Sabah, Malaysia)

Dr. Suraya Alias (Universiti Malaysia Sabah, Malaysia)Dr. Tan Soo Fun Universiti Malaysia Sabah, Malaysia)

Dr. Zamhar Ismail Iswandono (Universiti Malaysia Sabah, Malaysia)Ir. Othman Bin Ahmad (Universiti Malaysia Sabah, Malaysia)

Organizing Committee Members

Prof. Park Yong-JinA/Professor Dr. Edmund Ng Giap Weng

A/Professor Dr. Mohd Hanafi Ahmad HijaziDr. Aslina BaharumDr. Chin Kim On

Dr. Christie Pei YeeDr. Jane Lai

Dr. Joe Henry ObitDr. Khalifa ChekimaDr. Lau HuiKeng

Dr. Norazlina Khamis Dr. Tan Soo FunDr. Zamhar Iswandono Ismail Asni Tahir

Albert Alex JimmyAzman Bin Ismail

Diana Binti KimmuanFadzli Bin Bahal

Ida Haslinda Bt MuljanaIza Azura Ahmad BaharLeopold Madies Faliz

Nordin SaadNorhayati Daut

Saidin Bin AhmadShaliza Hayati A. Wahab

Siti Hasnah TanalolWinnie Edward Alau

Zaidatol Haslinda Bt Abdullah Sani

iv

ABOUT THE CONFERENCE

Computational Science and Engineering (CSE) is a relatively new paradigm for scientific researchand engineering design in which large-scale simulation, data analysis, and high performancecomputing play a central role. In fact, the applications of CSE can be seen in almost alldisciplines.

Engineering and science problems have been solved historically using experimental testingand/or mathematical analysis. Some examples of engineering problems are fluid flows andstructural properties associated with aircraft, ships, submarines, automobiles, spacecraft, jet androcket propulsion engines, buildings and other structures. Other examples relate to electricalpower generation, weather, rivers and oceans, electrical equipment, computer hardware, radar,antennas, chemical reactions and processes, fuel cells, petroleum recovery and refining, agriculturaland construction equipment, refrigeration and air conditioning, air and water pollution, energyconversion and storage, and many others. Many of these problems can now be solved efficientlyas computational simulations of mathematical models that represent the relevant physicalphenomena arising in each problem.

The Program Committees are looking for original research contributions on a broad-range oftopics related to High performance computing, Modeling and simulation, Algorithms, Big DataAnalysis and visualization, Data Science, CSE Education, Advanced Networking and Applicationsand Intelligent and Bio-Inspired Computing. Topics of interest include, but are not limited tothe following Research and Application Tracks:

• High performance computing

• Modeling and simulation

• Numerical and Discrete Algorithms

• Big Data Analysis and visualization

• Data Science

• CSE Education and Finance

• Advanced Networking and Applications

• Intelligent and Bio-inspired Computing

• Applications of CSE

We hope the participants will find this conference interesting, useful, and informative, and itwill give them a valuable inspiration for original and innovative research.

v

MESSAGE FROM THE CONFERENCECHAIRS

A/Prof. Dr. Rayner A/Prof. Dr. Ag Asri Dr. Joe Henry Obit Dr. Raymond Alfred

Conference Chairs

A/Prof. Dr. Rayner Alfred (Universiti Malaysia Sabah, Malaysia)A/Prof. Dr. Ag. Asri Ag. Ibrahim (Universiti Malaysia Sabah, Malaysia)Dr. Joe Henry Obit (Universiti Malaysia Sabah, Malaysia)Dr. Raymond Alfred (Borneo Wildlife Alliance)

On behalf of the conference program committee, we would like to extend a warm welcome toall honorable guests, keynote speakers, session chairs and participants of the Third InternationalConference on Computational Science and Engineering 2018 (ICCSE2018)!

Computational Science and Engineering (CSE) is a relatively new paradigm for scientificresearch and engineering design in which large-scale simulation, data analysis, and high perfor-mance computing play a central role. In fact, the applications of CSE can be seen in almost alldisciplines.

The Third International Conference on Computational Science and Engineering 2018 (ICCSE2018)will provide a unique forum to exchange innovative research ideas, recent results, and share experi-ences among researchers and practitioners in a broad-range of topics related to High performancecomputing, Modeling and simulation, Algorithms, Big Data Analysis and visualization, DataScience, CSE Education, Advanced Networking and Applications and Intelligent and Bio-InspiredComputing.

In this conference, we have three distinguished keynote speakers, Professor Chong Nak-Youngfrom the Japan Advanced Institute of Science and Technology (JAIST), Professor Kenji Suzukifrom World Research Hub Initiative (WRHI) & Laboratory for Future Interdisciplinary Researchof Science and Technology (FIRST), Institute of Innovative Research (IIR), Tokyo Institute ofTechnology and finally Mr Chai Sung Choong, who will share their knowledge in computationalscience and technology. We hope that the keynote sessions and the parallel sessions add value toyour knowledge and research career.

vi

We would also like to express our deep gratitude and appreciation to all the program com-mittee members, panel reviewers, organizing committees and volunteers for your efforts to makethis conference a successful event. We sincerely hope that ICCSE2018 will provide a venuefor knowledge sharing and establishing more research collaborations among us. Last but notleast, We wish everyone an enjoyable and memorable stay in Kota Kinabalu for the ICCSE2018Conference. Thank you.

vii

KEYNOTE SPEAKER 1

Keynote Speaker : Professor Chong Nak-Young

Affiliation : Professor of Computer Science, School of Information Science, Japan AdvancedInstitute of Science and Technology (JAIST)

Bibliography : Nak Young Chong received the B.S., M.S., and Ph.D. degrees in mechan-ical engineering from Hanyang University, Seoul, Korea, in 1987, 1989, and 1994, respectively.From 1994 to 2007, he was a member of research staff at Daewoo Heavy Industries and KIST inKorea, and MEL and AIST in Japan. In 2003, he joined the faculty of Japan Advanced Instituteof Science and Technology, where he currently is a Professor of Information Science. He alsoserved as Vice Dean for Research and Director of the Center for Intelligent Robotics at JAIST.He was a Visiting Scholar at Northwestern University, Georgia Institute of Technology, Universityof Genoa, and Carnegie Mellon University, and also served as an Associate Graduate Facultyat the University of Nevada, Las Vegas, International Scholar at Kyung Hee University, andDistinguished Invited Research Professor at Hanyang University. He serves as Senior Editor ofthe IEEE Robotics and Automation Letters, Journal of Intelligent Service Robotics, InternationalJournal of Advanced Robotic Systems, and servesserved as Senior Editor of UR 2018, IEEEICRA, IEEE Ro-Man, IEEE CASE Conference Editorial Boards, and Associate Editor of theIEEE Transactions on Robotics. He served as Program ChairCo-Chair for JCK Robotics 2009,ICAM 2010, IEEE Ro-Man 2011, IEEE CASE 2012, IEEE Ro-Man 2013, URAI 2012/2013,and DARS 2014. He was a General Co-Chair of URAI 2017. He also served as Co-Chair forIEEE-RAS Networked Robots Technical Committee from 2004 to 2006, and Fujitsu ScientificSystem Working Group from 2004 to 2008.

Specialization : Human-Friendly Robots, Teleoperation, Intelligent Mechanical Systems, Knowl-

viii

edge Networking Robot Control.

Title : Culture-Aware Elderly Care Robots in a Smart ICT Environment

Abstract : Rapid demographic change constitutes an unprecedented societal challenge forJapan. I will shed light on the issues of Japan’s super-aging society and introduce the human-robot interaction work package of our ongoing EC Horizon 2020 project “CARESSES”, aimingat developing culturally competent elderly care robots, jointly commissioned by the Ministry ofInternal Affairs and Communications of Japan. We envision a future where care robots are ableto interact with the elderly with different cultural and personality traits through personalizedemotion generation and facial/vocal/body expression. I will share some of our preliminary resultsof multi-modal human-robot interaction and explore opportunities for future collaboration withuniversities and research institutes in Malaysia. Furthermore, I will introduce a smart ICTenvironment testbed iHouse, and a user speech activated interface to enable care robots to gaincontrol over the iHouse devices and functions and provide data through verbal interaction withthe user. I am hoping to discuss the technical feasibility of a robotic smart care home interfacetoward supporting independent living of the elderly.

ix

KEYNOTE SPEAKER 2

Keynote Speaker : Professor Kenji Suzuki

Affiliation : Professor of Computer Science, World Research Hub Initiative (WRHI) &Laboratory for Future Interdisciplinary Research of Science and Technology (FIRST), Instituteof Innovative Research (IIR), Tokyo Institute of Technology

Bibliography : Kenji Suzuki, Ph.D. (by Published Work; Nagoya University) worked atHitachi Medical Corp., Japan, Aichi Prefectural University, Japan, as a faculty member, andin Department of Radiology, University of Chicago, as Assistant Professor. In 2014, he joinedDepartment of Electric and Computer Engineering and Medical Imaging Research Center, IllinoisInstitute of Technology, as Associate Professor (Tenured). Since 2017, he has been jointly ap-pointed in World Research Hub Initiative, Tokyo Institute of Technology, Japan, as Full Professor.He published more than 320 papers (including 110 peer-reviewed journal papers). He has beenactively studying deep learning in medical imaging and computer-aided diagnosis in the past 20years. He is inventor on 30 patents (including ones of earliest deep-learning patents), which werelicensed to several companies and commercialized. He published 10 books and 22 book chapters,and edited 13 journal special issues. He was awarded more than 25 grants as PI including NIHR01 and ACS. He served as the Editor of a number of leading international journals, includingPattern Recognition and Medical Physics. He served as a referee for 83 international journals, anorganizer of 35 international conferences, and a program committee member of 150 internationalconferences. He received 26 awards, including Springer-Nature EANM Most Cited Journal PaperAward 2016 and 2017 Albert Nelson Marquis Lifetime Achievement Award.

Specialization : Machine learning, deep learning, computer-aided diagnosis, medical imaging,artificial intelligence.

x

Title : Deep Learning in Medical Image Processing and Diagnosis

Abstract : Machine leaning (ML) in artificial intelligence has become one of the mostactive areas of research in the biomedical imaging field including medical image analysis andcomputer-aided diagnosis (CAD), because “learning from examples or data” is crucial to handlinga large amount of data (“Big data”) coming from medical imaging informatics systems. Recently,as the available computational power increased dramatically, image-based ML or “deep learn-ing” emerged. Deep learning, including our original massive-training artificial neural networks(MTANNs) and the most popular convolutional neural networks (CNNs), is an end-to-end MLmodel that enables a direct mapping from the raw input data to the desired outputs, eliminatingthe need for handcrafted features in feature-based ML. Deep learning (or image-based ML) is aversatile, powerful framework that can acquire image-processing and analysis functions throughtraining with image examples. In this talk, deep learning in medical imaging is overviewed tomake clear a) what has changed in machine learning after the introduction of deep learning,b) differences and advantages over conventional feature-based ML, and c) its applications to 1)separation of bones from soft tissue in chest radiographs, 2) CAD for lung nodule detection inchest radiography and thoracic CT, 3) distinction between benign and malignant nodules in CT,4) polyp detection and classification in CT colonography, and 5) radiation dose reduction in CTand mammography.

xi

KEYNOTE SPEAKER 3

Keynote Speaker : Chai Sung Choong

Affiliation : Transformative Executive at the Infomina Sdn Bhd, Kuala Lumpur, Malaysia

Bibliography : Mr. Chai Sung Choong graduated from Western Australia of TechnologyAustralia in Information Processing. He has a working experince of 32 years in convergenceICT. His consulting strength include financial process, solution design and architecture and ICTtrend and direction. He started his career as a Technical Specialist in 1982-1998. Then, hewas appointed as the Country Manager for BMC Malaysia in 1999-2000, and then became theRegional Executive at the ASEAN, IBM Singapore in 2001-2006. After that he was promotedto lead the Infrastructure Technical Services when he was appointed as the Country Managerfor IBM Malaysia for 2010-2012. In 2012-2013 he lead the Global Technical System at IBMMalaysia, when he was appointed as Business Development Executive. In 2016-2018, he thenmoved to FSI, Microlink Systems Bhd where he was appointed as the Solution Director wherehe lead a solution team to win a major financial solution of a public sector organization. Heis currently the transformative executive at the Infomina Sdn Bhd in which he leads projectsrelated to transforming major public entity traditional system into web services.

Specialization : Financial Process, Solution Design and Architecture and ICT Trend andDirection.

Title : How Industry 4.0 and Information System impacting Finance industry?

Abstract : Next generation technologies such as Robotics Process Automation (RPA), artificialintelligence (AI), and the internet of things (IoT) is driving the next industrial revolution – alsoreferred to as Industry 4.0. TECHNOLOGY has been changing how businesses work and itsno different for the finance industry. To reap the benefits that technology can bring, financialinstitutions would have to understand not only the advantages of implementing a particular

xii

technology but also how it will impact their businesses. The modern finance institutions mustbridge the information gap between internal and external stakeholders, helping to rebuild trustbetween institutions and the public. To play this role, they need to do more than just respondto technological changes, but also to embrace the rapid development of digital technologies thatcharacterizes Industry 4.0. With these new technologies, Finance institutions no longer needs tofocus on the past. Instead, they focus on the current quarter and provide intelligent projectionsfor the future. The arrival of blockchain or distributed ledger technology including the technicalinnovation called cyber-currencies has made possible for peer-to-peer (P2P) transactions. Inarchitectural terms, this is a bottom-up system which technically can avoid any official oversightor addressing the public mistrust of the official system.

xiii

CONFERENCE PROGRAM

29 August 2018

07:30 # Registration08:00 # Arrival of VIPs08:15 # Arrival of Datuk Seri Panglima Haji Mohd. Shafie bin Haji Apdal

(Chief Minister, Sabah)08:30 # Opening Ceremony (Ballroom 2)

# Opening Remark: Associate Professor Dr. Rayner Alfred(Conference Chair, Head of Knowledge Technology Research Unit, UMS)

# Welcoming Remark: Associate Professor Dr. Ag. Asri Ag. Ibrahim(Dean, Faculty of Computing and Informatics, Universiti Malaysia Sabah)

# Welcoming Remark: Professor Datuk Dr. D Kamarudin D Mudin(Vice Chancellor, Universiti Malaysia Sabah)

# Officiating Remark: Datuk Seri Panglima Haji Mohd. Shafie bin Haji Apdal(Chief Minister, Sabah)

# Photography Session# VIP Coffee Break and Press Conference

09:15 # Keynote Speech (Ballroom 2)# Professor Chong Nak-Young (JAIST)# Title: Culture-Aware Elderly Care Robots in a Smart ICT Environment

10:00 # Keynote Speech (Ballroom 2)# Chai Sung Choong (Infomina Sdn Bhd)# Title: How Industry 4.0 and Information System impacting Finance industry?

10:45 # Coffee Break11:00 # Parallel Session 1 (Event 2)13:00 # Lunch Break (Vista 1)14:00 # Parallel Session 2 (Event 2)15:45 # Coffee Break16:00 # Parallel Session 3 (Event 2)18:00 # Adjourn18:30 # Conference Dinner (Vista 1, Hilton Kota Kinabalu)

xiv

30 August 2018

08:00 # Registration (Cont)09:00 # Keynote Speech (Ballroom 2)

# Professor Kenji Suzuki (Tokyo Institute of Technology)# Title: Deep Learning in Medical Image Processing and Diagnosis

09:40 # Coffee Break10:00 # Session 4 (Ballroom 2)13:00 # Lunch14:00 # Session 5 (Ballroom 2)15:00 # Coffee Break15:30 # Session 6 (Ballroom 2)17:00 # Lucky Draw and Closing Ceremony (Ballroom 2)

xv

PARALLEL SESSIONS

ALL SESSIONSSESSION (29/08) CATEGORY TIME VENUE

1 Data Science [11:00 - 12:45] Event 22 Data Science [14:00 - 15:45] Event 23 Modeling / CSE Education [16:00 - 18:00] Event 2

SESSION (30/08)4 Numerical and Discrete Algorithms [10:00 - 12:30] Ballroom 25 Application of CSE [14:00 - 15:00] Ballroom 26 Advanced Networking and Applications [15:30 - 17:45] Ballroom 2

xvi

SESSION 1 : Data Science (Event 2)

(CHAIRPERSON: Dr. Mohd Fahmi Mohamad Amran)Time PID Paper Title Presenter

11:00 - 11:15 14

Using PERT to Optimise theCompletion Time of Air- Conditioningand Mechanical Ventilation (ACMV)Project

Chuan Kian Pang

11:15 - 11:30 10 Animal Detection in Highly ClutteredNatural Scenes by Using Faster R-CNN Wenjun Yu

11:30 - 11:45 11A Literature Review on MachineLearning Techniques Used for TrainingAnnotated Corpus

Dr. HalizahBasiron

11:45 - 12:00 31Pushing Visualization Effects intoPushed Schema EnumeratedTree-Based Support Constraints

Dr. Doreen YingYing Sim

12:00 - 12:15 38A Robust Geometric Skin Colour FaceDetection Method under UnconstrainedEnvironment of Smartphone Database

Assistant Prof.Fatima Khalid

12:15 - 12:30 40

Information Extraction fromMultifaceted Unstructured Big data –A Short Review in IT, Engineering andSciences

Dr. Rehan Akbar

12:30 - 12:45 42 CSMiner-TBM:Tree-Based MiningContrast Subspace Florence Sia

12:45 - 13:00 44Improved BVBUC Algorithm toDiscover Closed Itemsets in LongBiological Datasets

Dr. Nurul FarizaZulkurnain

xvii

SESSION 2 : Data Science (Event 2)

(CHAIRPERSON: Dr. Halizah Basiron)Time PID Paper Title Presenter

14:00 - 14:15 9Testing Design for Pixel ValueGraphical Password Scheme on ImageSpecimens

Dr. Mohd FahmiMohamad Amran

14:15 - 14:30 30 A Bicolano-to-Tagalog Transfer-BasedMachine Translation System

Assoc. Prof. Dr.Ria AmbrocioSagum

14:30 - 14:45 49 An Intelligent Crisis-mappingFramework for Flood PredictionDr. Siti AzirahBinti Asmai

14:45 - 15:00 50 Tree-Based Method for MiningContrast Subspace in Categorical Data Florence Sia

15:00 - 15:15 51A Review on Visual Water QualityMonitoring System in PrecisionAquaculture

Alter Jimat Embug

15:15 - 15:30 52

Designing A Model Based on WeightedRange Fingerprint Method ForTriangulation Analytics UsingCrowdsourcing Data

Aaron FranklinSoon

SESSION 3 : Modeling and Simulation / CSE Education and Finance (Event 2)

(CHAIRPERSON: A/Professor Dr. Rodziah Binti Atan)Time PID Paper Title Presenter

16:00 - 16:15 3 An Analysis of Gamification on Dota2’s Business ModelSagguneswaraanThavamuni

16:15 - 16:30 16Modeling a Novel Uni-MOSFET SolidState Relay for a Single Phase SystemCompensation and Load Control

Ir. KrishnanShashikumar

16:30 - 16:45 7 Designing Emotion Interface inE-Learning Application for Children Magrizef Gasah

16:45 - 17:00 17 Planning Approach For BlendedLearning: A Literature Review Pei Yee Chong

17:00 - 17:15 21 The Effects of System Quality onSocial Commerce Nona M Nistah

17:15 - 17:30 36 Estimation of CO2 Emission from theFerries in the Urban AreaDr. Chien-ChangChou

xviii

SESSION 4 : Numerical and Discrete Algorithms (Ballroom 2)

(CHAIRPERSON: Dr. Doreen Ying Ying Sim)Time PID Paper Title Presenter

10:00 - 10:15 4 A New Extrapolation Used for IterativeMethodsUthaiPrasopchingchana

10:20 - 10:35 26Formal Verification of Digital CircuitsUsing Simulator with MathematicalFoundation

Dr. Wilayat Khan

10:40 - 10:55 35 A Framework for Halal KnowledgeMetadata RepresentationsA/Professor Dr.Rodziah Binti Atan

11:00 - 11:15 37

Needleman-Wunsch AlgorithmComparison Using CPUImplementation and GPUImplementation

Yun Sup Lee

11:20 - 11:35 41

Static Profiling of Assembly CodePerformance and OptimizationEffectiveness Using InstructionsPerformed and Program Latency

Roger Luis Uy

11:40 - 11:55 45

Utilizing AVX/AVX2 Instructions toImplement Hyyrö’s Bit-vectorAlgorithm for DNA SequenceAlignment

Kyle MatthewChua

12:00 - 12:15 48 Issues in Lightweight EncryptionAlgorithm For mHealthDr. Azni HaslizanAb Halim

12:20 - 12:35 13Numerical Investigation of BootstrapArtificial Neural Network (p,d,q) Modelfor Malaysia Exchange Rate

Dr. Nurul HilaZainuddin

xix

SESSION 5 : Application of CSE (Ballroom 2)

(CHAIRPERSON: Dr. Aslina Baharum)

14:00 - 14:15 2Fitting Statistical Distribution of DailyTorrential Rainfall in PeninsularMalaysia for the Purpose of Simulation

Dr. ShazlynMilleanaShaharudin

14:15 - 14:30 8Temporal Frequency Spread of PlaneWave Propagation Through Moderateto Strong Turbulence

Bing Guan

14:30 - 14:45 18 Music Recommendation on CommonArtist Basis for Musical Applications Chethan Sharma

14:45 - 15:00 20 Flow Induced Vibration (FIV) Analysisof Oil and Gas Process Piping SystemMohd Fazril IrfanAhmad Fuad

15:00 - 15:15 24Performance Evaluation of FractionalOrder PID and Sliding Mode Controlwith Optimization Tuning Approach

Dr. RozaimiGhazali

15:15 - 15:30 25Development of Novel Gamified OnlineElectrocardiogram Learning Platform(GaMED ECG@TM )

Dr. May HoneyOhn

15:30 - 15:45 27

Molecular Structures, RotationalBarriers, and Density FunctionalTheory studies on 3,5-Diamino-6-(2,3-Dichlorophenyl)-1,2,4-Triazine

Dr. Pek Lan Toh

SESSION 6 : Advanced Networking and Applications (Ballroom 2)

(CHAIRPERSON: Dr. Rozaimi Ghazali)Time PID Paper Title Presenter

16:15 - 16:30 12 Evaluation of 5G Key Technique:Non-Orthogonal Multiple AccessProf. Dr. JaehoChoi

16:30 - 16:45 34

Cloud-based Education as a Service(CEaaS) System RequirementsSpecification Model of HigherEducation Institutions in IndustrialRevolution 4.0

Prof. Dr. RusliHaji Abdullah

16:45 - 17:00 46

A High Efficiency GyrostabilizerAntenna Platform for Real-time UAVSynthetic Aperture Radar (SAR)Motion Error Compensation

Ir Zahari AwangAhmad

17:00 - 17:15 47A Wireless Sensor Network (WSN)Implementation using Low-power WideArea Network (LPWAN)

Renann Baldovino

xx

Contents

Organizing Committee ii

About the Conference v

Message from the Conference Chair vi

Keynote Speakers viiiProfessor Chong Nak-Young (Japan Advanced Institute of Science and Technology

(JAIST)) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viiiProfessor Kenji Suzuki (Tokyo Institute of Technology) . . . . . . . . . . . . . . . . . xChai Sung Choong () . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

Conference Program xiv

Parallel Sessions xvi

Contents xxi

SESSION 1 : DATA SCIENCE 1

1 Using PERT to Optimise the Completion Time of Air- Conditioning andMechanical Ventilation (ACMV) Project 2Chuan-Kian Pang, Noor-Ajian Mohd-Lair and Yi-Sheng Chua . . . . . . . . . . . . . 2

2 Animal Detection in Highly Cluttered Natural Scenes by Using Faster R-CNN 3Wenjun Yu, Sumi Kim, Jeong-Hyu Lee, Jaeho Choi . . . . . . . . . . . . . . . . . . . 3

3 A Literature Review on Machine Learning Techniques Used for Training An-notated Corpus 4Fitrah Rumaisa, Halizah Basiron and Zurina Saaya . . . . . . . . . . . . . . . . . . . 4

4 Pushing Visualization Effects into Pushed Schema Enumerated Tree-BasedSupport Constraints 5SIM Doreen Ying Ying, TEH Chee Siong and ISMAIL Ahmad Izuanuddin . . . . . . . 5

5 A Robust Geometric Skin Colour Face Detection Method under UnconstrainedEnvironment of Smartphone Database 6Noor Amjed, Fatimah Khalid, Rahmita Wirza O.K. Rahmat, Hizmawati Bint Madzin . 6

xxi

CONTENTS CONTENTS

6 Information Extraction from Multifaceted Unstructured Big data – A ShortReview in IT, Engineering and Sciences 7Kiran Adnan, Rehan Akbar and Khor Siak Wang . . . . . . . . . . . . . . . . . . . . . 7

7 CSMiner-TBM:Tree-Based Mining Contrast Subspace 8Florence Sia and Rayner Alfred . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

8 Improved BVBUC Algorithm to Discover Closed Itemsets in Long BiologicalDatasets 9Fatimah Audah Md Zaki and Nurul Fariza Zulkurnain . . . . . . . . . . . . . . . . . . 9

SESSION 2 : DATA SCIENCE 10

9 Testing Design for Pixel Value Graphical Password Scheme on Image Speci-mens 11Mohd Fahmi Mohamad Amran, Mohd Afizi Mohd Shukran, Mohd Sidek Fadhil Mohd

Yunus, Omar Zakaria, Nurhafizah Moziyana Mohd Yusop, Mohd Nazri Ismail,Mohd Rizal Mohd Isa, Mohammad Adib Khairuddin, Kamaruzaman Maskat,Yamunah Kathiravan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

10 A Bicolano-to-Tagalog Transfer-Based Machine Translation System 12Ria Ambrocio Sagum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

11 An Intelligent Crisis-mapping Framework for Flood Prediction 13Siti Azirah Asmai, Zaheera Zainal Abidin, Halizah Basiron, Sabrina Ahmad, Muhammad

Ameer Faisal Adanan and Nor Azim Abdullah . . . . . . . . . . . . . . . . . . . . 13

12 Tree-Based Method for Mining Contrast Subspace in Categorical Data 14Florence Sia and Rayner Alfred . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

13 A Review on Visual Water Quality Monitoring System in Precision Aquacul-ture 15Alter Jimat Embug, Ag Asri Ag Ibrahim, Muzaffar Hamzah, and Mohammad Fadhli Asli 15

14 Designing A Model Based on Weighted Range Fingerprint Method For Tri-angulation Analytics Using Crowdsourcing Data 16Aaron Franklin Soon, Siti Nurulain Mohd Rum, Hamidah Ibrahim, Rohaya Latip ,Razali

Yaakob and Lilly Suriani Affendey . . . . . . . . . . . . . . . . . . . . . . . . . . 16

SESSION 3 : MODELING AND SIMULATION / CSE EDUCATION ANDFINANCE 17

15 An Analysis of Gamification on Dota 2’s Business Model 18Sagguneswaraan Thavamuni, Hadzariah Ismail and Hiroyuki Iida . . . . . . . . . . . . 18

xxii

CONTENTS CONTENTS

16 Modeling a Novel Uni-MOSFET Solid State Relay for a Single Phase SystemCompensation and Load Control 19K. Shashikumar, C. Venkataseshaiah, K.S. Sim . . . . . . . . . . . . . . . . . . . . . . 19

17 Designing Emotion Interface in E-Learning Application for Children 20Magrizef Gasah, Aslina Baharum, Nurul Hidayah May Zain . . . . . . . . . . . . . . . 20

18 Planning Approach For Blended Learning: A Literature Review 21Chong Pei Yee, Zamhar Iswandono bin Awang Ismail and Esmadi Abu Abu Seman . . 21

19 The Effects of System Quality on Social Commerce 22Nona M Nistah, Suaini Sura and Ook Lee . . . . . . . . . . . . . . . . . . . . . . . . . 22

20 Estimation of CO2 Emission from the Ferries in the Urban Area 23Chien-Chang Chou, Tsun-Lirng Yang, Hsien-Pin Hsu . . . . . . . . . . . . . . . . . . 23

SESSION 4 : NUMERICAL AND DISCRETE ALGORITHMS 24

21 A New Extrapolation Used for Iterative Methods 25Uthai Prasopchingchana and Tul Manewattana . . . . . . . . . . . . . . . . . . . . . . 25

22 Formal Verification of Digital Circuits Using Simulator with MathematicalFoundation 26Wilayat Khan, Basim Azam, Noman Shahid, Abdul Moeed Khan, Ahtisham Shaheen . 26

23 A Framework for Halal Knowledge Metadata Representations 27Rodziah Atan and Nur Adila Azram . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

24 Needleman-Wunsch Algorithm Comparison Using CPU Implementation andGPU Implementation 28Roger Luis Uy, Yun Sup Lee, Yu Sin Kim . . . . . . . . . . . . . . . . . . . . . . . . . 28

25 Static Profiling of Assembly Code Performance and Optimization Effective-ness Using Instructions Performed and Program Latency 29Chudrack Shalym Y. Salinas, Jonathan Paul C. Cempron and Roger Luis Uy . . . . . 29

26 Utilizing AVX/AVX2 Instructions to Implement Hyyrö’s Bit-vector Algo-rithm for DNA Sequence Alignment 30Janz Aeinstein F. Villamayor, Kyle Matthew C. Chua, Lorenzo C. Bautista and Roger

Luis Uy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

27 Issues in Lightweight Encryption Algorithm For mHealth 31Tasnuva Ali, A.H. Azni and Nur Hafiza Zakaria . . . . . . . . . . . . . . . . . . . . . 31

xxiii

CONTENTS CONTENTS

28 Numerical Investigation of Bootstrap Artificial Neural Network (p,d,q) Modelfor Malaysia Exchange Rate 32Hila N Z, Muhamad Safiih L and Shazlyn Milleana S . . . . . . . . . . . . . . . . . . 32

SESSION 5 : APPLICATION OF CSE 33

29 Fitting Statistical Distribution of Daily Torrential Rainfall in Peninsular Malaysiafor the Purpose of Simulation 34S.M. Shaharudin, N. Ahmad and N.H. Zainuddin . . . . . . . . . . . . . . . . . . . . . 34

30 Temporal Frequency Spread of Plane Wave Propagation Through Moderateto Strong Turbulence 35Bing Guan and Jaeho Choi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

31 Music Recommendation on Common Artist Basis for Musical Applications 36Manish Agnihotri, Aditya Rathod, Aditya Jajodia and Chethan Sharma . . . . . . . . . 36

32 Flow Induced Vibration (FIV) Analysis of Oil and Gas Process Piping System 37Fazril Irfan Ahmad Fuad, Naimah Lukman and Ahmad Danial Zulhilmi . . . . . . . . 37

33 Performance Evaluation of Fractional Order PID and Sliding Mode Controlwith Optimization Tuning Approach 38Chong Chee Soon, Rozaimi Ghazali, Shin Horng Chong, Chai Mau Shern and Yahaya

Md. Sam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

34 Development of Novel Gamified Online Electrocardiogram Learning Platform(GaMED ECG@TM) 39May Honey Ohn, Zamhar Iswandono, Issa Mchucha, Khin Maung Ohn, Shahril Yusof,

Urban D’Souza . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

35 Molecular Structures, Rotational Barriers, and Density Functional Theorystudies on 3,5-Diamino-6-(2,3-Dichlorophenyl)-1,2,4-Triazine 40Pek-Lan Toh and Suh-Miin Wang . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

SESSION 6 : ADVANCED NETWORKING AND APPLICATIONS 41

36 Evaluation of 5G Key Technique: Non-Orthogonal Multiple Access 42Haiyang Yu, Jaeho Choi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

37 Cloud-based Education as a Service (CEaaS) System Requirements Specifica-tion Model of Higher Education Institutions in Industrial Revolution 4.0 43Yousef A. M. Qasem, Rusli Abdullah, Rodziah Atan and Yusmadi Yah Jusoh . . . . . 43

xxiv

CONTENTS CONTENTS

38 A High Efficiency Gyrostabilizer Antenna Platform for Real-time UAV Syn-thetic Aperture Radar (SAR) Motion Error Compensation 44Zahari Awang Ahmad, Lim Tien Sze, Koo Voon Chet, Shuhaizar Daud, Muhamad Asmi

Romli, Rafikha Aliana A. Raof . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

39 A Wireless Sensor Network (WSN) Implementation using Low-power WideArea Network (LPWAN) 45Renann G. Baldovino and Elmer P. Dadios . . . . . . . . . . . . . . . . . . . . . . . . 45

Sponsors 46

1

(SESSION 1)DATA SCIENCE

TIME: 11:00 - 13:00VENUE: EVENT 2CHAIRPERSON:

DR. MOHD FAHMIMOHAMAD AMRAN

1

Using PERT to Optimise the Completion Timeof Air- Conditioning and Mechanical Ventilation

(ACMV) ProjectChuan-Kian Pang1, Noor-Ajian Mohd-Lair22 and Yi-Sheng Chua3

1Kuala Lumpur Learning Center, Open University Malaysia, Jalan Tun Ismail, 50480 KualaLumpur, Malaysia.

2Faculty of Engineering, University Malaysia Sabah, Jalan UMS, 88400 Kota Kinabalu, Sabah,Malaysia.

3Pansar Engineering Services Sdn Bhd, 18, Jalan Karamunsing, 88818 Kota Kinabalu, Sabah,Malaysia.

kian8808@gmail.com, nrajian@ums.edu.my, chua.yisheng@pansar.com.my

AbstractAn air-conditioning and mechanical ventilation (ACMV) construction project involved manyinterrelated construction activities with varying durations and multiple dependencies. This paperfocuses on the development of a best possible project planning and scheduling technique thatcan help the project manager to manage and complete the ACMV project in optimal time. Theprogram evaluation and review technique (PERT) is used in this research. The PERT techniquehelps to determine the probabilities of various stages of the project by specified deadlines andidentify the activities on the critical path that have high potential for causing delays in theproject completion time. The completion times of 72, 76, 79, and 80 weeks were taken for thePERT analysis. Deadline of 72 weeks is the contractual period of the ACMV project as specifiedin the contract. Completion time of 76 weeks is the critical path of the project as analyzed bythe PERT technique. The deadline of 79 weeks is the proposed completion time for the ACMVproject whereas 80 weeks are the actual completion date of the project. The PERT analysisrevealed that the project completion time of 72 weeks, which is as stated in the contractualcompletion time, yielded the probability of completion of 2.67% only. This analysis shows thatthe contractual period is impossible to be achieved and it is an unrealistic time setting. Furthernegotiation is required including changing or extending of the contractual completion time. Theother completion times are able to be achieved but the company may need to face the highpenalty costs due to the delays in completion time.

Keywords: ACMV, PERT Network, Project Planning and Scheduling.

2

Animal Detection in Highly Cluttered NaturalScenes by Using Faster R-CNN

Wenjun Yu1, Sumi Kim2, Jeong-Hyu Lee3, Jaeho Choi11Department of Electronic Engineering, CAIIT, Chonbuk National University, Chonju, Korea

561-765.2Seoyeong University, Gwangju, Korea.

3Dept. of SW Engineering, Chonbuk National University, Chonju, Korea.wave@jbnu.ac.kr

AbstractWith the increasing awareness of environmental protection, people are paying more and moreattention to the protection of wild animals. Their survival is closely related to human beings. Asprogress in target detection has achieved unprecedented success in computer vision, we can moreeasily target animals. Animal detection based on computer vision is an important branch ofobject recognition, which is applied to intelligent monitoring, smart driving, and environmentalprotection. At present, many animal detection methods have been proposed. However, animaldetection is still a challenge due to the complexity of the background, the diversity of animalposes, and the obstruction of objects. An accurate algorithm is needed. In this paper, the fastRegion-based Convolutional Neural Network (Faster R-CNN) is used. The proposed methodwas tested using the CAMERA_TRAP DATASET. The results show that the proposed animaldetection method based on Faster R-CNN performs better in terms of detection accuracy whenits performance is compared to conventional schemes.

Keywords: deep learning, convolutional neural networks, faster r-cnn, animal recognition,RPN.

3

A Literature Review on Machine LearningTechniques Used for Training Annotated Corpus

Fitrah Rumaisa1, Halizah Basiron2 and Zurina Saaya21Information Technology Department, Widyatama University, Indonesia.

2Faculty of Information And Communication Technology, Universiti Teknikal Malaysia Melaka(UTeM) Melaka, Malaysia.

fitrah.rumaisa@widyatama.ac.id, halizah@utem.edu.my, zurina@utem.edu.my

AbstractThe development of research in the annotation area is growing. Researchers perform annotationtask using various forms of datasets such as text, sound, images, and videos. Various algorithmsare used to perform the tasks. The purpose of this survey is to find out algortihms that are oftenused by researchers to perform annotation tasks especially on text data.. The literature surveysthirteen research papers on text annotation from the last 5 years. The results of this reviewindicate that SVM is the algorithm used for all three annotation methods: manual, automaticand semi-automatic annotation, with a significant accuracy above 80%. The result of this surveywill be referred by the authors as the basis for a subsequent research that will be conducted,especially in the semi-automatic annotation method.

Keywords: Annotation, Algorithm, Text, Survey.

4

Pushing Visualization Effects into PushedSchema Enumerated Tree-Based Support

ConstraintsSIM Doreen Ying Ying1, TEH Chee Siong1 and ISMAIL Ahmad Izuanuddin2

1Faculty of Cognitive Sciences and Human Development, Universiti Malaysia Sarawak, Kuching,Sarawak, Malaysia.

2Respiratory Medicine Unit, Department of Respiratory Medicine, UiTM Medical SpecialistCentre, Faculty of Medicine, Universiti Teknologi MARA, Selangor, Malaysia.

dsdoreenyy@gmail.com, csteh@unimas.my, izuanuddin@salam.uitm.edu.my

AbstractBased on the datasets from UCI and Obstructive Sleep Apnea, a disparate methodology ofuncovering the visualization effects into the pushed support constraints of schema enumeratedtree-based classification techniques is proposed and presented in this paper. This is to actively‘wipe out’ the redundant growing effects of decision trees through itemset generation whenvisualization techniques are applied using Principal Component Analysis (PCA) and/or PrincipalComponent Variable Grouping (PCVG) algorithms. Enumeration specification is based on theschema enumerated tree (SET) drawn after sorting out the features and characteristics on eachdataset applied. The linchpin is to streamline the pre-tree classification effects for post-treeclassification by using visualization techniques, i.e. PCA and/or PCVG, which are appliedduring the SET development. The over-fitting effects done during the SET development bythe pushed support constraints can be counter-corrected by fewer PCA and/or PCVG imposedduring visualization processes. The under-fitting effects done by the imprecise ‘early stopping’of the SET development can be counter-corrected by greater PCA and/or PCVG imposedduring the post-tree classification techniques through pushed SET support constraint learning.Research outcome on all the investigated datasets showed that the prediction accuracies havebeen profoundly improved after applying visualization of PCA and/or PCVG algorithms intothe pushed SET-based or SET-based support constraints.

Keywords: pushed support constraints; visualization effects; Principal Component Analy-sis (PCA) algorithms; schema enumerated tree (SET); Principal Component Variable Grouping(PCVG); error-sensitive learning.

5

A Robust Geometric Skin Colour Face DetectionMethod under Unconstrained Environment of

Smartphone DatabaseNoor Amjed, Fatimah Khalid, Rahmita Wirza O.K. Rahmat, Hizmawati Bint Madzin

Multimedia Department, Faculty of Computer Science and Information Technology, UniversityPutra Malaysia, Serdang, 43400, Malaysia.

noor.alzuhairi32@gmail.com, {fatimahk, rahmita, hizmawati}@upm.edu.my

AbstractFace detection is the primary task in building a vision-based human-computer interaction systemand in special applications such as face recognition, face tracking, face identification, expressionrecognition and also content-based image retrieval. A potent face detection system must beable to detect faces irrespective of illuminations, shadows, cluttered backgrounds, orientationand facial expressions. In previous literature, many approaches for face detection had beenproposed. However, face detection in outdoor images with uncontrolled illumination and imageswith complex background are still a serious problem. Hence, in this paper, we had proposeda Geometric Skin Colour (GSC) method for detecting faces accurately in real world image,under capturing conditions of both indoor and outdoor, and with a variety of illuminationsand also in cluttered backgrounds. The selected method was evaluated on two different facevideo smartphone databases and the obtained results proved the outperformance of the proposedmethod under the unconstrained environment of these databases.

Keywords: face recognition, face detection, illumination variation, background, skin colorsegmentation, smartphone database.

6

Information Extraction from MultifacetedUnstructured Big data – A Short Review in IT,

Engineering and SciencesKiran Adnan, Rehan Akbar and Khor Siak Wang

Faculty of Information & Communication Technology, Universiti Tunku Abdul Rahman, 31900,Kampar, Perak Malaysia.

kiranadnan@1utar.my, rehan@utar.edu.my, khorsw@utar.edu.my

AbstractIn the era of digital globalization, huge volume and variety of data are being produced at a veryhigh rate. Every day, the world is producing around 2.5 quintillion bytes of data. Accordingto IDC, by 2020, over 40 zettabytes of data will be generated and reproduced. Digital datahave become a deluge, overwhelming in every field of information technology (IT), business,science and engineering. These fields are shifting to smart and advanced technologies such assmart manufacturing industries, data-aware medical sciences, and other smart applications ofscience, engineering and technology. These applications are facilitating the industries in contextof data-driven decision making, big data storage, and complex analysis of large data sets. Also,these applications are contributing to generate big data deluge where a variety of data necessitatethe industries to use advanced IT approaches. 95% of the digital universe is unstructured data.It is rich data as it contains information that can play a vital role to improve big data analytics.The heterogeneity, complexity, lack of structured information, poor quality and scalability ofunstructured data generates difficulties in adapting traditional information extraction techniques.Numerous studies on information extraction from different types of data i.e. text, image, audioand video have been conducted but limited to single data type only. Information extractioncan play a vital role in transformation of unstructured data into useful information. A multi-step pipeline with data preprocessing steps, extraction methods and representation are utmostrequirement to improve the unstructured data analytics. In this regard, this paper presents ashort review of information extraction process w.r.t. input data type, extraction methods withtheir corresponding techniques, and representation of extracted information. The issues withunstructured data and the challenges to information extraction from multifaceted unstructuredbig data as well as the future research directions have also been discussed.

Keywords: unstructured data, information extraction, big data.

7

CSMiner-TBM:Tree-Based Mining ContrastSubspace

Florence Sia and Rayner AlfredKnowledge Technology Research Unit, Universiti Malaysia Sabah, Jalan UMS, 88400, Kota

Kinabalu, Sabah, Malaysiaflorence.sfs@ums.edu.my, ralfred@ums.edu.my

AbstractIn recent years, mining contrast subspace of a query object has been introduced due to its manycrucial real life applications. Given a multidimensional data set of two classes, mining contrastsubspace task finds contrast subspaces of a particular query object that is subspaces where thequery object is most likely similar to a target class while dissimilar to other class. All existingmining contrast subspace methods employ density to measure the likelihood of a query object toa target class and other class in a subspace. However, the density tends to decrease when thedimensionality of subspaces increases causes it bounds to identify inaccurate contrast subspacesfor the given query object. This paper proposes a novel tree-based contrast subspace miningmethod which is not affected by the dimensionality of subspaces. It recursively binary partitionsthe subspace space in the way that objects belong to target class are grouped together andseparated from objects belong to other class. Thus, in a contrast subspace, the query objectshould be in a group having higher number of objects of target class than other classes. Theexperiments conducted show that the proposed method is able to mine contrast subspaces of aquery object and outperform the existing method on several real world data sets.

Keywords: Contrast Subspace Mining, Contrast Subspace, Feature Selection.

8

Improved BVBUC Algorithm to DiscoverClosed Itemsets in Long Biological Datasets

Fatimah Audah Md Zaki and Nurul Fariza ZulkurnainDepartment of Electrical and Computer Engineering, Kuliyyah of Engineering, International

Islamic University Malaysia.fatimah.audah@gmail.com, nurulfariza@iium.edu.my

AbstractThe task in mining closed frequent itemsets requires the algorithm to mine the frequent ones thendetermine its closure. The efficiency of closure computation is very important as it will determinethe total mining time and the required memory. Over the years, many closure computationmethods have been proposed to achieve these goals. However, to the best of our knowledge, thereis no suitable method that can be adapted for algorithms that enumerate the rowset lattice,which is effective for biological datasets. Therefore, this paper proposed a method for computingclosure compare with the method used in BVBUC algorithm method. Finally, BVBUC_I isproposed and the performances of these algorithms were evaluated using two synthetic datasetsand three real datasets. The results of these tests proved the efficiency of the proposed method.

Keywords: Data mining, association rules, closed itemset, bioinformatics.

9

(SESSION 2)DATA SCIENCE

TIME: 14:00 - 15:45VENUE: EVENT 2CHAIRPERSON:

DR. HALIZAH BASIRON

10

Testing Design for Pixel Value GraphicalPassword Scheme on Image Specimens

Mohd Fahmi Mohamad Amran, Mohd Afizi Mohd Shukran, Mohd Sidek Fadhil Mohd Yunus,Omar Zakaria, Nurhafizah Moziyana Mohd Yusop, Mohd Nazri Ismail, Mohd Rizal Mohd Isa,

Mohammad Adib Khairuddin, Kamaruzaman Maskat, Yamunah KathiravanComputer Science Department, Faculty of Defense Science and Technology, Universiti

Pertahanan Nasional Malaysia (UPNM), 57000 Kuala Lumpur, Malaysia.fahmiamran@upnm.edu.my

AbstractIn decades, authentication system is relying on username and password as pass-phrase objectfor authentication process. The longer, wider the variety of symbol choices, and randomlyarranged phrase considered as a strong password. A strong password is hard to memorize andmake it less preferred by most of the users in most of the cases. To make pixel value graphicalpassword scheme a test-able subject, a coded prototype was developed and deployed. Thecoded pro-totype was called as Pixel Value Access Control (PVAC). The study is focused on thetesting design and result of experimental testing on different scenario using 24 specimens are usedto test the functionality of PVAC pixel value extraction process to produce an accurate pixel value.

Keywords: Testing Design; Pixel Value; Authentication.

11

A Bicolano-to-Tagalog Transfer-Based MachineTranslation System

Ria Ambrocio SagumFaculty of College of Computer and Information Sciences, Polytechnic University of the

Philippines.rasagum@pup.edu.ph

AbstractThe Bicolano-Tagalog Transfer-based Machine Translation System is an unidirectional machinetranslator for languages Bicolano and Tagalog. The transfer-based approach is divided into threephase: Pre-Processing Analysis, Morphological Transfer, and Sentence Generation. The systemhas to first analyze the source language (Bicolano) input to create some internal representation.This includes the tokenizer, stemmer, POS tag and parser. Through transfer rules, it thentypically manipulates this internal representation to transfer parsed source language syntacticstructure into target language syntactic structure. Finally, the system generates Tagalog sentencefrom own morphological and syntactic information. Each phase will undergo training andevaluation test for the competence of end-results. Overall performance shows a 71.71% accuracyrate.

Keywords: Machine Translation, Transfer-based, Pre-processing analysis, Stemming, POStagging, Morphological Transfer, language model, Bicolano-to-Tagalog.

12

An Intelligent Crisis-mapping Framework forFlood Prediction

Siti Azirah Asmai, Zaheera Zainal Abidin, Halizah Basiron, Sabrina Ahmad, Muhammad AmeerFaisal Adanan and Nor Azim Abdullah

Faculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka,Hang Tuah Jaya, 76100 Durian Tunggal, Melaka, Malaysia.

azirah@utem.edu.my, zaheera@utem.edu.my, halizah@utem.edu.my,

sabrina@utem.edu.my, ameerfaisaladanan@gmial.com, azim.abdullah.58@gmail.com

AbstractThis paper proposes a new framework for crisis-mapping with flood prediction model based on thecrowdsourcing data. Crisis-mapping is still at infancy stage development and offers opportunitiesfor exploration. In fact, the application of the crisis-mapping gives fast information deliveryand continuous updates for crisis and emergency evacuation using sensors. However, currentcrisis-mapping is more to the information dissemination of flood-related information and lackof flood prediction capability. Therefore, this paper applied artificial neural network for floodprediction model in the proposed framework. Sensor data from the crowdsourcing platformcan be used to predict the flood-related measures to support continuous flood monitoring. Inaddition, the proposed framework makes used of the unstructured data from the Twitters tosupport the flood warnings dissemination to locate flood area with no sensors installation. Basedon the results of the experiment, the fitted model from the optimization process gives 90.9% ofaccuracy performance. The significance of this study is that we provide a new alternative in floodwarnings dissemination that can be used to predict and visualized the flood occurrence. Thisprediction is significant to agencies and authorities to identify the flood risk before its occurrenceand crisis-maps can be used as an analytics tool for future city planning.

Keywords: Crisis-mapping, flood prediction, crowdsourcing, artificial neural network andvisualization.

13

Tree-Based Method for Mining ContrastSubspace in Categorical Data

Florence Sia and Rayner AlfredKnowledge Technology Research Unit, Universiti Malaysia Sabah, Jalan UMS, 88400, Kota

Kinabalu, Sabah, Malaysia.florence.sfs@ums.edu.my, ralfred@ums.edu.my

AbstractMining contrast subspace has emerged to find subspaces where a particular query object is mostsimilar to the target class against the other class in a multidimensional data set of two classes.It is important for discovering those subspaces, which are known as contrast subspaces, in manyreal life applications. By now, all existing mining contrast subspace methods (e.g. CSMinerand CSMiner-BPR) focus on numerical data. These methods are based on distance measuresand thus, cannot be directly applied to categorical data. This is due to the fact that there islittle sense in computing distance among categorical data. In this paper, we propose a noveltree-based method for mining contrast subspace in categorical data. It recursively partitions thesubspace space in the way that objects belong to target class are grouped together and separatedfrom objects belong to other class. Hence, the query object should fall in a group having higherprobability of objects from the target class than the other class in a contrast subspace space.An experiment has been conducted on several real world categorical data sets to evaluate theeffectiveness of the proposed method in terms of classification accuracy. The experimental resultsdemonstrate that the contrast subspaces obtained by using the proposed method is capable toimprove the accuracy of most classification tasks. This corresponds to our proposed tree-basedmethod has ability to mine contrast subspaces of the given query object in categorical data.

Keywords: Mining Contrast Subspace, Contrast Subspace, Categorical Data.

14

A Review on Visual Water Quality MonitoringSystem in Precision Aquaculture

Alter Jimat Embug, Ag Asri Ag Ibrahim, Muzaffar Hamzah, and Mohammad Fadhli AsliFaculty of Computing and Informatics Universiti Malaysia Sabah, Malaysia.alterjimat@gmail.com, awgasri@ums.edu.my, muzaffar@ums.edu.my,

ssilfadhli@gmail.com

AbstractThis paper presents the review of available visual water quality monitoring and proposes aconceptual sonification model of audiovisual analytics for precision aquaculture. This studyreviews the current practice of the visual water quality monitoring system used to interpret thecomplex fish farming data. This study also explores the possibility of using an auditory display,by using sound as complementary elements to communicate information from the system to theuser.

Keywords: Sonification, visual analytics, water quality monitoring, real-time, precision aqua-culture, fish farming.

15

Designing A Model Based on Weighted RangeFingerprint Method For Triangulation Analytics

Using Crowdsourcing DataAaron Franklin Soon, Siti Nurulain Mohd Rum, Hamidah Ibrahim, Rohaya Latip ,Razali Yaakob

and Lilly Suriani AffendeyFaculty of Computer Science & Information Technology, University Putra Malaysia.aaronfs1@gmail.com, snurulain@upm.edu.my, hamidah.ibrahim@upm.edu.my,

rohayalt@upm.edu.my, razaliy@upm.edu.my, lilly@upm.edu.my

AbstractThe vast amount of mobile smartphone users provides an infinite source of data for crowd-sourcing. Crowdsourcing provides an economical method of gathering data to cover a largegeographical area compared to traditional methods. However, the inaccurate predictions forbase station localization derived from mobile crowdsourcing impacts its effectiveness for usein radio planning. Therefore, the purpose of this study is to design a model that can yielda more accurate localization through the introduction of a rule-based weighted classification.The methodology deployed is a permutation series based on fingerprint of the cell site withweightage derived from rule-based classification. DeLaunay triangulation and Voronoi diagramsare used to determine the positions of the existing base stations and the prediction of new sitelocation respectively. The expected results are better accuracy of the classification model in the lo-calization prediction of the base station leading to a more accurate prediction of new site location.

Keywords: Crowdsourcing, Triangulation, Fingerprint, Weighted range.

16

(SESSION 3)MODELING AND

SIMULATION / CSEEDUCATION AND FINANCE

TIME: 16:00 - 17:30VENUE: EVENT 2CHAIRPERSON:

A/PROFESSOR DR. RODZIAHBINTI ATAN

17

An Analysis of Gamification on Dota 2’sBusiness Model

Sagguneswaraan Thavamuni1, Hadzariah Ismail1 and Hiroyuki Iida21Faculty of Computing and Informatics, Universiti Malaysia Sabah Labuan International

Campus, 87000 FT Labuan, Malaysia.2School of Information Science, Japan Advanced Institute of Science and Technology, Nomi,

Ishikawa, Japan 923-1211.bi14110317@student.ums.edu.my, had@ums.edu.my, iida@jaist.ac.jp

AbstractGamification has been widely applied to various fields to boost the user experience and en-gagement. Most businesses now have implemented gamification elements into their businessmodels to increase their customer retention. These elements, often called the Customer LoyaltyPrograms have helped businesses retain their customer base by ensuring that their customersget rewarded for their support to the business. This paper analyses the business model of Dota2, a Free-to-Play game which sells various in-game items including the Compendium annually.The compendium allows the players to earn or buy levels and receive rewards based on theirachievements in the compendium. The compendium applies various concepts of gamification intoit. The compendium is then annually improved by adding new rewards, milestones and features.We can expect that the sales of the compendium will increase with the implementation andimprovement of gamification, which in turn can help to sustain the Free-to-Play model of the game.

Keywords: Dota 2, Compendium, gamification, business model, game refinement theory,force-in-mind.

18

Modeling a Novel Uni-MOSFET Solid StateRelay for a Single Phase System Compensation

and Load ControlK. Shashikumar, C. Venkataseshaiah, K.S. Sim

Multimedia University, Faculty of Engineering, Cyberjaya, Malaysia.shashikumar@mmu.edu.my, venkataseshaiah@mmu.edu.my, kssim@mmu.edu.my

AbstractThe concept of replacing mechanical switcher using Solid State Devices (SSR) as the mediumof relaying high power AC load has already been recommended five decades ago using DualThyristor or Triac. The rapid growth in SSR field has now resulted in using dual IGBTs, dualJFETs, dual MOSFETs or dual HEMTs semiconductor devices to control high powered ACloads. One of the drawbacks of all the systems above is that they all using two semiconductordevices to operate for both half cycle of the single phase AC waveform, with either anti-parallelor anti series-arrangement incorporated. We have modeled a novel SSR, using only one MOSFETsemiconductor device with a new SSR module packaging. This novel Uni-MOSFET SSR iscapable of operating on both half cycles of an AC waveform and eliminating problems associatedwith dual semiconductor SSR system. Moreover, in this paper we have explained the simulationsproving that the Uni-MOSFET SSR is able to control resistive load and able to compensatecapacitive and inductive as a SVC, which is as good as its predecessor the Dual-Thyristor SSR.The draw back would be the complexity of the control signals of the Uni-MOSFET SSR for TCRbut we have managed to simplify using equations. Our novel SSR is cost efficient, low noise andworks well as a load switcher, circuit breaker or compensation switch for single phase electricalnetwork system.

Keywords: SSR, MOSFET, SVC, PFC, TCR, TSC.

19

Designing Emotion Interface in E-LearningApplication for Children

Magrizef Gasah1, Aslina Baharum1, Nurul Hidayah May Zain21Faculty of Computing and Informatics, Universiti Malaysia Sabah, 88400 Kota Kinabalu,

Sabah, Malaysia.2Faculty Computer and Mathematical Sciences, Universiti Technology MARA (UiTM),

Cawangan Melaka, Kampus Jasin, 77300 Merlimau, Melaka, Malaysia.magrizef_zone@yahoo.com, aslina@ums.edu.my, nurulmz@fskm.uitm.edu.my

AbstractThis paper shows an overview of methods in designing emotional interfaces for mobile applica-tion. The aim for this paper is to explore and adapt a method, namely as emotional interfacemethodology for creating user interfaces that could trigger user’s positive emotions. The positiveemotions elicited through the user interface design could influences users’ perception such astheir interest and curiosity wherever the design was applied in web-based or mobile-based. Themethod consist of three phases; phase one is the design of a conceptual framework, phase two isthe development of mobile application, and phase three is the evaluation of the implementedconceptual framework in mobile application. By adapting the method, the mobile applicationmay create positive emotions towards children learning process.

Keywords: Emotional, Interface, Methodology, Design, Emotion.

20

Planning Approach For Blended Learning: ALiterature Review

Chong Pei Yee1, Zamhar Iswandono bin Awang Ismail2 and Esmadi Abu Abu Seman21Labuan Matriculation College, Malaysia.

2Universiti Malaysia Sabah, Malaysia.cpykml@gmail.com, zamhar@ums.edu.my, esmadi@ums.edu.my

AbstractHigher learning institutions favor implementing blended learning to achieve their educationalgoals and learning objectives. The purpose of this review is to analyze the ideal blended learningmodel and the recommendations for planning future blended learning activities. Nine articlespublished from 2013 to 2018 which provided the frameworks or models and future suggestionswere appraised. This review revealed that blended learning is not about using Information,Communication, and Technology (ICT) tools anymore but how the learning process is beingblended to deliver effectiveness in learning. The review also discovered that in future blendedlearning design has to include collaboration among educators and how is the learning processcan be applied to actual life. The culture of an educational organization is also an important focus.

Keywords: Blended Learning, Technology in Education, Educational Framework, CurriculumDesign, Higher Education.

21

The Effects of System Quality on SocialCommerce

Nona M Nistah1, Suaini Sura1 and Ook Lee21Faculty Computing & Informatics, UMS, 87000 Labuan, Malaysia.

2Department of Information Systems, Hanyang University, 04763 Seoul, R.O. Korea.nona@ums.edu.my, su_sura@ums.edu.my, ooklee@hanyang.ac.kr

AbstractThis study examines the effect of system quality (SQ) determinants on social commerce (s-commerce) context from the s-commerce consumers’ perspective by adapting IS success model.220 data samples from online survey were analyzed using confirmatory factor analysis (CFA)and the structural equitation model (SEM) to test the research model. The findings show thatvisibility has a significant effect on perceived usefulness, and ease of use and responsiveness hasa significant effect on customer satisfaction. In addition, the finding indicates the significantrelationship among perceived usefulness, customer satisfaction and net benefits in s-commerce.

Keywords: System Quality, S-commerce, E-commerce, Customer Satisfaction.

22

Estimation of CO2 Emission from the Ferries inthe Urban Area

Chien-Chang Chou1,2, Tsun-Lirng Yang3, Hsien-Pin Hsu41Department of Shipping Technology, National Kaohsiung University of Science and Technology

& Department of Shipping Technology, National Kaohsiung Marine University, Taiwan.2Chou’s Science Research Center, Taiwan.

3Department of Marine Engineering, National Kaohsiung University of Science and Technology.4Department of Supply Chain Management, National Kaohsiung University of Science andTechnology, 482, Chung-Chou 3rd Road, Chi-Chin District 805, Kaohsiung City, Taiwan.

ccchou@mail.nkmu.edu.tw, ccchou@nkust.edu.tw

AbstractAir pollutant emission is one of critical environment issues all over the world. This studycalculates the CO2 emission from the ferryboats in the urban area by using Energy EfficiencyDesign Index (EEDI) and Energy Efficiency Operation Index (EEOI) approaches, respectively.A comparison of the CO2 emission estimation results for EEDI and EEOI is also presented toillustrate the difference between EEDI and EEOI approaches. Finally, some conclusions are givento the managers of ferryboats and the governmental department of the urban area.

Keywords: CO2 emission, Air pollution, Ferryboat

23

(SESSION 4)NUMERICAL AND DISCRETE

ALGORITHMSTIME: 10:00 - 13:00

VENUE: BALLROOM 2CHAIRPERSON:

DR. DOREEN YING YING SIM

24

A New Extrapolation Used for IterativeMethods

Uthai Prasopchingchana and Tul ManewattanaDepartment of Mechanical Engineering, Chulalongkorn University, Bangkok 10330, Thailand.

Uthai.P@Student.chula.ac.th, Tul.M@chula.ac.th

AbstractThis paper presents a new extrapolation, the Weighting Factor with Lagrange Interpolating Poly-nomial (WF-LIP), for determining initial guess values in each time step of iterative methods. Themain advantage of the WF-LIP is that it accelerates solution convergence of transient problemsto a steady-state condition. A performance testing of the WF-LIP is carried out by comparingcomputational time based on the “lid-driven cavity flow” problem at the Reynolds number equalto 1000. From the performance testing, the WF-LIP can decrease the computational time of theproblem up to 36.5% compared with the conventional method.

Keywords: Extrapolation, initial guess value, iterative method, Weighting Factor with LagrangeInterpolating Polynomial.

25

Formal Verification of Digital Circuits UsingSimulator with Mathematical Foundation

Wilayat Khan, Basim Azam, Noman Shahid, Abdul Moeed Khan, Ahtisham ShaheenCOMSATS University Islamabad, Wah Campus, Pakistan

wilayat@ciitwah.edu.pk, {accessedon, noman.552769, abdulmoeedkhan79,

ahtisham.shaheen}@gmail.com

AbstractTo ease hardware design process, circuits are normally designed in description languages such asVerilog and VHDL. The correctness of circuits is normally checked by exhaustive simulation insimulators such as Icarus and VCS. Both the description languages Verilog/VHDL and simulatorsIcarus/VCS do not have mathematical foundations and hence are not reliable and cannot beused to mathematically prove correctness of circuit designs. Hardware description languageswith mathematical (formal) foundation such as VeriFormal, on the other hand, are more reliable,trustworthy and can be used for robust design. In this paper, we report our results of formal ver-ifications of two simple hardware circuits designed in the formal description language VeriFormal.Using the VeriFormal simulator and the accompanied type checker tools, we prove reliabilityproperties type safety, functional correctness and functional equivalence of the digital circuits.

Keywords: Formal verification, computer-aided design, theorem provers, equivalence checking,digital logic circuits.

26

A Framework for Halal Knowledge MetadataRepresentations

Rodziah Atan1,2 and Nur Adila Azram31Laboratory of Halal Policy and Management, Halal Products Research Institute, Universiti

Putra Malaysia, Malaysia.2Department of Software Engineering and Information Systems, Faculty of Computer Science

and Information Technology, Universiti Putra Malaysia, Malaysia.3Laboratory of Halal Science Research, Halal Products Research Institute, Universiti Putra

Malaysia, Malaysia.rodziah@upm.edu.my, nuradila.azram@yahoo.com

AbstractEnd users and consumers of halal industry are facing difficulties in finding verified halal infor-mation. This occurred due to information that is stored in silos at every point of activity forevery process chain, employing different structures and models, creating an issue of informationverification. Integration of multiple information systems generally aims at combining selectedsystems so that information can be easily retrieved and manage by users. A proposed fivecomponents metadata representation development methodology is presented in this paper sothat they form a unified new whole and give users the illusion of interacting with one singleinformation system, therefore, data can be represented using the same abstraction principles(unified global data model and unified semantics) without any physical restructuring.

Keywords: Metadata modelling, knowledge representations, ontology, data integration.

27

Needleman-Wunsch Algorithm ComparisonUsing CPU Implementation and GPU

ImplementationRoger Luis Uy, Yun Sup Lee, Yu Sin Kim

College of Computer Science, De La Salle University, Manila, Philippines.roger.uy@dlsu.edu.ph, yun_sup_lee@dlsu.edu.ph, yusin_kim@dlsu.edu.ph

AbstractThe Graphics Processing Unit has become one of the competitive ways to increase the efficiencyin computation performance. To be specific, the parallel computation such as CUDA is nowbeing adapted to various fields for that. One domain that requires higher performance for itsheavy computation is pairwise sequence alignment. This is because as the length of sequencesand the number of data in database increase, the more computations time is involved. To ad-dress such issues, we implemented CUDA based implementation of Needleman-Wunsch dynamicprogramming algorithm aiming to speed up its CPU implementation. This paper has proved thatit outperforms its CPU implementation by approximately three times in terms of computation.Moreover, the increase in sequence size causes computation of CPU implementation to growrapidly, while CUDA implementation only increments gradually.

Keywords: Bioinformatics, Global Alignment, Needleman-Wunsch, GPU, CUDA.

28

Static Profiling of Assembly Code Performanceand Optimization Effectiveness Using

Instructions Performed and Program LatencyChudrack Shalym Y. Salinas, Jonathan Paul C. Cempron and Roger Luis Uy

Computer Technology, College of Computer Studies De La Salle University, Manila Philippineschudrack_salinas@dlsu.edu.ph, jonathan_cempron@dlsu.edu.ph,

roger.uy@delasalle.ph

AbstractSoftware program optimization for improved execution speed can be achieved through modi-fying the program. Programs are usually written in high level languages then translated intolow level assembly language. More coverage of optimization and performance analysis can beperformed on low level than high level language. Optimization improvement is measured inthe difference in program execution performance. Several methods are available for measuringprogram performance are classified into static approaches and dynamic approaches. This paperpresents an alternative method of more accurately measuring code performance statically thancommonly used code analysis metrics. New metrics proposed are designed to expose effectivenessof optimization performed on code, specifically unroll optimizations. An optimization method,loop unroll is used to demonstrate the effectiveness of the increased accuracy of the proposedmetric. The results of this study show that measuring Instructions Performed and InstructionLatency is a more accurate static metric than Instruction Count and subsequently those basedon it.

Keywords: Assembly Programming, Code Profiling, Performance Metrics. Instruction SetArchitecture, Loop Unroll, Vectorization, Compiler, Software Optimization, Time Complexity.

29

Utilizing AVX/AVX2 Instructions to ImplementHyyrö’s Bit-vector Algorithm for DNA

Sequence AlignmentJanz Aeinstein F. Villamayor, Kyle Matthew C. Chua, Lorenzo C. Bautista and Roger Luis Uy

Computer Technology Department, College of Computer Studies, De La Salle University,Manila, Philippines.

janz_aeinstein_villamayor@dlsu.edu.ph, kyle_matthew_chua@dlsu.edu.ph,

lorenzo_bautista@dlsu.edu.ph, roger.uy@dlsu.edu.ph

AbstractDNA sequence alignment is a significant research field in biometrics because locating and analyz-ing the similarities or differences yield deeper knowledge and discovery of biological relationshipswhich contributes to scientific, biological, and medical advancements. DNA sequence alignmentis the process of matching a query sequence with length m against a reference sequence withlength n to obtain the edit distance. However, the process demands a significant amount oftime, thus, researchers are challenged to formulate efficient solution so that it can keep upwith the exponential growth of sequence database. This study presents an implementation ofa bit-vector algorithm for pairwise DNA sequence alignment that takes advantage of SIMDcomputing capabilities on modern processors. The implementation was tested using varyinglengths of query and reference sequences; moreover, the I/O load time, computation time, andmemory consumption were measured. The experimental results show that the implementationhas achieved a I/O load time of θ(n), a computation time of θ(n+m), and memory consumptionof θ(n).

Keywords: DNA sequence alignment, Bit-vector algorithm, SIMD computing capabilities,Modern processors.

30

Issues in Lightweight Encryption Algorithm FormHealth

Tasnuva Ali1,2, A.H. Azni1 and Nur Hafiza Zakaria11Faculty of Science and Technology, Institute Science Islam, Universiti Sains Islam Malaysia

(USIM), Malaysia.2Department of ETE, Daffodil International University, Bangladesh.

tasnuva@daffodilvarsity.edu.bd, ahazni@usim.edu.my, mzhafiza@usim.edu.my

AbstractThe mHealth is the future of next era of Mobile Health that can modify the current paradigm ofcare and allow to improve access of patients’ data with doctors and healthcare industries. Thedata security of all layers of mHealth architecture is considered as immensely significant becauseof handling patients’ data over wireless medium. The main challenges of mHealth are the overallarchitecture, security and privacy concerns with data performance, availability and reliabilityof the devices. Thus, it is indispensable to have a definite security mechanism in mHealtharchitecture to resolve all these limitations. In addition, the cryptography may lead to minimizethe amount of loss done on patients’ data from unauthorized users but their utilization in mHealthis still disputed. Therefore, this paper discusses the issues in lightweight encryption algorithmsto make a comparative analysis among them regarding data privacy in wireless perception layer.From the review, the proposed algorithm of 3D RECTANGLE may give better cipher security forits rotation compared to other lightweight algorithms. This comparative review also concludeswith some research directions in mHealth cryptography for designing secured mHealth networkin future.

Keywords: mHealth, Lightweight Algorithm, Attacks, Security.

31

Numerical Investigation of Bootstrap ArtificialNeural Network (p,d,q) Model for Malaysia

Exchange RateHila N Z1, Muhamad Safiih L2 and Shazlyn Milleana S3

1Department of Mathematics, Faculty of Scince and Mathematic, Universiti Pendidikan SultanIdris, Malaysia.

2School of Informatics & Applied Mathematics, Universiti Malaysia Terengganu, Malaysia.3Department of Mathematics, Faculty of Scince and Mathematic, Universiti Pendidikan Sultan

Idris, Malaysia.nurulhila@fmst.upsi.edu.my, safiihmd@umt.edu.my, shazlyn@fsmt.upsi.edu.my

AbstractIn this paper, a residual bootstrap is plugged-in onto a hybrid neural network autoregressiveintegrated moving average (p,d,q) model in order to improve accuracy of model prediction.Sampling the non-parametric data can help reducing the inherent residual of any statistics model,even though it is a hybrid model. Adding algorithm of B replication of bootstrap sampling tohybrid model can improve the model estimation and forecast precision. In order to investigatenumerically, this paper has considered Malaysia exchange rate information data to be forecasted.By comparing the mean square error result with hybrid model, this plug-in model has higherforecasting accuracy.

Keywords: Accuracy, ANN, ARIMA, Bootstrap, Hybrid model.

32

(SESSION 5)APPLICATION OF CSE

TIME: 14:00 - 15:45VENUE: BALLROOM 2

CHAIRPERSON:DR. ASLINA BAHARUM

33

Fitting Statistical Distribution of DailyTorrential Rainfall in Peninsular Malaysia for

the Purpose of SimulationS.M. Shaharudin11, N. Ahmad2 and N.H. Zainuddin1

1Department of Mathematics, Faculty of Science and Mathematics, Universiti Pendidikan SultanIdris, 35900 Tanjung Malim, Perak.

2Department of Mathematics, Faculty of Science, Universiti Teknologi Malaysia, 81310 Skudai,Johor.

shazlyn@fsmt.upsi.edu.my, norhaiza@utm.my, nurulhila@fsmt.upsi.edu.my

AbstractIn this study, several types of probability distributions was used to fit the daily torrential rainfalldata from 15 monitoring stations of Peninsula Malaysia for the period from 1975 to 2007. Theaim of the study is to determine which distribution fit well the daily torrential Malaysian rainfalldata. Generalized Pareto, Lognormal and Gamma distributions are the distributions that havebeen tested to fit daily torrential rainfall amount in Peninsular Malaysia. First, the appropriatedistribution of the daily torrential rainfall was identified within the selecting distributions forrainfall stations. Then, data sets are generated based on probability distributions that mimica daily torrential rainfall data. Graphical representation and goodness of fit tests were usedin finding the best fit model. The Generalized Pareto was found to be the most appropriatedistribution in describing the daily torrential rainfall amounts of Peninsula Malaysia. The outputscan be beneficial for the purpose of generating several sets of simulated data matrices that mimicto the same characteristics of rainfall data.

Keywords: Daily torrential rainfall amount, Generalized Pareto distribution, Lognormal distri-bution, Gamma distribution, Simulation, Chi-square test, Anderson Darling Test.

34

Temporal Frequency Spread of Plane WavePropagation Through Moderate to Strong

TurbulenceBing Guan1,2, and Jaeho Choi2

1Dept. of Information Engineering, Eastern Liaoning University, Dandong 18003, China.2Dept. of Electronic Engineering, CAIIT, Chonbuk National University, Chonju, 561-756, Rep.

of Korea.wave@jbnu.ac.kr

AbstractIn this study, we derive new expressions for the atmospheric-induced frequency spread of planewaves propagating through moderate to strong turbulence in a horizontal path based on themodified anisotropic non-Kolmogorov power spectrum instead of classic Kolmogorov power spec-trum. The power law value varies from 3 to 4 instead of the classic Kolmogorov power law value;the general amplitude value differs from the classic Kolmogorov constant value 0.033. These newexpressions are based on moderate to strong fluctuation region and anisotropic non-Kolmogorovturbulence. The simulation results show that temporal frequency spread will decrease when theanisotropic factor ζ2−α is increasing.

Keywords: moderate to strong turbulence, frequency spread, anisotropic turbulence.

35

Music Recommendation on Common ArtistBasis for Musical Applications

Manish Agnihotri, Aditya Rathod, Aditya Jajodia and Chethan SharmaDepartment of Information and Communication Technology, Manipal Institute of Technology,

MAHE, Manipal, Karnataka, India.{manish.agnihotri, aditya.rathod, aditya.jajodia}@learner.manipal.edu,

chethan.sharma@manipal.edu

AbstractIn today’s world of growing number of songs, finding apposite music content according to a user’sinterest might be important in many cases. Furthermore, recommendations which might seemsuitable to one user may be contextually irrelevant to another. So there is a need for generatingmusic recommendation specific to users with certain music listening patterns making their taskof choosing music easier. In this paper, we propose a recommendation system for users withcommon-artist music listening patterns. We use random walk with restart algorithm to obtainrelevant recommendations and conduct various experiments to find the optimal values of multipleparameters involved in the procedure.

Keywords: music recommendation, collaborative filtering, random walk.

36

Flow Induced Vibration (FIV) Analysis of Oiland Gas Process Piping System

Fazril Irfan Ahmad Fuad1, Naimah Lukman2 and Ahmad Danial Zulhilmi31Faculty of Chemical Engineering, Universiti Teknologi MARA, Shah Alam, Malaysia.2Faculty of Chemical Engineering, Universiti Teknologi MARA, Shah Alam, Malaysia.

3DM Consultancy Services, Kuala Lumpur, Malaysia.fazrilirfan@salam.uitm.edu.my, naimah.speuitmsc@gmail.com,

ahmad.danial.zulhilmi@googlemail.com

AbstractPiping systems in crude oil production facilities tend to handle high pressure and high velocityflowing fluid at certain flow rate limit resulted in generation of turbulence flow. This turbulencecan generate high levels of broad band kinetic energy which can propagate through the system.Even though the energy is distributed across a wide frequency range, the majority of the excitationis concentrated at low frequency (typically below 100 Hz); the lower the frequency, the higher thelevel of excitation from turbulence. This leads to excitation of the low frequency vibration modesof the pipe work, in many cases causing visible motion of the pipe and, in some cases, the pipesupports and this phenomenon is called Flow induced vibration (FIV) and if excessive can resultin fatigue failure. This objective of the paper is to identify potential sources of FIV on piping bypredicting the Likelihood of failure (LOF) and advising on the necessity for system modificationsto minimise or eliminate any FIV induced piping or line failure from Flow Induced Turbulence.These are done by conducting a FIV assessment for piping systems in oil production facilitieswhich are potentially affected by vibration due to both process conditions and mechanicallyinduced pipework vibration. Another scope of this paper is to apply the recommendation actionrequired in order to reduce the likelihood of failure by changing pipe diameter and/or wallthickness by looking at the fluid structure interaction (FSI). By changing the outer diameter andthe wall thickness of the pipe, the formation of the excitation in flow regime and the presence ofthe critical flow disturbance will decrease. These changes are able to reduce the formation ofFIV thus increasing the lifespan of the piping system.

Keywords: Flow induced Vibration, Excitation, Fatigue failure, Likelihood of Failure, Turbu-lence.

37

Performance Evaluation of Fractional Order PIDand Sliding Mode Control with Optimization

Tuning ApproachChong Chee Soon1, Rozaimi Ghazali1, Shin Horng Chong1, Chai Mau Shern1 and Yahaya Md.

Sam21Centre for Robotics and Industrial Automation, Faculty of Electrical Engineering, Universiti

Teknikal Malaysia Melaka, 76100 Durian Tunggal, Melaka, Malaysia.2Department of Control and Mechatronic Engineering, Faculty of Electrical Engineering,

Univer