2022 6th International Conference on Intelligent Systems, Metaheuristics & Swarm Intelligence - Seoul, Korea (Virtual Conference) | April 9-10 ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
2022 6th International Conference on Intelligent Systems, Metaheuristics & Swarm Intelligence Seoul, Korea (Virtual Conference) | April 9-10, 2022 Organized by Technically Sponsored by
WELCOME MESSAGE It is our great pleasure to welcome you to attend 2022 6th International Conference on Intelligent Systems, Metaheuristics & Swarm Intelligence (ISMSI 2022), an annual international event of India International Congress on Computational Intelligence (IICCI). This event will provide an excellent opportunity for researchers, scientists and technologists who are working in the emerging areas of intelligent systems, metaheuristics & swarm intelligence, to assemble and share their latest research efforts and findings. We had been looking forward to seeing everyone at the conference in Seoul, Republic of Korea, but due to the continued Covid- 19 pandemic, the organizing committee has decided to convert the conference to a full-fledged virtual event, because the safety and well-being of our participants can't be compromised. The conference programme includes oral paper presentations along with keynote speeches by leading researchers. We are confident that over these two days you will get the theoretical grounding, practical knowledge and personal contacts that will help you in building long-term, profitable and sustainable communication among researchers and practitioners working in a wide variety of scientific areas with a common interest in intelligent systems, metaheuristics & swarm intelligence. It is hoped that this conference will provide each one of you with a good platform for networking opportunities and interactions with other delegates from both the academics and industry. We look forward to seeing you all next year. Prof. Suash Deb (General Chair, ISMSI22) March 18, 2022
01. Online Presentation Guideline
02. Conference Committees
03. Program at a Glance
04. Keynote Speakers
Table of Contents
05. Contents of Sessions
06. Oral Presentation Abstracts
07. Listener
01
01
Online Presentation Guideline
Install the Zoom application to join virtual conference
& some tips for presentation
The meeting invitation link: https://us06web.zoom.us/j/81527236343 & https://us06web.zoom.us/j/87828084933
Name Setting before Entry
Online Presentation Guideline
Keynote Speaker: Keynote-Name Author: Paper ID-Name
Committee: Position-Name Listener: Listener-Name
Zoom Pre-Test on April 9
Participants who are going to do an online
presentation are required to join the Zoom test
session on Saturday, April 9 start from 10:00am.
Please download the Zoom App and prepare
URL: https://zoom.us/download your presentation slides before you do the test.
Note
The meeting room normally will be opened 30
minutes before the scheduled time. Please
enter the room 10-15 minutes earlier.
For online participants, the certificate will be
sent to you by e-mail after the conference.
01
02
Conference Committees
Honorary Co-chairs
Prof. Juergen Branke, Warwick Business School, UK
Prof. Alice E. Smith, Joe W. Forehand/Accenture Distinguished Professor, Auburn University, USA
General Chair
Suash Deb, Secretary General - India Intl. Congress on Computational Intelligence
International Advisory Board
Conference
Jorn Altmann, Seoul National University, South Korea
Christian Blum, Spanish National Research Council, Spain
Aboul Ella Hassanien, Cairo University, Egypt
Tzung-Pei Hong, National University of Kaohsiung, Taiwan
Committees
Nikola K. Kasabov, Auckland University of Technology, New Zealand
Laszlo T. Koczy, Budapest University of Technology & Economics, Hungary
Meng-Hiot Lim, Nanyang Technological University, Singapore
Nadia Nedjah, State University of Rio de Janeiro, Brazil
Yaroslav D. Sergeyev, Universita della Calabria, Renede, Italy
Patrick Siarry, Universite Paris-Est Creteil, France
Jaya Sil, Indian Institute of Engineering Science and Technology Shibpur, India
Kenneth Sorensen, Universiteit Antwerpen, Belgium
Qingfu Zhang, City University of Hong Kong, Hong Kong
Organizing Co-chairs
Zong Woo Geem, Gachon University, South Korea
Joong Hoon Kim, Korea University, South Korea
Ashutosh Mishra, Yonsei University, South Korea
Toshiaki Omori, Kobe University, Japan
Organizing Co-chairs
Rakesh Shrestha , Yonsei University, South Korea
Ka-Chun Wong, City University of Hong Kong, Hong Kong
Shixiong Zhang, Xidian University, China
Program Co-chairs
Amir H. Alavi, University of Pittsburgh, USA
Thomas Hanne, University of Applied Sciences & Arts Northwestern Switzerland
Deepak Mishra, Indian Institute of Space Science and Technology Trivandrum
Conference Publications Co-chairs
Andries P. Engelbrecht, Stellenbosch University, South Africa
Committees
Iztok Fister Jr., University of Maribor, Slovenia
Celso C. Ribeiro, Universidade Federal Fluminense, Institute of Computing, Brazil
Prashant Singh, Umea University, Sweden
Xin-She Yang, Middlesex University, UK
Monica Chis, Freelance IT consultant & Trainer, Romania
V. Susheela Devi, Indian Institute of Science Bangalore, India
Mohammed El-Abd, American University of Kuwait, Kuwait
Amir H. Gandomi, University of Technology Sydney, Australia
Andres Iglesias, University of Cantabria, Spain
Sameerchand Pudaruth, University of Mauritius, Mauritius
International Program Committee
Michel Aldanondo, Toulouse University, France
Ankit Chaudhary, University of Missouri, USA
Monica Chis, Freelance IT consultant & Trainer, Romania
Marco Cococcioni, University of Pisa, ItalyInternational Program Committee
V. Susheela Devi, Indian Institute of Science Bangalore, India
Kei Eguchi, Fukuoka Institute of Technology, Japan
Mohammed El-Abd, American University of Kuwait, Kuwait
Iztok Fister Jr., University of Maribor, Slovenia
Amir H. Gandomi, University of Technology Sydney, Australia
Jun Hu, Harbin University of Science & Technology, China
Andres Iglesias, University of Cantabria, Spain
Donghwi Jung, Korea University, South Korea
Tee Yee Kai, Universiti Tunku Abdul Rahman, Malaysia
Joongheon Kim, Korea University, South Korea
Manoj Kumar, University of Petroleum and Energy Studies, India
Conference
Marat Mukhametzhanov, University of Calabria, Italy
Anand Nayyar, Duy Tan University, Vietnam
Toshiaki Omori, Kobe University, Japan
Committees
Elisha Opiyo, University of Nairobi, Kenya
Olutomilayo Petinrin, City University of Hong Kong, Hong Kong
Mikhail Posypkin, Russian Academy of Sciences, Russia
Sameerchand Pudaruth, University of Mauritius, Mauritius
Yain Whar Si, University of Macau, Macau
Prashant Singh, Umea University, Sweden
Pritpal Singh, The Jagiellonian University, Poland
M. Tanveer, Indian Institute of Technology Indore, India
Boldizsar Tuu-Szabo, Szechenyi Istvan University, Hungary
Ka-Chun Wong, City University of Hong Kong, Hong Kong
Yew Kee Wong, JiangXi Normal University, China
Xin-She Yang, Middlesex University, UK
Shixiong Zhang, Xidian University, China01
03
Program at a Glance
The meeting invitation link: https://us06web.zoom.us/j/81527236343 & https://us06web.zoom.us/j/87828084933Program at a Glance
April 9 (Saturday), [Greenwich Mean Time (GMT+9) – Seoul Local Time]
Test Session
Time Agenda venue
10:00am-4:00pm Keynote speakers & Session 1 Meeting ID: 815 2723 6343
10:00am-4:00pm Session 2, 3 Meeting ID: 878 2808 4933
For ALL Online Presentations
Participants who are going to do an online presentation are required to join the Zoom test session on Saturday,
April 9.
Duration: 3~5 minutes apiece. Feel free to leave after your rehearsal is done.Program at a Glance
April 10 (Sunday), [Greenwich Mean Time (GMT+9) – Seoul Local Time]
Time Agenda venue
Welcome Address-Prof. Suash Deb
10:00am-10:10am General Chair, ISMSI2022
Founding Secretary General, IICCI
Welcome Address-Prof. Panos M. Pardalos
10:10am-10:15am
University of Florida, USA
Welcome Address-Prof. Carlos A. Coello Coello
10:15am-10:20am
Department of Computer Science CINVESTAV-IPN, Mexico
Welcome Address-Prof. Ke Tang Meeting ID:
10:20am-10:25am
Southern University of Science and Technology, China 815 2723 6343
Keynote Speech 1-Prof. Panos M. Pardalos
10:25am-11:05am University of Florida, USA
“Less is More Approach in Optimization and the Power of Heuristics”
11:05am-11:20am Break Time
Keynote Speech 2-Prof. Carlos A. Coello Coello
11:20am-12:00pm Department of Computer Science CINVESTAV-IPN, Mexico
“Where is the Research on Evolutionary Multi-objective Optimization Heading to?”
12:00pm-1:30pm Lunch TimeProgram at a Glance
April 10 (Sunday), [Greenwich Mean Time (GMT+9) – Seoul Local Time]
Time Agenda venue
Keynote Speech 3-Prof. Ke Tang
1:30pm-2:10pm Southern University of Science and Technology, China
“Evolving Generalizable Parallel Algorithm Portfolios”
2:10pm-2:20pm Break Time
Meeting ID:
Session 1-Image Processing and Algorithm 815 2723 6343
2:20pm-4:05pm
Chair-Prof. Ke Tang
4:05pm-4:20pm Break Time
Session 2-Data Model and Intelligent Computing
4:20pm-6:20pm
Chair-Prof Andries Engelbrecht
Session 3-Algorithm Design and Optimization Meeting ID:
4:20pm-6:20pm
Chair-Prof Thomas Hanne 878 2808 493301
04
Keynote Speakers
The meeting invitation link: https://us06web.zoom.us/j/81527236343 & https://us06web.zoom.us/j/87828084933Keynote Speakers
Dr. Panos Pardalos is a Distinguished Professor in the Department of Industrial and Systems
Engineering at the University of Florida, and an affiliated faculty of Biomedical Engineering and
Computer Science & Information & Engineering departments. In addition, he is the director of the
Center for Applied Optimization. Dr. Pardalos is a world renowned leader in Global Optimization,
Mathematical Modeling, Energy Systems, and Data Sciences. He is a Fellow of AAAS, AIMBE, and
INFORMS and was awarded the 2013 Constantin Caratheodory Prize of the International Society of
Global Optimization. In addition, Dr. Pardalos has been awarded the 2013 EURO Gold Medal prize
bestowed by the Association for European Operational Research Societies. This medal is the preeminent
European award given to Operations Research (OR) professionals for "scientific contributions that stand
the test of time." Dr. Pardalos has been awarded a prestigious Humboldt Research Award (2018-2019).
Prof. Panos M. The Humboldt Research Award is granted in recognition of a researcher's entire achievements to date -
Pardalos fundamental discoveries, new theories, insights that have had significant impact on their discipline. Dr.
University of Florida, USA Pardalos is also a Member of several Academies of Sciences, and he holds several honorary PhD
degrees and affiliations. He is the Founding Editor of Optimization Letters, Energy Systems, and Co-
10:25am-11:05am April 10 (GMT+9) Founder of the International Journal of Global Optimization, Computational Management Science, and
Meeting ID: 815 2723 6343 Springer Nature Operations Research Forum. He has published over 500 journal papers, and
edited/authored over 200 books. He is one of the most cited authors and has graduated 64 PhD
students so far. Details can be found in www.ise.ufl.edu/pardalos.Abstract of Keynote Speech
Less is More Approach in Optimization and the Power of
Heuristics
Large scale problems in the design and analysis of networks, energy systems, biomedicine, finance, and
engineering are modeled as optimization and control problems. Both humans and nature are constantly
optimizing to minimize costs or maximize profits, to maximize the flow in a network, or to minimize the
probability of a blackout in the smart grid. The resulting optimization problems very often are
nonconvex, hard to solve, or of very large scale. Exact algorithms are of very limited use in these cases.
Due to new algorithmic developments in heuristics, as well as the computational power of machines,
Prof. Panos M. optimization heuristics have been used to solve problems in a wide spectrum of applications in science
and engineering. In this talk, we are going to address new developments including the "Less is more
Pardalos
Approach in Optimization", as well as discuss their power to solve hard problems and new
University of Florida, USA developments for their evaluation.
10:25am-11:05am April 10 (GMT+9)
Meeting ID: 815 2723 6343Keynote Speakers
Dr. Carlos Artemio Coello Coello received a PhD in Computer Science from Tulane University (USA) in
1996. He currently has over 500 publications which, according to Google Scholar, report over 54,800
citations (with an h-index of 94). He has received several awards, including the National Research Award
(in 2007) from the Mexican Academy of Science (in the area of exact sciences), the 2012 National Medal
of Science in Physics, Mathematics and Natural Sciences from Mexico's presidency (this is the most
important award that a scientist can receive in Mexico). the prestigious 2013 IEEE Kiyo Tomiyasu
Award, "for pioneering contributions to single- and multiobjective optimization techniques using
bioinspired metaheuristics", of the 2016 The World Academy of Sciences (TWAS) Award in
"Engineering Sciences" and of the 2021 IEEE CIS Evolutionary Computation Pioneer Award. Since
January 2011, he is an IEEE Fellow. Since 2010, he is a Full Professor with distinction at the Computer
Prof. Carlos A. Science Department of CINVESTAV-IPN in Mexico City, Mexico. He specializes on the design of
Coello Coello metaheuristics for solving nonlinear multi-objective problems. He is currently the Editor-in-Chief of the
Department of Computer IEEE Transactions on Evolutionary Computation.
Science CINVESTAV-IPN, Mexico
11:20am-12:00pm April 10 (GMT+9)
Meeting ID: 815 2723 6343Abstract of Keynote Speech
Where is the Research on Evolutionary Multi-objective
Optimization Heading to?
The first multi-objective evolutionary algorithm was published in 1985. However, it was not until the late
1990s that so-called evolutionary multi-objective optimization began to gain popularity as a research area.
Throughout these 36 years, there have been several important advances in the area, including the
development of different families of algorithms, test problems, performance indicators, hybrid methods
and real-world applications, among many others. In the first part of this talk we will take a quick look at
some of these developments, focusing mainly on some of the most important recent achievements. In
Prof. Carlos A. the second part of the talk, a critical analysis will be made of the by analogy research that has
proliferated in recent years in specialized journals and conferences (perhaps as a side effect of the
Coello Coello
abundance of publications in this area). Much of this research has a very low level of innovation and
Department of Computer almost no scientific input, but is backed by a large number of statistical tables and analyses. In the third
Science CINVESTAV-IPN, Mexico and final part of the talk, some of the future research challenges for this area, which, after 36 years of
11:20am-12:00pm April 10 (GMT+9) existence, is just beginning to mature, will be briefly mentioned.
Meeting ID: 815 2723 6343Keynote Speakers
Dr. Ke Tang is a Professor at the Department of Computer Science and Engineering, Southern
University of Science and Technology (SUSTech). Before joining SUSTech in January 2018, he was with
the School of Computer Science and Technology, University of Science and Technology of China
(USTC), first as an Associate Professor (2007-2011) and then as a Professor (2011-2017). His major
research interests include evolutionary computation and machine learning, particularly in large-scale
evolutionary computation, integration of evolutionary computation and machine learning, as well as
their applications.He has published more than 180 papers, which have received over 10000 Google
Scholar citations with an H-index of 48. Professor Tang is a recipient of the IEEE Computational
Intelligence Society Outstanding Early Career Award (2018), the Newton Advanced Fellowship (Royal
Society, 2015) and the Natural Science Award of Ministry of Education of China (2011 and 2017). He is
Prof. Ke Tang an Associate Editor of the IEEE Transactions on Evolutionary Computation and served as a member
of Editorial Boards for a few other journals.
Southern University of Science
and Technology, China
1:30pm-2:10pm April 10 (GMT+9)
Meeting ID: 815 2723 6343Abstract of Keynote Speech
Evolving Generalizable Parallel Algorithm Portfolios
Parallel Algorithm Portfolios (PAPs), being generally applicable to nearly all kinds of computation
(optimization/decision/counting/learning) problems and friendly to modern parallel computing facilities,
has become a framework adopted by many industrial software systems. On the other hand, to configure
a good PAP in practice has emerged as a tedious and challenging problem. For it population-based
search nature, Evolutionary Computation, in particular, co-evolution, offers some off-the-shelf ideas for
automated PAP configuration, which will be introduced in this talk. Specifically, we will show that, when
the training instances are sufficient, high-performance PAPs can be automatically constructed with little
Prof. Ke Tang human effort involved, by a co-evolutionary approach. In case the sample is of small size or biased,
which is often encountered in practice, we propose to use competitive co-evolution of the PAPs and the
Southern University of Science instance set to tackle such challenges. The codes and datasets are available at
and Technology, China https://github.com/senshineL/CEPS.
1:30pm-2:10pm April 10 (GMT+9)
Meeting ID: 815 2723 634301
05
Contents of Sessions
Note: Please find out which session your paper is included in and
enter the meeting room at least 10 minutes before the session starts.
The meeting invitation link: https://us06web.zoom.us/j/81527236343 & https://us06web.zoom.us/j/87828084933Session 1: Image Processing and Algorithm
2:20pm-4:05pm (GMT+9), April 10, Sunday, Meeting ID: 815 2723 6343
Session Chair: Prof. Ke Tang
Paper ID Time Title & Authors
Evolutionary Algorithm for Solving Supervised Classification Problems: An Experimental
Study
SI012 2:20pm-2:35pm
Daniel Soto and Wilson Soto
Politécnico Grancolombiano, Colombia
A New Discrete Whale Optimization Algorithm with a Spiral 3-opt Local Search for
Solving the Traveling Salesperson Problem
SI019 2:35pm-2:50pm
Elias Rotondo and Steffen Heber
North Carolina State University, USA
Structured Pruning with Automatic Pruning Rate Derivation for Image Processing Neural
Networks
SI021 2:50pm-3:05pm
Yasufumi Sakai, Akinori Iwakawa and Tsuguchika Tabaru
Fujitsu Limited, JapanSemanTV: A Content-Based Video Retrieval Framework
China Marie Lao, Juan Miguel Mendoza, Antolin Alipio, Anne Camille Maupay, Charito
SI023 3:05pm-3:20pm
Molina, Criselle Centeno, Dan Michael Cortez, Jonathan Morano
Pamantasan ng Lungsod ng Maynila, Philippines
A Novel Approach to Low Light Object Detection Using Exclusively Dark Images
SI014 3:20pm-3:35pm Ankit Kumar, Dr Bijal Talati, Mihir Rajput, Harshal Trivedi
Softvan Pvt Ltd, India
Automation of Fabric Pattern Construction using Genetic Algorithm
SI013 3:35pm-3:50pm Omema Ahmed, Muhammad Salman Abid, Aiman Junaid and Syeda Saleha Raza
Habib University, Pakistan
N-Gram-based Machine Learning Approach for Bot or Human Detection from Text
Messages
SI011 3:50pm-4:05pm
Durga Prasad Kavadi, Chandra Sekhar Sanaboina, Rizwan Patan, Amir H. Gandomi
B V Raju Institute of Technology, IndiaSession 2: Data Model and Intelligent Computing
4:20pm-6:20pm (GMT+9), April 10, Sunday, Meeting ID: 815 2723 6343
Session Chair: Prof Andries Engelbrecht
Paper ID Time Title & Authors
A Framework for Estimating Integrated Information of Brain Based on Deep Neural
Network
SI024-A 4:20pm-4:35pm
Ryo Omae, Toshiaki Omori
Kobe University, Japan
Set-based Particle Swarm Optimization for Data Clustering
SI016 4:35pm-4:50pm Lienke Brown, AP Engelbrecht
Stellenbosch University, South Africa
Identifying the Best Combination of Crossover and Mutation Operators in NSGA-II for
Redundancy-based Optimal Design of Water Network
SI004-A 4:50pm-5:05pm
Jaehyun Kim, Soyeon Lim and Donghwi Jung
Korea University, KoreaStatic Polynomial Approximation Using Set-based Particle Swarm Optimisation
SI017 5:05pm-5:20pm D Edeling, AP Engelbrecht
Stellenbosch University, South Africa
AxDFM:Position prediction system based on the importance of high-order features
SI018 5:20pm-5:35pm Chang Su, Haoxiang Feng and Xianzhong Xie
Chongqing University of Posts and Telecommunications, China
Estimating Dynamical Nonlinear System with Nonstationarity by Gaussian Process Self-
Organizing Generalized State-Space Model
SI026-A 5:35pm-5:50pm
Takashi Terayama, Toshiaki Omori
Kobe University, Japan
Stability-Guided Multi-Guide Particle Swarm Optimization Algorithm
SI015 5:50pm-6:05pm W Steyn, AP Engelbrecht
Stellenbosch University, South Africa
Application of Hybrid PSO and SQP Algorithm in Optimization of the Retardance of
Citrate Coated Ferrofluids
SI002 6:05pm-6:20pm
Jing-Fung Lin, Jer-Jia Sheu
Far East University, TaiwanSession 3: Algorithm Design and Optimization
4:20pm-6:20pm (GMT+9), April 10, Sunday, Meeting ID: 878 2808 4933
Session Chair: Prof Thomas Hanne
Paper ID Time Title & Authors
Coevolutionary Algorithm for Evolving Competitive Strategies in the Weapon Target
Assignment Problem
SI008 4:20pm-4:35pm Ehab Elfeky, Madeleine Cochrane, Luke Marsh, Saber Elsayed, Brendan Sims, Simon
Crase, Daryl Essam and Ruhul Sarker
University of New South Wales, Australia
Understanding the Effects of Ant Algorithms on Path Planning with Gain-Ant Colony
Optimization
SI005 4:35pm-4:50pm
V Sangeetha, R Krishankumar, K S Ravichandran and Amir H Gandomi
Amrita School of Engineering, India
Genetic Algorithm with Machine Learning to Estimate the Optimal Objective Function
Values of Subproblems
SI006 4:50pm-5:05pm
Yohei Hazama and Hitoshi Iima
Kyoto Institute of Technology, JapanResource Prediction of Virtual Network Function Based on Traffic Feature Extraction
SI028 5:05pm-5:20pm Chang Su, Ya Tan, Xianzhong, Xie, Yong Liu
Chongqing University of Posts and Telecommunications, China
Assessing the Quality of Car Racing Controllers in a Virtual Setting Under Changed
Conditions
SI007 5:20pm-5:35pm
Sebastian Minder, Marc Funken, Rolf Dornberger and Thomas Hanne
University of Applied Sciences and Arts Northwestern Switzerland, Basel/Olten, Switzerland
Unsupervised Deep Video Interpolation Based on Spatio-Temporal Autoregressive Neural
Network
SI025-A 5:35pm-5:50pm
Koki Nakashima, Toshiaki Omori
Kobe University, Japan
Robotic Path Planning by Q Learning and a Performance Comparison with Classical Path
Finding Algorithms
SI003 5:50pm-6:05pm
Phalgun Chintala, Rolf Dornberger and Thomas Hanne
University of Applied Sciences and Arts Northwestern Switzerland, Basel/Olten, Switzerland
A Hybrid Multi-Objective Teaching Learning-Based Optimization Using Reference Points
and R2 Indicator
SI010 6:05pm-6:20pm
Farajollah Tahernezhad-Javazm, Debbie Rankin, Damien Coyle
Ulster University, UK01
06
Oral Presentation Abstracts
The meeting invitation link: https://us06web.zoom.us/j/81527236343 & https://us06web.zoom.us/j/87828084933Session Chair: Prof. Ke Tang
Southern University of Science and Technology, China S1
Image Processing
Note: and Algorithm
Greenwich Mean Time (GMT+9) – Seoul Local Time
Time: 2:20pm-4:05pm, April 10, 2022
Meeting ID: 815 2723 6343
Please enter the room 5-10 minutes earlierSession 1: Image Processing and Algorithm
2:20pm-2:35pm 2:35pm-2:50pm
Evolutionary Algorithm for Solving Supervised A New Discrete Whale Optimization Algorithm with a
SI012 Classification Problems: An Experimental Study SI019 Spiral 3-opt Local Search for Solving the Traveling
Salesperson Problem
Wilson Soto
Steffen Heber
Politécnico Grancolombiano, Colombia
North Carolina State University, USA
Abstract—Evolutionary Algorithms (EAs) are population-based, stochastic search Abstract—The whale optimization algorithm is a metaheuristic inspired by the hunting
algorithms that mimic natural evolution. Over the years, EAs have been successfully strategy of humpback whales. This paper proposes a new discrete spiral whale
applied to many classification problems. In this paper, we propose to demonstrate the optimization algorithm (DSWOA) to solve the traveling salesperson problem (TSP). Our
performance of an improved evolutionary algorithm for synthesizing classifiers in approach uses sequential consecutive crossover and spiral 3-opt search, a modified
supervised data scenarios. This approach generates an arithmetic expression DAG version of the popular 3-opt local search. Spiral 3-opt search works like the original 3-
(Directed Acyclic Graph) for each training class in order to adjust each test class to one opt heuristic but only uses part of the tour to generate 3-opt moves. We show that
of them. We compare our approach with well-known machine learning methods, such spiral 3-opt achieves results similar to the original 3-opt technique and significantly
as SVM and KNN. The performance of the improved algorithm for evolving classifiers is reduces runtime. We evaluate DSWOA’s performance on 19 TSP instances against six
competitive with respect to the solution quality. benchmark algorithms. Our results suggest that DSWOA produces TSP solutions that
are as good or better than our competitors. For five of the six benchmark algorithms,
we demonstrated statistically significant improvements.Session 1: Image Processing and Algorithm
2:50pm-3:05pm 3:05pm-3:20pm
Structured Pruning with Automatic Pruning Rate SemanTV: A Content-Based Video Retrieval
SI021 Derivation for Image Processing Neural Networks SI023 Framework
Yasufumi Sakai Juan Miguel Mendoza
Fujitsu Limited, Japan Pamantasan ng Lungsod ng Maynila, Philippines
Abstract—With the increased adaption of CCTV for surveillance, challenges in terms of
Abstract—Structured pruning has been proposed for network model compression.
retrieval have recently gained attention. Most Surveillance Video Systems can only
Because most of existing structured pruning methods assign pruning rate manually,
retrieve footage based on its metadata, (date, time, camera location, etc.) which limits
finding appropriate pruning rate to suppress the degradation of pruned model
the diversity of meaningful footage intended to be retrieved by the user. To solve this, a
accuracy is difficult. Although we have been proposed the automatic pruning rate
content-based video retrieval framework was proposed to retrieve relevant videos
search method, the pruned model performances for complex image processing task
based on their content and match it to the user’s query. This framework composes of
such as ImageNet have not been evaluated. In this paper, we demonstrate a
two (2) methods: A method for Video Content Extraction that utilizes Google’s Video
performance of the pruned model on ImageNet task using our proposed structured
Intelligence API for Optical Character Recognition and Label Detection, and a method
pruning method. Furthermore, we evaluate our pruning method in comparison of the
for Video Retrieval. Various setups for the Video Retrieval method are explored; this
pruned model performance using CIFAR-10 and ImageNet. When using ResNet-34 on
includes the usage of SBERT and Okapi BM25. Each setup was tested against various
ImageNet task, our proposed method reduces model parameters of ResNet-34 by
text queries with equivalent test video results based on the MSVD dataset. To measure
44.0% with 72.99% accuracy.
each setup’s performance in terms of relevance, Recall and Precision at K and Median
and Mean Rank were used. It was concluded that the framework composed of the
Video Intelligence API along with SBERT alone performed better than the other
proposed setup for returning videos relevant to the user’s text query more accurately
than the other setups of the method.Session 1: Image Processing and Algorithm
3:20pm-3:35pm 3:35pm-3:50pm
A Novel Approach to Low Light Object Detection Automation of Fabric Pattern Construction using
SI014 Using Exclusively Dark Images SI013 Genetic Algorithm
Ankit Kumar Omema Ahmed
Softvan Pvt Ltd, India Habib University, Pakistan
Abstract—The efficiency of our vision highly depends on the light’s intensity. In dark Abstract—This paper introduces the use of Genetic Algorithms to evolve fabric patterns
images, the intensity of light in our surroundings is generally lower, reducing the from randomly generated seeds. The patterns are evolved from random, often dull
efficiency of vision and the capability to distinguish different objects. An analysis of coloring of the image, to bright multi-color patterns that are aesthetically pleasing in
lowlight images is possible with handcrafted and learned features. This process of nature. The main problem that this paper intends to solve is to introduce complete
object recognition also needs to take into consideration the intensity of light that is automation in the design process of patterns, which have historically been dependent
produced by a particular pixel varies depending on the color space used for a particular upon human arbitrators to judge the quality of intermediate outputs. In its stead, the
image since different colors produce different intensities of light. Therefore, the proposed algorithm evaluates the quality of the image using inherent latent features
exclusively dark dataset has been used recently as a benchmark dataset for object present in the image itself. Our algorithm takes into account the distribution of color,
recognition in the dark that contains 10 low light illumination types and 12 different global contrast, and the overall dullness score of the image to evaluate the quality of
categories of objects, and it has the potential to be used as the standard database for the generated patterns. To create diverse patterns that feel more natural, different
benchmarking research in the domain of low light. CSPNet is essential for the purpose approaches are experimented with. These include the use of L-systems and image
of feature extraction. This reduces the computational load required by our model and processing techniques, in a bid to construct a pattern which seems more human-like,
also ensures that the accuracy does not significantly reduce. When it is coupled with the rather than just rudimentary digital art.
CNN, the results show potential for practical applications. The goal of this paper is to
further improve the recognition rate of various objects in the dark.Session 1: Image Processing and Algorithm
3:50pm-4:05pm
N-Gram-based Machine Learning Approach for Bot
SI011
or Human Detection from Text Messages
Durga Prasad Kavadi
B V Raju Institute of Technology, India
Abstract—Social bots are computer programs created for automating general human activities like
the generation of messages. The rise of bots in social network platforms has led to malicious activities
such as content pollution like spammers or malware dissemination of misinformation. Most of the
researchers focused on detecting bot accounts in social media platforms to avoid the damages done
to the opinions of users. In this work, n-gram based approach is proposed for a bot or human
detection. The content-based features of character n-grams and word n-grams are used. The
character and word n-grams are successfully proved in various authorship analysis tasks to improve
accuracy. A huge number of n-grams is identified after applying different pre-processing techniques.
The high dimensionality of features is reduced by using a feature selection technique of the Relevant
Discrimination Criterion. The text is represented as vectors by using a reduced set of features.
Different term weight measures are used in the experiment to compute the weight of n-grams
features in the document vector representation. Two classification algorithms, Support Vector
Machine, and Random Forest are used to train the model using document vectors. The proposed
approach was applied to the dataset provided in PAN 2019 competition bot detection task. The
Random Forest classifier obtained the best accuracy of 0.9456 for bot/human detection.Session Chair: Prof Andries Engelbrecht
S2 Stellenbosch University, South Africa
Data Model and
Note:
Greenwich Mean Time (GMT+9) – Seoul Local Time
Intelligent Computing
Time: 4:20pm-6:20pm, April 10, 2022
Meeting ID: 815 2723 6343
Please enter the room 5-10 minutes earlierSession 2: Data Model and Intelligent Computing
4:20pm-4:35pm
A Framework for Estimating Integrated Information of Brain Based on Deep Neural
SI024-A Network
Ryo Omae
Kobe University, Japan
Abstract—Analyzing higher brain functions such as emotion, thinking, and consciousness is still challenging since it requires knowing the
relation between partial and entire dynamics of neurons. The integrated information theory says the integration of information is the key to
this relation. This theory provides a method for quantifying the amount of information generated by the whole system. While the
integrated information would help us to know the characteristic of the neural system, its computation cost prevents us from making it
practical.
Some approaches, such as Queranne's Algorithm and the Spectral Clustering(SC)-based method, have been presented to solve this
problem.
Because of the nonlinearity in neural processes, SC is a linear transformation of the adjacency matrix, which may result in inaccurate values.
We present a framework based on deep neural networks for calculating integrated information at a lower cost than the full-search
technique. Our method consists of three parts, 1) calculation of adjacency matrix, 2) deep neural network-based autoencoder for node
embedding, and 3) clustering each node for cutting.
We analyzed chimpanzee's ECoG high dimensional time-series data, and compared the result with the conventional methods. As a result,
the proposed method calculated better values in 95% of the datasets. We also compared the clustering results with the
electroencephalographic coordinates and found that the clustering results generally corresponded to the electrode positions. These results
indicate that the proposed method is useful for integrated information estimation.Session 2: Data Model and Intelligent Computing
4:35pm-4:50pm 4:50pm-5:05pm
Set-based Particle Swarm Optimization for Data Identifying the Best Combination of Crossover and
SI016 Clustering SI004-A Mutation Operators in NSGA-II for Redundancy-
based Optimal Design of Water Network
Lienke Brown
Jaehyun Kim
Stellenbosch University, South Africa
Korea University, Korea
Abstract—Computational intelligence approaches to data clustering have been Abstract—The Non-dominated Sorting Genetic Algorithm II (NSGA-II) is a fast sorting
successful in producing compact and well-separated clusters. In particular, particle and elite optimization algorithm to find a set of the optimal solutions of multi-objective
swarm optimization (PSO) is deemed an effective approach to data clustering. This problem. Outstanding nature of the algorithm is to simultaneously achieve the diversity
paper develops and evaluates a discrete-valued variation of PSO, namely the set-based and convergence of the final Pareto-optimal solutions. Since the initial introduction, a
PSO (SBPSO) algorithm, to cluster data. The SBPSO algorithm is evaluated on six variety of operators have been employed for crossover (e.g. single-point crossover, N-
standard data sets and nine artificially generated data sets. The clustering results of the point crossover, and simulated binary crossover) and the mutation (e.g. single-point
SBPSO algorithm is compared to the performance of established clustering algorithms mutation, adaptive mutation, and polynomial mutation) of NSGA-II. The best crossover
and a PSO clustering algorithm. It is concluded that the results of the SBPSO algorithm and mutation operators would differ for different problems. In this study, we investigate
varies with the data set characteristics. Nonetheless, the SBPSO is deemed a successful the performances of NSGA-II with several different combinations of mutation and
approach to clustering data. crossover operators in a multi-objective design of water distribution system. Two
objectives considered are to minimize the total construction cost (Objective 1) and
maximize the system redundancy (Objective 2). Anytown network is used for the
redundancy based multi-objective optimal design in this paper. The results and
discussions in this study can be referred when selecting the proper operators for similar
engineering optimization problems.Session 2: Data Model and Intelligent Computing
5:05pm-5:20pm 5:20pm-5:35pm
Static Polynomial Approximation Using Set-based AxDFM:Position prediction system based on the
SI017 Particle Swarm Optimisation SI018 importance of high-order features
AP Engelbrecht Haoxiang Feng
Stellenbosch University, South Africa Chongqing University of Posts and Telecommunications,
China
Abstract—Recently, a set-based particle swarm optimisation (SBPSO) algorithm was Abstract—The exploration and combination of high-level features is crucial for many
developed to find optimal polynomials for univariate polynomial approximation machine learning tasks. At the same time, we cannot ignore the different importance of
problems. This SBPSO algorithm employed a computational costly adaptive coordinate high-level features. In traditional machine learning predictive models, analyzing and
descent (ACD) algorithm to find optimal monomial coefficients. In addition, the ACD combining the original data and manually making these features will undoubtedly increase
algorithm prematurely converged in coefficient space. This paper presents a variation of the complexity and cost of the system. The emergence of factorization machines can use
the SBPSO polynomial approximation algorithm where the ACD algorithm is replaced the vector product to represent the interaction of features, and automatically learn features
with a standard particle swarm optimisation (PSO) algorithm, which is applied to find The combination of to get high-order feature interactions not only reduces the complexity
optimal monomial coefficients only after an optimal polynomial architecture has been of the system, but also increases the diversity of high-order features. We refer to the depth
found. This results in a significant reduction in computational costs and prevents factorization machine (xDeepFM) to generate high-level feature interactions at the display
premature stagnation in coefficient space. The results show that the new SBPSO mode and vector level, and The importance of different features is dynamically learned
algorithm for polynomial approximation performs well on univariate, static polynomial through the squeeze-incentive (SENET) mechanism, and different weights are used for
approximation problems. interaction.Then, use the attention mechanism to extract the importance of the obtained
high-order features and assign weights, and finally get the prediction classification through
the fully connected layer. We further summarized these methods into a unified model, and
named the model the Advanced Attention Depth Factorization Machine (AxDFM).Session 2: Data Model and Intelligent Computing
5:35pm-5:50pm
Estimating Dynamical Nonlinear System with Nonstationarity by
SI026-A Gaussian Process Self-Organizing Generalized State-Space Model
Takashi Terayama
Kobe University, Japan
Abstract—Elucidating the dynamic systems behind time series data is an important task in time series analysis. In
particular, many dynamical systems have nonlinearity and nonstationarity in the nonlinear dynamics of latent
variables. Therefore, it is important to establish a method to estimate nonlinear and nonstationary dynamics from
observed time series data.
Algorithms for estimating dynamical systems for nonlinear and nonstationary time series data have been proposed
separately for either nonlinearity or nonstationarity. For nonstationary time series data, algorithms have been
proposed to estimate nonstationary dynamic systems by introducing dynamically varying system noise. However,
these previous studies assume parametric estimation where the dynamics is assumed to be known. However, these
previous studies assume that the dynamics are known, which makes it difficult to accurately estimate the dynamics in
realistic situations where the nonlinear dynamics of the time series data are unknown. On the other hand, some
algorithms have been proposed for nonparametric estimation of dynamical systems for nonlinear time series data,
but they assume stationary time series data and it is difficult to estimate nonstationary systems. Thus, a method for
nonparametric estimation of nonlinear and nonstationary dynamic systems and simultaneous estimation of
dynamically varying noise components has not been established, and there is a need to construct an algorithm for
this purpose.Session 2: Data Model and Intelligent Computing
5:50pm-6:05pm 6:05pm-6:20pm
Stability-Guided Multi-Guide Particle Swarm Application of Hybrid PSO and SQP Algorithm in
SI015 Optimization Algorithm SI002 Optimization of the Retardance of Citrate Coated
Ferrofluids
AP Engelbrecht
Jing-Fung Lin
Stellenbosch University, South Africa
Far East University, Taiwan
Abstract—This paper proposes a multi-guide particle swarm optimization (MGPSO) Abstract—The citrate (citric acid, CA) coated ferrofluids with great magneto-optical
algorithm which does not require tuning of its control parameters. Control parameter retardance can meet the high magnetic responsive demand, especially in widely potential
values are randomly sampled to satisfy theoretically derived stability conditions, biomedical applications such as hyperthermia and magnetic resonance imaging. In this
eliminating the need for computatinally expensive parameter tuning. In addition, the study, the measured retardances are based on the Taguchi method with nine tests for four
feasibility of utilizing dynamically decreasing tournament sizes in the selection of the parameters, including pH of suspension, molar ratio of CA to Fe3O4, CA volume, and
archive guide, aswell as a ring neighbourhood topology, is investigated. The results coating temperature. The retardance obtained from the double centrifugation test is also
show that random control parameter sampling is a viable alternative to static tuning, included. Three optimization algorithms including the particle swarm optimization (PSO),
most notably when applied to higher numbers of objectives. However, the results show the sequential quadratic programming (SQP), and a hybrid PSO-SQP algorithm are
no clear benefit or detriment to utilizing dynamic tournament selection sizes and ring executed to obtain high retardance. The comparisons are made among the retardance
neighbourhood topologies. results obtained from these algorithms. Seven start points chosen from the orthogonal test
are input into the SQP, the PSO is applied to the stepwise regression equation, and while
executing the hybrid PSO-SQP algorithm, the parametric combination obtained by the PSO
is adopted as the start point in the SQP simulation. The global optimum retardance and the
corresponding parameter values are effectively assured by the global search ability of the
PSO and the local search ability of the SQP.Session Chair: Prof Thomas Hanne
University of Applied Sciences and Arts Northwestern Switzerland, Switzerland S3
Algorithm Design
Note: and Optimization
Greenwich Mean Time (GMT+9) – Seoul Local Time
Time: 4:20pm-6:20pm, April 10, 2022
Meeting ID: 878 2808 4933
Please enter the room 5-10 minutes earlierSession 3: Algorithm Design and Optimization
4:20pm-4:35pm 4:35pm-4:50pm
Coevolutionary Algorithm for Evolving Competitive Understanding the Effects of Ant Algorithms on Path
SI008 Strategies in the Weapon Target Assignment Problem SI005 Planning with Gain-Ant Colony Optimization
Ehab Elfeky V Sangeetha
University of New South Wales, Australia Amrita School of Engineering, India
Abstract—This paper considers a non-cooperative real-time strategy game between Abstract—With the advent of more automated and unmanned systems, there is an
two teams; each has multiple homogeneous players with identical capabilities. In increasing need for path planners. Intelligent path planners play an important role in the
particular, the first team consists of multiple land vehicles under attack by a team of navigation of automated systems. In this work, the performance of an enhanced gain-ant
drones, and the vehicles are equipped with weapons to counterattack the drones. colony optimization has been tested with the most popularly used ant algorithms – Ant
However, with the increase in the number of drones, it may become difficult for human system, Ant colony system and Min-Max ant system in the application of path planning.
operators to coordinate actions across vehicles in a timely manner. Therefore, we The pheromone update mechanism of traditional ant metaheuristic is enhanced with a
explore a coevolutionary approach to simultaneously evolve competitive weapon target local optimization mechanism and simulated with popular ant algorithms for an efficient
assignment strategies for the land vehicles and drone threats to address this problem. choice of update rule. Evaluation is done using performance measures like path length
Different scenarios involving a different number of land vehicles and drone threats have and computation time taken. The results are statistically verified and analyzed. Path
been considered to evaluate the performance of the proposed approach. Results planned by proposed algorithm was found to be 3.25% shorter than existing algorithms.
showed some advantages of applying such a coevolutionary approach.Session 3: Algorithm Design and Optimization
4:50pm-5:05pm 5:05pm-5:20pm
Genetic Algorithm with Machine Learning to Resource Prediction of Virtual Network Function Based on
SI006 Estimate the Optimal Objective Function Values SI028 Traffic Feature Extraction
of Subproblems
Ya Tan
Yohei Hazama Chongqing University of Posts and Telecommunications, China
Kyoto Institute of Technology, Japan
Abstract—With the continuous innovation of the Internet, the development of Cloud Computing
Abstract—This paper addresses an optimization problem with two decision technology and standard server promotes the development of Network Function Virtualization (NFV).
variable vectors. This problem can be divided into multiple subproblems Although NFV solves the shortcomings of traditional network function equipment such as high cost
when an arbitrary value is given to the first decision variable vector. In and difficult operation, it also brings certain challenges. Resource management in NFV is a complex
conventional genetic algorithms (GAs) for the problem, an individual is often problem because the resource requirements of Virtual Network Function (VNF) vary with the dynamic
expressed by the value of the first decision variable vector. In evaluating the traffic, so it is necessary to understand the resource requirements of VNF. Due to the limited physical
individual, the value of the remaining decision variable vector is determined network resources, it is very important to find an effective resource prediction method. Based on
by metaheuristics or greedy algorithms. However, such GAs are time- Heterogeneous Information Network (HIN) and Multilayer Perceptron (MLP), we propose VNF-RPHIN,
consuming or not general-purpose. We propose a GA with a neural network a method of the VNF resource requirement prediction based on traffic feature extraction. Firstly, we
model to estimate the optimal objective function values of the subproblems. construct the HIN by the correlation between traffic features. Secondly, we use the HIN2Vec model to
Experimental results compared to other GAs show that the proposed method obtain the feature representation of each traffic feature. Finally, the attention mechanism is used to
is effective. measure the importance of each feature, and different weights are assigned to each feature, and then
they are input into the MLP model. The hidden relationship between traffic features is mined by HIN
to predict the resource requirement of the VNF. The experimental results show that the proposed
method has good performance and is superior to the traditional machine learning model and
common deep learning model.Session 3: Algorithm Design and Optimization
5:20pm-5:35pm 5:35pm-5:50pm
Assessing the Quality of Car Racing Controllers in a Unsupervised Deep Video Interpolation Based on Spatio-
Virtual Setting Under Changed Conditions SI025-A Temporal Autoregressive Neural Network
SI007
Thomas Hanne Koki Nakashima
University of Applied Sciences and Arts Northwestern Kobe University, Japan
Switzerland, Switzerland
Abstract—In recent years, methods that apply machine learning techniques to temporal
Abstract—This paper discusses several controllers based on fuzzy logic and super-resolution have been proposed. Temporal super-resolution aims to improve the
evolutionary concepts applied to a car racing simulation and their robustness to frame rate by interpolating new images between the images. By using machine learning
changing physics of the cars. The challenge is to design a car controller that passes the for temporal super-resolution, it is becoming possible to develop frame interpolation
next three arising waypoints faster than an opponent car controller. Two fuzzy techniques that can deal with complex motion in video. However, these previous
controllers are compared to two evolutionary optimized controllers in solo races as well methods assume that there is a high-frame-rate video image corresponding to the
as in head-to-head competitions, where all controllers compete head-to-head against supervisory data, and may not be able to deal with videos without supervisory data.
the other controllers. The influence of some parameter settings is investigated as well. In this study, we propose a unsupervised temporal super-resolution technique that can
The results emphasize the robustness of the fuzzy controllers, not differing much from cope with video without supervisory data. By applying the spatio-temporal
each other. Overall, the fuzzy controllers perform better with different parameter autoregressive model twice, we introduce an extended model in which the low frame rate
settings of the driving physics except when reverse speed is equal to forward speed. video itself is used as the substantial supervisory data. In particular, we introduce a
spatio-temporal autoregressive model based on deep learning. In other words, we
propose a video temporal super-resolution technique that uses a convolutional neural
network, which has shown high performance in areas such as image analysis, and a
network that utilizes a spatio-temporal 3D convolutional…Session 3: Algorithm Design and Optimization
5:50pm-6:05pm 6:05pm-6:20pm
Robotic Path Planning by Q Learning and a Performance A Hybrid Multi-Objective Teaching Learning-Based
Comparison with Classical Path Finding Algorithms SI010 Optimization Using Reference Points and R2 Indicator
SI003
Thomas Hanne Farajollah Tahernezhad-Javazm
University of Applied Sciences and Arts Northwestern Ulster University, UK
Switzerland, Switzerland
Abstract—Hybrid multi-objective evolutionary algorithms have recently become a hot
Abstract—Q Learning is a form of reinforcement learning for path finding problems topic in the domain of metaheuristics. Introducing new algorithms that inherit other
that does not require a model of the environment. It allows the agent to explore the algorithms’ operators and structures can improve the performance of the algorithm.
given environment and the learning is achieved by maximizing the rewards for the set Here, we proposed a hybrid multi-objective algorithm based on the operators of the
of actions it takes. In the recent times, Q Learning approaches have proven to be genetic algorithm (GA) and teaching learning-based optimization (TlBO) and the
successful in various applications ranging from navigation systems to video games. This structures of reference point-based (from NSGA-III) and R2 indicators methods. The new
paper proposes a Q learning based method that supports path planning for robots. The algorithm (R2-HMTLBO) improves diversity and convergence by using NSGA-III and R2-
paper also discusses the choice of parameter values and suggests optimized based TLBO, respectively. Also, to enhance the algorithm performance, an elite archive is
parameters when using such a method. The performance of the most popular path proposed. The proposed multi-objective algorithm is evaluated on 19 benchmark test
finding algorithms such as A* and Dijkstra algorithm have been compared to the Q problems and compared to four state-of-the-art algorithms. IGD metric is applied for
learning approach and were able to outperform Q learning with respect to computation compression, and the results reveal that the proposed R2-HMTLBO outperforms MOEA/D,
time and resulting path length. MOMBI-II, and MOEA/IGD-NS significantly in 16/19 tests, 14/19 tests and 13/19 tests,
respectively. Furthermore, R2-HMTLBO obtained considerably better results compared to
all other algorithms in 4 test problems, although it does not outperform NSGA-III on a
number of tests.01
07
Listener
The meeting invitation link: https://us06web.zoom.us/j/81527236343 & https://us06web.zoom.us/j/87828084933Listener
Danlord Ancheta
Listener 1
Eulogio “Amang” Rodriguez Institute of Science
and Technology, PhilippinesISMSI 2023
2023 7th International Conference on
Intelligent Systems, Metaheuristics An annual international event of
India International Congress on
& Swarm Intelligence Computational Intelligence (http://www.iicci.in)
Kuala Lumpur, Malaysia
April 23-24
KEYNOTE SPEAKERS
Prof. Yaroslav D. Sergeyev
Distinguished Professor, University of
Calabria, Italy
PUBLICATIONS Prof. Andries Engelbrecht
Accepted Papers after registration and presenta- Stellenbosch University, South Africa
tion at ISMSI 2023 will be published in the Inter-
national Conference Proceedings, which will be
indexed by EI Compendex, Scopus, and some Prof. Shengxiang Yang,
major databases. De Montfort University, UK
A special issue of Neural Computing & Applica-
tions, a Springer Publication [SCIE indexed, SUBMISSION METHOD
2020 lmpact Factor: 5.606, 5 Year Impact Factor: Please log in the Electronic Submission System; (
5.573; ISSN: 0941-0643 (print version) ISSN: .pdf only) to submit your full paper or abstract. For any
1433-3058 (electronic version)], will publish a inquiry about the conference, please feel free to
selected set of extended versions of ISMSI23 contact us at: sub@ismsi.org
papers (to be shortlisted after the conference),
after the usual reviewing of those papers.
ABOUT KUALA LUMPUR
Kuala Lumpur is a federal territory and the capital city
IMPORTANT DATES of Malaysia. It is the largest city in Malaysia. It is
Submission Deadline November 20, 2022 among the fastest growing metropolitan regions in
Southeast Asia, both in population and economic
Acceptance Notification December 15, 2022 development. Since the 1990s, the city has played
host to many international sporting, political and
Camera-Ready Paper Due January 5, 2023 cultural events including the 1998 Commonwealth
Games and the 2017 Southeast Asian Games. Kuala
Conference Dates April 23-24, 2023
Lumpur has undergone rapid development in recent
decades and is home to the tallest twin buildings in
the world, the Petronas Towers, which have since
Organized by Technically Sponsored by EU/ME,
become an iconic symbol of Malaysian development.
the Euro Working Group on Metaheuristics
Kuala Lumpur is one of the leading cities in the world
for tourism and shopping, the 6th most-visited city in
the world in 2019. The city houses three of the world's
ten largest shopping malls.
Conference Secretary of ISMSI: Ms. Nancy Liu; Email: sub@ismsi.org; Tel: +86-13709044746 CONTACT USYou can also read
