Invited Speakers
Plenary 1: Prof. Dr. Sushmita Mitra, Indian Statistical Institute, Kolkata, India
Plenary 2: Prof. Dr. Bassem Jarboui, Sfax University, Tunisia
Plenary 3: Prof. Dr. Tzung-Pei Hong, National University of Kaohsiung, Taiwan
Plenary 4: Prof. Dr. André Rossi, Université Paris-Dauphine, France
Plenary 5: Prof. Dr. Amine Nait-Ali, University Paris-Est Créteil, France
Plenary 6, Dr. Rudy A. Oude Vrielink, University of Twente & TwiQel, Netherlands
Plenary 7: Prof. Dr. Frederico Gadelha Guimar?es, Federal University of Minas Gerais, Brazil
Plenary 8: Prof. Dr. Fakhri Karray, University of Waterloo, Canada
Plenary 9: Prof. Dr. Saman K. Halgamuge, University of Melbourne, Australia
Plenary 10: Prof. Dr. Salvador García, University of Granada, Granada, Spain
Plenary 11: Prof. Dr. Vincenzo Piuri, Università degli Studi di Milano, Italy
Plenary 12: Prof. Dr. Dries F. Benoit, Ghent University, Belgium
Plenary 13: Prof. Dr. Dijiang Huang, Arizona State University, USA
Plenary 14: Prof. Dr.Ketan Kotecha, Symbiosis Institute of Technology, India
Plenary 15: Prof. Dr. Sheela Ramanna, University of Winnipeg, Canada
Prof. Dr. Sushmita Mitra
Machine Intelligence Unit (MIU),
Indian Statistical Institute, Kolkata
Title: Intelligent Analysis of Brain Images
Abstract: Medical imaging inherently entails imperfection, and is therefore an appropriate domain for involving computational intelligence. We introduce the concepts of quantitative imaging and radiomics, followed by radiogenomics and deep learning. Next we describe an automated and fast detection algorithm for brain tumor in MRI, and its efficient segmentation both in two- and three-dimensions. Visual saliency is utilized for a fast localization and detection of the tumor. Just a single user-provided seed for efficient delineation of the GBM tumor is also elaborated. The second part of the talk focuses on the application of deep learning for the detection, segmentation and survival analysis in brain tumors. Comparative study with related machine learning algorithms demonstrates its effectiveness on medical image data, both quantitatively and qualitatively.
Biography: Sushmita Mitra is full Professor at the Machine Intelligence Unit (MIU), Indian Statistical Institute, Kolkata. From 1992 to 1994 she was in the RWTH, Aachen, Germany as a DAAD Fellow. She was a Visiting Professor in the Computer Science Departments of the University of Alberta, Edmonton, Canada in 2004, 2007; Meiji University, Japan in 1999, 2004, 2005, 2007; and Aalborg University Esbjerg, Denmark in 2002, 2003. Dr Mitra received the National Talent Search Scholarship from NCERT, India, the University Gold Medal in 1988, the IEEE TNN Outstanding Paper Award in 1994 for her pioneering work in neuro-fuzzy computing, the CIMPA-INRIA-UNESCO Fellowship in 1996, and Fulbright-Nehru Senior Research Fellowship in 2018-2020. She is the author of the books ``Neuro-Fuzzy Pattern Recognition: Methods in Soft Computing" and ``Data Mining: Multimedia, Soft Computing, and Bioinformatics" published by John Wiley, and ``Introduction to Machine Learning and Bioinformatics", Chapman & Hall/CRC Press, beside a host of other edited books. Dr Mitra has guest-edited special issues of several journals, is an Associate Editor of ``IEEE/ACM Trans. on Computational Biology and Bioinformatics", ``Information Sciences", ``Fundamenta Informatica", and is a Founding Associate Editor of ``Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery (WIRE DMKD)". She has more than 150 research publications in refereed international journals. Dr Mitra is a Fellow of the IEEE, Indian National Science Academy (INSA), International Association for Pattern Recognition (IAPR), and Fellow of the Indian National Academy of Engineering (INAE) and The National Academy of Sciences, India (NASI). She is an IEEE CIS Distinguished Lecturer and the current Chair-Elect, IEEE Kolkata Section. She has visited more than 30 countries as a Plenary/Invited Speaker or an academic visitor, and served as General Chair and/or Program Chair of several international conferences. Her current research interests include data science, deep learning, soft computing, medical image processing, and Bioinformatics.
Prof. Dr. Bassem Jarboui
Professor of Operational Research
Sfax University, Tunisia
Higher Colleges of Technology, UAE
Title : Metaheuristics from optimization to machine learning
Abstract: Metaheuristics are well-known techniques for solving hard optimization problems. In addition, using metaheuristics as search strategies for the automatic generation of computer programs is becoming a growing research area. The interest in this direction has been initially promoted by the genetic programming paradigm. Afterward, other metaheuristic algorithms, especially those of population-based type, have been extended in order to produce adequate programs for a given problem.
In the first part of the presentation, I will discuss possible ways of using metaheuristic optimization techniques for solving automatic programming problems. In the second part, I will talk about a relatively new technique called Variable Neighborhood Programming (VNP) that was inspired by the principles of Variable Neighborhood Search metaheuristic. The VNP methodology and its variants will be illustrated on Symbolic regression, Classification and prediction problems.
Biography: Professor Bassem Jarboui is a full Professor of Operational Research at Sfax University, Tunisia, where he also got his Ph.D. degree. Currently, he is working at Higher Colleges of Technology, Abu Dhabi, UAE.
He had edited 7 books, and two special journal issues. He also organized and chaired five international conferences. He published over 140 scientific papers, including books, academic journals, edited proceedings and book chapters. According to Google scholar, his current h index is 22.
Prof. Dr. Tzung-Pei Hong
National University of Kaohsiung,
Taiwan
Title: Some Interesting Issues in Data Mining
Abstract: Data mining discovers useful rules or patterns from a set of data. Its role is more and more important in the era of big data because it can provide compact and relevant knowledge instead of detailed data to users. In this speech, I would like to talk some interesting issues about data mining, including fuzzy data mining, incremental data mining, duality of data mining, knowledge warehouse, and so on.
Biography: Tzung-Pei Hong received his Ph.D. degree in computer science and information engineering from National Chiao-Tung University in 1992. He served as the first director of the library and computer center, the Dean of Academic Affairs and the Vice President in National University of Kaohsiung, Taiwan. He is currently a chair and distinguished professor at the Department of Computer Science and Information Engineering in NUK, and a joint professor at the Department of Computer Science and Engineering, National Sun Yat-sen University, Taiwan. He got the first national flexible wage award from Ministry of Education in Taiwan. He has published more than 600 research papers in international/national journals and conferences. His current research interests include knowledge engineering, data mining, soft computing, and management information systems.
Prof. André Rossi
Université Paris-Dauphine,
France
Title: Matheuristics: Taking the Challenge of Hard Operational Research Problems by Mixing Exact and Approximate Algorithms.
Abstract: This talk aims at providing an introduction to matheuristics, which are combinations of exact and approximate algorithms. These algorithmic strategies can take various forms, and three of them are presented on case studies. We first show how a local search algorithm can benefit from an exact approach to address the multi-dimensional knapsack problem. Second, we present the problem of maximizing the lifetime of a wireless sensor network, and we address it with a column generation algorithm whose performances are boosted by a genetic algorithm. Finally, we show a win-win collaboration of a tabu search with an integer linear program to address the problem of minimizing the number of branch vertices in a tree, which is a problem that arises in the context of multicast fiber optic network design.
Biography: André Rossi holds an engineering degree in Automatic Control and Production Sciences from Institut National Polytechnique de Grenoble, France, and a PhD degree in Robust Scheduling from the same institution. He is currently full professor of Computer Sciences at LAMSADE, Université Paris-Dauphine, France, since 2018. Before that, he was professor of Operations Research at the Computer Sciences Department of Université d'Angers, France, from 2015 to 2018. Previously, he was Associate Professor at Université de Bretagne-Sud in Lorient, France, from 2005 to 2015 and served as an invited professor in the Department of Computer and Information Sciences in University of Hyderabad, India, from 2009 to 2014. He is now co-director of the Graduate School of Computer Sciences of PSL University, and is a researcher at LAMSADE laboratory in Université Paris-Dauphine.
- Preferred date for the plenary talk:
14 December: 09:00 GMT
Prof. Amine Nait-Ali
University Paris-Est Créteil (UPEC)
France
Title: Facial biometrics: human being perception vs. artificial intelligence, is it a competition?
Abstract: Biometrics is known to be a discipline dealing with the extraction of physical or behavioural characteristics from the human body. Although facial biometrics is commonly associated with the identification and verification of individuals, this presentation addresses a collection of applications highlighting new challenges in forensics, and healthcare and wellness. Specifically, we consider skin health analysis, facial paralysis, facial ageing process, as well as other promising biometric applications. The purpose is to analyse and to evaluate the ‘roadmap’ of computer vision, and more specifically Artificial Intelligence (AI). Nonetheless, AI is perceived by some as the future, and by others as a serious threat. One may raise the question, Is there a real competition?
Abstract: Amine NAIT-ALI is a full Professor at the University of Paris-Est Créteil (UPEC). He obtained his Ph.D. degree in 1998, in Biosignal processing and obtained the “Habilitation à Diriger des Recherches” (HDR) from the University Paris XII, in 2007. His research interests are focused on biosignal processing, biometrics, medical signal and image compression, and artificial intelligence applications. He has co-authored hundreds of international peer-reviewed papers and edited and co-edited seven books in the biomedical engineering and biometrics field (Springer, ISTE-Wiley and Hermes). He has organized and run several national, european and international workshops. He is currently, the General Chair of the International Conference of Bio-engineering for smart technologies (BioSMART). He is also a member of: IEEE, SFGBM, GDR ISIS and STIC-Santé. Prof. Amine NAIT-ALI has served as a representative of IEEE Biomedical Engineering Society in IEEE Biometric Council.
14 December: 14:00 GMT
Rudy A. Oude Vrielink
University of Twente & Company TwiQel
Faculty of Behavioural, Management & Social Sciences
Department of Industrial Engineering & Business Information Systems
Enschede, Netherlands
Title: Adaptive scheduling: towards a living smart campus
Abstract: Capacity shortages are a common threat. This is also the case at universities in the Netherlands. After manually counting the number of students present during lectures, it is striking how big the deviation is between what is planned and what is actually in the classroom. That seems a threat, because housing and m2 are so expensive, but at the same time it is also an opportunity. As a public institution you can try to use the deviations, by adaptive scheduling. You can choose to increase efficiency by allocating smaller classrooms in consecutive weeks or by emptying entire buildings of education. You can also choose to support the effectiveness of education, by focusing the timetable more on community building by students, or by allocating more suitable classrooms more often. How you can measure these indicators of efficiency and effectiveness is not easy, but once they can be measured, they can be continuously improved, leading to a ever better timetables and a flexible organisation of education. Also now, during the covid-19 pandemic, it is important to make optimal use of the available spaces.
Biography: Rudy Oude Vrielink is a PhD student at the University of Twente in the field of education logistics. He is the inventor of adaptive timetabling, which continuously adapts student timetables to changing circumstances. For example, during the course of a quartile, student attendance often changes, which opens the possibility of allocation smaller or larger educational spaces. The goal of this is to increase occupancy and utilization. It may also be that a different educational space is more suitable than the one in which the lesson is planned. This can increase the effectiveness of the education; all through the implementation of adaptive timetabling. In 2019 Rudy founded the company TwiQel (www.twiqel.nl <http://www.twiqel.nl/>) together with 2 partners. Twiqel focuses on increasing both the efficiency and the effectiveness in the use of spaces.
Prof. Frederico Gadelha Guimarães
Federal University of Minas Gerais
Brazil
Title: New developments in fuzzy time series methods
Abstract: Fuzzy Time Series (FTS), introduced in the early 1990s by Song and Chissom, can be viewed as a way to represent time series from the perspective of fuzzy logic. Numerical data is translated into fuzzy sets generating a fuzzy representation of the time series. Then, temporal patterns are extracted according to the number of past observations (lags) that are considered in the model to produce a rule-based knowledge base. The knowledge base is the model that is then used for forecasting. In this talk, I will review the latest developments in FTS methods, including interval and probabilistic forecasting, multivariate forecasting, adaptive and evolving models and hybrid methods. I will also present the pyFTS library, an open source Python library for FTS developed in the MINDS laboratory.
Biography: Frederico Gadelha Guimarães received his B.Eng. and M.Sc. degrees in electrical engineering from the Federal University of Minas Gerais (UFMG), Belo Horizonte, Brazil, in 2003 and 2004, respectively. He received a Ph.D. degree in Electrical Engineering from the Federal University of Minas Gerais in 2008, with a one-year visiting student scholarship at McGill University, Montreal, Canada (2006–2007). He had postdoctoral fellowship (2017–2018) at the Laboratoire Images, Signaux et Systèmes Intelligents (LiSSi), linked to the Université Paris-Est Créteil (UPEC), Paris, France. In 2010, he joined the Department of Electrical Engineering, UFMG, and in 2018 he became an Associate Professor. He has been responsible for the Machine Intelligence and Data Science Laboratory (MINDS) for computational intelligence research since 2014. Dr. Guimarães has published more than 200 papers in journals, congresses and chapters of national and international books. He has experience in Electrical Engineering and Computer Engineering, with emphasis on optimization, computational intelligence, machine learning, genetic algorithms and evolutionary computation. He is a Senior member of the IEEE Computational Intelligence Society (CIS) and the IEEE Systems, Man, and Cybernetics Society (SMCS).
15 December: 14:00 GMT
Prof. Fakhri Karray
University of Waterloo
Canada
Title: Recent Advances in Operational Artificial Intelligence
Abstract: The talk presents an overview on the origins and the major advances accomplished lately in the field of Artificial Intelligence (AI) and specifically Operational Artificial Intelligence. Future directions are also highlighted. As demonstrated by the significant accomplishments made in this field, many are heralding a new technological era that may well correspond to the dawn of the Fourth Industrial Revolution. It is expected that AI will grow the world ‘s GDP by 15% by 2025. This amounts to more than 15 Trillion dollars per year of growth. We are indeed on the cusp of revolutionary technological developments fuelled by advances made in the field of machine learning and artificial intelligence. These developments have impacted other technological innovations made in the field of Internet of Things, self-driving machines, virtual assistants, human machine intelligent interface, natural language and speech understanding, cognitive robotics, virtual care systems, eHealth and Fintech, to name a few. Although AI constitutes an umbrella of several interrelated technologies, all of which are aimed at imitating to a certain degree intelligent human behavior or decision making, deep learning algorithms are considered to be the driving force behind the explosive growth of AI and their applications in almost every scientific and technological sector: disease diagnosis, remote health care monitoring, financial market prediction, self-driving vehicles, social robots with cognitive skills, intelligent manufacturing, surveillance, cybersecurity, intelligent transportation systems, to name a few. The talk highlights the milestones that led to the current growth in AI, discusses some of the major achievements in the field of Operational AI, future directions and enumerates challenges in making the field of AI strictly useful to humanity and safe for society.
Biography: Fakhri Karray is the University Research Chair Professor in Electrical and Computer Engineering and the co-director of the Institute of Artificial Intelligence at the University of Waterloo. He holds the Loblaw’s Research Chair in Artificial Intelligence. Dr. Karray’s research work spans the areas of intelligent systems and operational artificial intelligence as applied to autonomous machines/devices and man machine interaction systems through speech, gesture, and natural language. He has authored extensively in these areas and has disseminated his work in journals, conference proceedings, and textbooks. He is the co-author of two dozen US patents, has chaired/co-chaired several international conferences in his area of expertise and has served as keynote/plenary speaker on numerous occasions. He has also served as the associate editor/guest editor for a variety of leading journals in the field, including the IEEE Transactions on Cybernetics, the IEEE Transactions on Neural Networks and Learning Systems, the IEEE Transactions on Mechatronics, the IEEE Computational Intelligence Magazine. His work has been featured on Discovery Channel, CBC, Globe and Mail, The Record, Reuters, the Daily Mail, Washington Post, Wired Magazine, and DigitalTrends portals. He has served as the University of Waterloo’s Academic Advisor for Amazon’s Alexa Fund Fellowship Program and is a Fellow of the Canadian Academy of Engineering, a Fellow of the Engineering Institute of Canada and a Fellow of the IEEE.
Prof Saman K. Halgamuge,
University of Melbourne, Australia
Fellow of IEEE and Distinguished Lecturer of IEEE Computational Intelligence Society.
Title: Dancing to the SONG: Dynamic Data Visualization with Self-Organizing Nebulous Growths
Abstract: Visualization of high dimensional data attempts to compensate for the limitation of human perception of high dimensions through mapping higher dimensions into two or three dimensions while keeping the loss of the mapping to a minimum. Nonparametric dimensionality reduction techniques, such as t-distributed Stochastic Neighbor
Embedding (t-SNE) and uniform manifold approximation and projection (UMAP), are proficient in visualizing data sets of fixed sizes. However, they cannot incrementally map and insert new data points into already existing data visualizations. Self-Organizing Nebulous Growths (SONG), a parametric nonlinear dimensionality reduction technique that supports incremental data visualization, i.e., incremental addition of new data while preserving the
structure of the existing visualization [1] is a new method that allows us to observe patterns “dancing” as the data are continuously added, for example, in case of acquiring epidemiological data in a pandemic, collecting electrical signals from “min-brains” in a wet lab [2], obtaining continuous signals from a solar assisted ground source heat pump [3], recording dynamic temperature profiles and the carbon foot print of data centres [4] or in observing the
gradual and persistent destruction of the environment by countries with less capable leaders. SONG is capable of handling new data increments, no matter whether they are similar or heterogeneous to the already observed data distribution. We test SONG on a variety of real and simulated data sets. The results show that SONG is superior to Parametric t-SNE, t-SNE, and UMAP.
[1] D Senanayake, W Wang, SH Naik, S Halgamuge. “Self Organizing Nebulous Growths for Robust and Incremental Data Visualization”, IEEE Transactions on Neural Networks and Learning Systems, 2020
[2] GDC Mendis, G Berecki, E Morrisroe, S Pachernegg, M Li, M Varney, PB Osborne, CA Reid, S Halgamuge, S Petrou, “Discovering the pharmacodynamics of conolidine and cannabidiol using a cultured neuronal network based workflow”, Scientific reports, Nature Publishing Group 9 (1), 121, 2019
[3] H Weeratunge, G Narsilio, J de Hoog, S Dunstall, S Halgamuge. “Model predictive control for a solar assisted ground source heat pump system”, Energy, Elsevier 152, 974-984, 2018
[4] J Siriwardana, SK Halgamuge, T Scherer, W Schott, “Minimizing the thermal impact of computing equipment upgrades in data centers”, Energy and Buildings 50, 81-92, 2012
Biography: Prof Saman Halgamuge received the B.Sc. Engineering degree in Electronics and Telecommunication from the University of Moratuwa, Sri Lanka, and the Dipl.-Ing and Ph.D. degrees in data engineering from the Technical University of Darmstadt, Germany. He is currently a Professor of the Department of Mechanical Engineering of the School of Electrical Mechanical and Infrastructure Engineering, The University of Melbourne, Australia. He is a Fellow of IEEE (2017-), a distinguished Lecturer of IEEE Computational Intelligence Society (2018-21) and listed as a top 2% most cited researcher for AI and Image Processing in Stanford database (2020-). His research interests are in AI, machine learning including deep learning, optimization, big data analytics and their applications in energy, mechatronics, bioinformatics and neural engineering. He graduated 45 PhD students at University of Melbourne. He is currently an honorary Professor of multiple institutions including ANU in Canberra and ITB in Bandung and a distinguished visiting professor of HEBUT, China. He has also held leadership roles including Director of Melbourne India Postgraduate program and Associate Dean International at University of Melbourne and Head of Engineering School at Australian National University.
Prof. Dr. Salvador García
University of Granada, Granada,
Spain
Title: Singular and Non-Standard Tasks in Supervised Machine Learning
Abstract: In recent years, new tasks and paradigms have appeared in the field of machine learning which, in some way, do not fit into the classical representation in supervised learning (input data vector, output value), or which do not follow the conventional dynamics of classification and/or regression paradigms.
Thus, as far as non-standard representations are concerned, we have examples such as multi-instance learning (MIL), in which each pattern is represented by a bag containing a variable number of instances, each instance having the same number of attributes. Moreover, the paradigms of multi-output learning present the output space in a more flexible way, namely with a vector that represents several simultaneous outputs, which correspond to behaviours or functionalities that are learned concurrently. In relation to tasks that do not fit into the classical dynamics of the learning process, there are tasks with semi-supervised learning, where both labeled and unlabeled data sets are used to carry out more effective learning. On the other hand, in monotonic ordinal classification, the target variable is not of a numerical type, but has an implicit ordering, which can be used to improve the learning process instead of considering this task as one of conventional classification.
In this talk we aim to describe each of these paradigms, as well as their relationships and current status. The challenges and future work in each of these disciplines will be highlighted.
Biography: Salvador García received the B.S. and Ph.D. degrees in Computer Science from the University of Granada, Granada, Spain, in 2004 and 2008, respectively. He is currently a Full Professor in the Department of Computer Science and Artificial Intelligence, University of Granada, Granada, Spain. Dr. García has published more than 100 papers in international journals (more than 70 in Q1), h-index 52. As edited activities, he is an associate editor in chief of “Information Fusion” (Elsevier), and an associate editor of “Swarm and Evolutionary Computation” (Elsevier) and “AI Communications” (IOS Press) journals. He is a co-author of the books entitled “Data Preprocessing in Data Mining”, “Learning from Imbalanced Data Sets” and “Big Data Preprocessing: Enabling Smart Data” published by Springer. His research interests include data science, data preprocessing, Big Data, evolutionary learning, Deep Learning, metaheuristics and biometrics. He belongs to the list of the Highly Cited Researchers in the area of Computer Sciences (2014-2020): http://highlycited.com/ (Clarivate Analytics).
Prof. Dr. Vincenzo Piuri,
Department of computer Science
Università degli Studi di Milano, Italy
Title: Ambient intelligence
Abstract Adaptability and advanced services for ambient intelligence require an intelligent technological support for understanding the current needs and the desires of users in the interactions with the environment for their daily use, as well as for understanding the current status of the environment also in complex situations. This infrastructure constitutes an essential base for smart living. Various technologies are nowadays converging to support the creation of efficient and effective infrastructures for ambient intelligence.
Artificial intelligence can provide flexible techniques for designing and implementing monitoring and control systems, which can be configured from behavioral examples or by mimicking approximate reasoning processes to achieve adaptable systems. Machine learning can be effective in extracting knowledge form data and learn the actual and desired behaviors and needs of individuals as well as the environment to support informed decisions in managing the environment itself and its adaptation to the people’s needs.
Biometrics can help in identifying individuals or groups: their profiles can be used for adjusting the behavior of the environment. Machine learning can be exploited for dynamically learning the preferences and needs of individuals and enrich/update the profile associated either to such individual or to the group. Biometrics can also be used to create advanced human-computer interaction frameworks.
Cloud computing environments will be instrumental in allowing for world-wide availability of knowledge about the preferences and needs of individuals as well as services for ambient intelligence to build applications easily.
This talk will analyze the opportunities offered by these technologies to support the realization of adaptable operations and intelligent services for smart living in an ambient intelligent infrastructures.
Biography: Vincenzo Piuri has received his Ph.D. in computer engineering at Politecnico di Milano, Italy (1989). He is Full Professor in computer engineering at the Università degli Studi di Milano, Italy (since 2000). He has been Associate Professor at Politecnico di Milano, Italy and Visiting Professor at the University of Texas at Austin and at George Mason University, USA.
His main research interests are: intelligent systems, artificial intelligence, signal and image processing, machine learning, pattern analysis and recognition, biometrics, intelligent measurement systems, industrial applications, cloud computing, and dependablity. Original results have been published in more than 400 papers in international journals, proceedings of international conferences, books, and book chapters.
He is Fellow of the IEEE, Distinguished Scientist of ACM, and Senior Member of INNS. He has been IEEE Vice President for Technical Activities (2015), IEEE Director, President of the IEEE Computational Intelligence Society, Vice President for Education of the IEEE Biometrics Council, Vice President for Publications of the IEEE Instrumentation and Measurement Society and the IEEE Systems Council, and Vice President for Membership of the IEEE Computational Intelligence Society.
He is Editor-in-Chief of the IEEE Systems Journal (2013-19), and Associate Editor of the IEEE Transactions on Computers and the IEEE Transactions on Cloud Computing, and has been Associate Editor of the IEEE Transactions on Neural Networks and the IEEE Transactions on Instrumentation and Measurement.
He received the IEEE Instrumentation and Measurement Society Technical Award (2002). He is Honorary Professor at Obuda University, Budapest, Hungary, Guangdong University of Petrochemical Technology, China, Muroran Institute of Technology, Japan, and the Amity University, India.
Prof. Dr. Dries F. Benoit
Head - Big Data Analytics Research
Ghent Univeristy, Belgium
Title: Educational data mining
Abstract Educational data mining is a rapidly evolving area that has attracted increasing interest of the machine and deep learning community in the last years. Multidisciplinary nature of the field leads to unique challenges in user behaviour and content modelling. One of the focal points of educational data mining research is building personalised learning systems which could mimic one-on-one tutoring on a large scale.
Our team is involved in developing a semantic-based approach to personalisation in language and mathematics education. As a part of this project, we implement deep learning architectures to predict student's success in the online course. Since there is a demand for educational applications to be highly interpretable, we rely on attention mechanisms and embedding visualisations to shed light on the learning strategies of the students, investigating both best and worst trajectories.
We also adapt natural language processing techniques to capture richer data about the content of mathematical exercises. The goal is to evaluate whether the syntax information contained in the formulas can lead to better prediction of student's performance. Such models have the potential to be used in building optimised learning trajectories that would allow the students to progress faster and achieve better results.
Biography:
Dries F. Benoit is head of the Big Data Analytics research group at
Ghent University, Belgium. He is associate professor of Data
Analytics at the Faculty of Economics and Business Administration. He
holds an MSc degree in Marketing Analytics and a PhD in Applied
Economics (Data Analytics). As of 2017, he is visiting professor at
Université de Namur (Belgium) in the Data Science program.
His main expertise is in Bayesian modeling with a focus on
high-dimensional data, robustness and decision making, where most
applications are in the field of business administration and
management (marketing, finance, operations research).
As a data-scientist, he often work together with researcher from other
fields such as medicine, energy, education, etc.
He has published in international peer-reviewed journals like Journal
of Statistical Software, Journal of Applied Econometrics,
Computational Statistics, Statistical Modelling, Journal of the
Operational Research Society, Decision Support Systems, and Expert
Systems with Applications, among others. He is currently guest editing
a special issue on Interpretable Data Science at Decision Support
Systems.
He regularly serves as reviewer member for international journals and
conferences and is co-organizer of the 2020 Ifors conference (Seoul,
Korea). Over the last years, he has been a stream organizer at the EURO
conferences. Moreover, he is the main developer and maintainer of the
R/Fortran software-package, bayesQR: Bayesian quantile regression
(available on CRAN). He was involved in company projects in different
industries such as financial services, education, retail, and
government among others.
December 16: 12:00 GMT
Prof. Dijiang Huang, PhD
Fulton Entrepreneurial Professor
School of Computing, Informatics and Decision Systems Engineering
Ira A. Fulton Schools of Engineering, Arizona State University, USA
Title : “Overcoming the Privacy Restraints for Future IoT services – A Software-Defined Edge Cloud Computing Approach”
Abstract : In this talk, we will focus on Software-Defined Edge Clouds. The Edge Computing landscape enables millions of IoT devices to shape an enormous intelligent network that can perform tasks that are usually only possible in large data centers. To support IoT modern services, software-defined networking, software-defined storage, software-defined systems, and the shift to merchant silicon/white box have poised to be the disruptive technology to revolutionize the IoT industry.
While software-defined solutions were not explicitly developed for IoT challenges, it can provide impetus to solve the complex issues and help in efficient IoT service orchestration. Using software-defined technologies and edge cloud computing to support IoT services has a few benefits: (a) edge cloud computing resolves the resource limitation problems by bringing computation closer to the edge of IoT devices; (b) software-defined networking provides flexible and adaptable, QoS-guaranteed networking services to distributed edge nodes across the distributed networking environment; (c) software-defined infrastructure provides both storage and in-memory computing support to enhance the real-time IoT service models, etc.
The current IoT paradigm of massive data generation, complex infrastructures, security and privacy vulnerabilities, and requirements from the newly developed technologies make IoT realization a challenging issue. In this talk, we focus on challenging privacy issues and present a software-defined solution to address the presented privacy issues. Furthermore, we will present open research issues, specifically based on edge-cloud-supported IoT services.
Biography: Dr. Huang received his Bachelor of Science degree in Telecommunications from Beijing University of Posts & Telecommunications, China. He received his Master of Science and PhD degrees from University of Missouri-Kansas City majored in Computer Science and Telecommunications. He is currently an associate professor in the School of Computing Informatics and Decision Systems Engineering. His research interests are in computer and network security, mobile ad hoc networks, network virtualization, and mobile cloud computing. Dr. Huang's research is supported by federal agencies NSF, ONR, ARO, and NATO, and organizations such as Consortium of Embedded System (CES), Hewlett-Packard, and China Mobile. He is a recipient of ONR Young Investigator Award, HP Innovation Research Program (IRP) Award, and JSPS Fellowship. He is a co-founder of Athena Network Solutions LLC (ATHENETS) and CyNET LLC. He is currently leading the Secure Networking and Computing (SNAC) research group at ASU. He is a senior member of IEEE and IEEE ComSoc Distinguished Lecturer. For more information, please refer to http://www.public.asu.edu/~dhuang8/
Prof. Dr.Ketan Kotecha
Head, Symbiosis Centre for Applied Artificial Intelligence (SCAAI)
Symbiosis Institute of Technology
Chief Executive Officer (CEO), Symbiosis Centre for Entrepreneurship and Innovation
Title : Domain Adaption in Deep Learning: Theories and Applications
Deep learning has to create revolutions in the field of Artificial Intelligence. However, there is an unrealistic assumption that training and test data comes from the same distribution. This assumption does not hold in practical applications. An example includes, an autonomous car designed for Europe won't work directly in Indian scenario being visual data will be quite different. Spam email classifier intended for the general purpose may not work well for a specific user. Not only that there are applications where domain adaption will save tremendous efforts if implemented with the proper insight. There are also attempts to generalise the algorithms for any domain leads to Domain Generalization(DG). Combining DA/DG to Few short learning and meta-learning also is a need of the day and a lot of work is in progress in those areas.
The talk shall throw light on defining Domain Adaption and allied areas like transfer learning, few-shot learning, meta-learning and multitask learning. Also, few states of the algorithms in these areas, along with applications, will be discussed. The talk will end, giving a few directions for research.
Biography: Recipient of Erasmus + faculty mobility grants from European Union LEAP (Leadership for Academicians Programme) grant from MHRD Govt of India in collaboration with IIT Kharagpur and University of Cambridge UK, November 2019-Jan 2020. DUO- INDIA Professor fellowship under Asia - Europe Meeting (ASEM-DUO) with Brunel University, UK, 2020
Two Research grants worth INR 1.66 crores under Promotion of Academic and Research Collaboration(SPARC) scheme by MHRD, Govt of India, in collaboration with Arizona State University USA and University of Queensland, Australia.
Invite and sponsorship from Embassy of the United States of America to participate in US-India Higher Education Collaboration Workshop to be held in Washington D.C on April 2020 for establishing international partnerships for research and academic collaborations with India and USA. An advocate and practitioner of emotional intelligence in workplaces, Dr Ketan Kotecha is at the helm of the administrative, academic and entrepreneurship affairs of the Symbiosis International (Deemed University). His 25 years of extraordinary career saw him serving in the finest of the engineering colleges in various higher technical education leadership positions.
An avid researcher, he has various prestigious transnational projects under him. He is recipient of the projects worth INR 1.6 crores on AI for Credibility Analysis of Information and Explainable AI for Health care in collaboration with the Arizona State University, USA and the University of Queensland, Australia under Scheme for Promotion of Academic and Research Collaboration(SPARC) by MHRD, GoI respectively. He is also a team member for the nationwide initiative on "AI and deep learning Skill and Research named Leadingindia.ai initiative sponsored by Royal Academy of Engineering, UK under Newton Bhabha Fund.
A researcher - teacher of Deep learning, his interest areas are Artificial Intelligence, Computer Algorithms, Machine Learning, Deep Learning Higher Order Thinking Skills, Critical Thinking and Ethics & Values. His research work can also be seen at https://scholar.google.co.in/citations?user=oNiE0gMAAAAJ&hl=en
He has more than 100 papers published /presented at international conferences around the world, to his credit and 3 patents filed. He is Guide to doctoral students working on the various field of Computer Engineering and especially in Machine Learning for PhD/MTech Degrees. His insightful articles on varying topics are also published in news dailies; CSI magazine; AIU University newsletter etc.
He was keynote speaker for events (recent ones) such as ACMA Technology Summit 2019, Pune; international conference on “Flexible Learning Pathways: Asia-Europe Conference on Lifelong Learning and the 2030 Agenda for Sustainable Development”, Hanoi, Vietnam; Artificial Intelligence & Data Analytics in Decision Making" at the 14th FICCI Higher Education Summit 2018. He was also a member of FICCI higher education delegation to S Korea in May 2019. He has also delivered a talk on Research directions for Artificial intelligence at various forums including at IIT Bombay.
Dr Kotecha was founding Provost of Gujarat's largest private university- Parul University. He was Director – Academic Development and Research Cell, Nirma University, Ahmedabad, where he also served as the Dean and Director of the Institute of Technology, Nirma University.
He was Governing Council Member of the USA based Global Engineering Dean’s Council – India Chapter and is also a member of National Advisory Council for Confederation of Indian Industry’s (CII) Engineering and Management Curriculum Restructuring Task Force. He is the member of Governing Council of various universities and member of Academic Council at few Universities. He was a Member of the Technical Advisory Committee of BRTS, Ahmedabad and is also a member of Metro–Link Express for Gandhinagar.
Dr Kotecha is also an independent director of Gujarat Informatics nominated by Government of Gujarat. A recipient of Erasmus + faculty mobility grant from European Union, Dr Kotecha was invited by Wroclaw University of Science and Technology and Poznan University of technology, Poland for delivering sessions on Machine learning. He was a visiting expert to University of Pretoria, S. Africa. He was also invited and visited various countries like USA, Canada, Singapore, Hong Kong, S. Africa, Spain, Poland, Germany, Cz Republic, Switzerland, Argentina, China, Vietnam, and South Korea. Dr Kotecha is pioneer in Education Technology, believes in drastic curriculum reforms and innovative teaching-learning practices. He is also a voracious reader and is passionate about travelling and indulging in good food.
Prof. Dr. Sheela Ramanna
Applied Computer Science
University of Winnipeg, Canada
Title: Tolerance-based Granular Computing
Abstract: This talk presents the theory and major applications of tolerance-based granular computing framework for different forms of knowledge representation and machine learning. This research bridges the gap between approximate and crisp forms of knowledge structures. Tolerance relations provide the most general tool for studying indiscernibility phenomena. Specifically, we employ the following granular methods: Tolerance form of rough sets, fuzzy rough sets and tolerance near sets. A direct result of this research is its utility in three distinct areas: natural language processing, social networks and perception-based image and audio information. Construction of knowledge repositories from web corpora by harvesting linguistic patterns is of benefit for many natural language-processing applications that rely on question-answering schemes. These methods require minimal or no human intervention and can recursively learn new relational facts/instances in a fully automated and scalable manner. This talk highlights a novel tolerance rough-set based model and fuzzy rough model for learning relational facts from web corpora and explores the issue of concept drift in a semi-supervised learning setting. A popular real-world application of community detection can be found in social networks, where networked communities are fundamental structures for understanding social behaviour. This talk presents a novel approach for discovering non-overlapping community structures based on tolerance spaces and near set theory. In addition, a soft clustering method based on a hybrid computational geometry approach with Voronoi diagrams and tolerance-based neighborhoods to detect overlapping communities is also presented. Lastly, this talk presents results from tolerance near sets-based supervised learning methods in music genre classification and solar-flare detection.
Biography: Dr. Sheela Ramanna is a Full Professor and past Chair of the Applied Computer Science Department. She is the co-founder of the ACS graduate studies program University of Winnipeg.
She received a Ph.D. in Computer Science from Kansas State University, U.S.A and a BS in
Electrical Engineering and MS in Computer Science and Engineering from Osmania University,
India. She serves on the Editorial Board of Springer Transactions on Rough Sets (TRS) Journal,
Elsevier Engineering Applications of AI Journal and Advisory Board of the International Journal
of Rough Sets and Data Analysis. She is the Managing Editor of the TRS and is a Senior
Member of the IRSS (Intl. Rough Set Society). She has co-edited a book on Emerging Paradigms in Machine Learning published in 2013 by Springer. She has served as Program Co-Chair for MIWAI 2013, RSKT 2011, RSCTC 2010 and JRS2007. She is currently the Program Co-Chair of IJCRS 2021 track of IFSA/EUSFLAT 2021. She is the recipient of a TUBITAK Fellowship (Turkey) for 2015. Her research is funded by Natural Sciences and Engineering Research
Council of Canada Discovery and Engage Grants Program. She has received more than
$1,130,000 in research funding since 1992. She has published over 50 peer-reviewed articles in the past 6 years in journals such as Pattern Recognition Letters, Knowledge-Based Systems,
Neural Computing and Applications, Granular Computing, Knowledge and Information Systems,
Intelligent Information Systems, Mathematics in Computer Science, Frontiers Advances in
Computational Neurosciences, and Neuroscience Letters. The focus of her research is in
fundamental and applied research in machine learning and granular computing. Her current
interests include: i) Tolerance–based granular computing techniques (fuzzy sets, rough sets and
near sets) with applications in social networks, natural language processing, computer vision and audio signal processing, ii) topological data analysis, and iii) application of descriptive
proximities.