Research
 @ SIT

ในวันพฤหัสบดีที่ 31 มีนาคม 2559 เวลา 10.00 น.- 15.00 น. ณ ห้องTraining 4/1 ชั้น 4
อาคาร คณะเทคโนโลยีสารสนเทศ มจธ. บางมด

Abstract Detail: Search Strategies: Classical to Modern Approaches (Artificial Intelligence through Search)

Almost all Artificial Intelligence (AI) programs perform some form of “problem solving” whether it be interpreting a visual scene, scheduling time tables, playing chess or planning a sequence of robot actions. For many complex problems one needs to find out a good solution from among a set of possible solutions; and hence some “search” strategy is required to obtain this solution in a reasonable time. The complexity increases when the system does not have any background (domain) knowledge about the problem. Further, even if the problem is solved, the solution, thus found, may not necessarily be the global optimum (best) one --- may be a local/near-optimal too.

Search procedures may be blind (uninformed) or heuristic based (informed); it can be based on state space graph or others. Search can also be viewed as an optimization process. Various performance measuring indices exist to evaluate the quality of search strategies. In this series of lectures, initially some classical blind (like BFS, DFS) and informed search (like A*) techniques will be discussed keeping in mind their evaluation aspects. This will be followed by some modern search and optimization methods which are iterative too (e.g., Hill Climbing, Simulated Annealing). In the next part, focus will be given on Evolutionary Algorithms (EAs), an iterative multi-solution based meta-heuristic search and optimization strategy, evolved based on natural systems. Finally, utility of EAs for solving some real life problems will be discussed.

Ashish Ghosh is a Professor at the Machine Intelligence Unit, Indian Statistical Institute. He has already published more than 170 research papers in internationally reputed journals and refereed conferences, and has edited eight books. His current research interests include pattern recognition and machine learning, data mining, big data analysis, image analysis, remotely sensed image analysis, video image analysis, soft computing, fuzzy sets and uncertainty analysis, neural networks, evolutionary computation, and bioinformatics. Dr. Ghosh received the prestigious and most coveted Young Scientists Award in Engineering Sciences from the Indian National Science Academy in 1995, and in Computer Science from the Indian Science Congress Association in 1992. He was selected as an Associate of the Indian Academy of Sciences, Bangalore, India, in 1997. He is a member of the founding team that established the National Center for Soft Computing Research at the Indian Statistical Institute, Kolkata, in 2004, with funding from the Department of Science and Technology, Government of India, and is currently the In-charge of the Center. He is acting as a member of the editorial boards of various international journals.

*Biography*
_Susmita Ghosh_,

Susmita Ghosh received the B.Sc. (with Honors) in Physics and the B.Tech. in Computer Science and Engineering from the University of Calcutta, India, in 1988 and 1991, respectively, the M.Tech. in Computer Science and Engineering from the Indian Institute of Technology, Bombay, in 1993, and the Ph.D. (Engineering) from Jadavpur University, Kolkata, in 2004. She has visited various universities/academic institutes for collaborative research/ international projects and delivered lectures in different countries including Australia, China, Italy, Japan and the USA. She has published many research papers in internationally reputed journals and referred conferences and has edited 3 books. She was the Principal Investigator of several research projects (including India Trento Programme for Advanced Research) funded by government agencies of National/ International repute (including India Trento Programme for Advanced Research). Presently, she is a Local Coordinator of two Erasmus Mundus (funded by European Union) Projects. Susmita Ghosh was a Lecturer and then an Associate Professor with the Department of Computer Science and Engineering, Jadavpur University since 1997. She is currently a Full Professor at the same department. Her research interests include genetic algorithms, neural networks, image processing, pattern recognition, soft computing, bioinformatics and remotely sensed image analysis.

If you are interested, Please contact Miss Paweena Mongkolpongsiri,
Tel.9815, or register at http://goo.gl/forms/mOnupnxWh3, or
email: paweena@sit.kmutt.ac.th
Thank you.

Details of Advance Workshop

09:00 a.m. - 10.30 a.m. Lecture and Workshop

10.30 a.m. -10.45 a.m. -- Coffee Break --

10.45 a.m. - 11.50 a.m. Lecture and Workshop

11.50 a.m. - 12.00 a.m. -- Lunch Break –

13:00 p.m. - 14.30 p.m. Lecture and Workshop

14.30 p.m. -14.45 p.m. -- Coffee Break --

14.45 p.m. - 17.00 p.m. Lecture and Workshop

Details:
In this workshop, we will have a hands-on data analysis by utilizing data mining and optimization techniques. The objective is to have an idea of how to implement these techniques and what is the benefit of them on current researches. The actual research problem will be used to enhance the learning experience.

Short Bio:
Assoc. Prof. Dr. FONG Simon James,
Department of Computer and Information Science
Faculty of Science and Technology, University of Macau.

URL: http://www.cis.umac.mo/~ccfong/introducton/#

Simon Fong graduated from La Trobe University, Australia, with a 1st Class Honours BEng. Computer Systems degree and a PhD. Computer Science degree in 1993 and 1998 respectively. Simon is now working as an Associate Professor at the Computer and Information Science Department of the University of Macau. He is also one of the founding members of the Data Analytics and Collaborative Computing Research Group in the Faculty of Science and Technology. Before joining the University of Macau, he worked as an Assistant Professor in the School of Computer Engineering, Nanyang Technological University, Singapore. Prior to his academic career, Simon took up various managerial and technical posts, such as systems engineer, IT consultant and e-commerce director in Melbourne, Hong Kong and Singapore. Some companies that he worked before include Hong Kong Telecom, Singapore Network Services, AES Pro-Data and United Oversea Bank, Singapore. Dr. Fong has published over 289 international conference and peer-reviewed journal papers, mostly in the areas of E-Commerce technology, Business Intelligence and Data-mining.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

There are details of these research workshop;

-Thursday 9th July 2015, 9.00 am. - 12.00 am.
    Workshop 1 Title:
    Analyzing Sensor Data Stream : Challenges and Prospects

- Friday 10th July 2015, 9.00 am. - 12.00 am.
     Workshop 2 Title:
    Bio-inspired Optimization Algorithms for
    Improving Data Mining and Other Applications

Schedule of these Workshops

09:00 a.m. - 09.10 a.m. Registration

09:10 a.m. - 10.30 a.m. Lecture and Workshop

10.30 a.m.. -10.45 a.m. -- Coffee Break –

10.45 a.m. - 11.50 a.m. Lecture and Workshop

11.50 a.m. - 12.00 a.m. Q & A

If you are interested to join this free Workshop,
please reserve a seat at:

http://goo.gl/forms/Tuu36sYz8k

Moreover, if you have any questions about the workshop,
please feel free to email us at
email: paweena@sit.kmutt.ac.th
School of Information Technology, KMUTT.

Abstract:
Workshop 1 Title: Analyzing Sensor Data Stream : Challenges and Prospects* By Assoc. Prof. Dr. FONG Simon James

Abstract:
Analyzing real-time sensor data is expected to be a feasible but challenging task even equipped with the latest frontier of information technology. The volume of sensor data is predicted by experts that will vastly surpass from various sources of Big Data, such as wireless sensor network, IOT, CCTV videos, social media over the next decade. Enterprises are eager to discover the true benefits by data mining such sensor data in terms of real-time intelligence. Technically, from the perspective of analytics, a new breed of data mining techniques known as data stream mining seems to be a possible solution though it has a relatively short history of development. With the number of new sensor data sources and issues of Big Data that are emerging, data stream mining massive data sets in real-time has a strong application motivation. In particular, data stream mining which is unlike its predecessor, traditional data mining, is able to process and incrementally induce a prediction model on the fly without the need of loading in the full archive of data. This attractive property makes it a potentially suitable candidate for the next generation real-time sensor data analtyics solution. In this workshop, the pros and cons as well as some recommended solutions are discussed on the possibilities of applying data stream mining methods for analyzing sensor data in real-time. Computer simulation demo will be conducted for showing the efficacy of some current data stream mining methods over sensor data.

Workshop 2 Title: Bio-inspired Optimization Algorithms for Improving Data Mining and Other Applications* By Assoc. Prof. Dr. FONG Simon James

Abstract:
Modern Bio-inspired computing and metaheuristics algorithms (BiCam) such as PSO, Fireflies, Ants, Bats and Bees algorithms etc. start to demonstrate their power in dealing with tough optimization problems and even NP-hard problems. The research momentum is picking up recently as optimal solutions for combinatorial optimization are possible to be sought by Metaheuristics. The research introduced in this workshop is about a new BiCam algorithm which can outperform the current ones, as well as modifying from some important existing BiCam algorithms for applying in data mining applications; so as to produce optimal solutions in the most efficient way. The benefits offered by BiCam algorithms complement very well the limitations of data mining because data mining models usually face multidimensional combinatorial problems which are typically NP-hard, with very large search space including about finding global optima and overcoming local optima. Data mining models are well known to be prone to suffer from the curse of dimensionality, which also makes them infeasible for exhaustive search or complex analytical methods. Three case studies will be described in this talk: possible integration of BiCam algorithms into some classical data mining algorithms - namely (1) Data Clustering; (2) Feature Selection in Classification, and (3) Highly Non-linear Regression. At the end of the talk, sample source codes will be provided for free, for fellow researchers to apply the integrated bio-inspired optimization and data mining algorithms to application problems of your own choice.

Short bio:
Assoc. Prof. Dr. FONG Simon James, Department of Computer and Information Science Faculty of Science and Technology, University of Macau. URL:http://www.cis.umac.mo/~ccfong/introducton/#
Simon Fong graduated from La Trobe University, Australia, with a 1st Class Honours BEng. Computer Systems degree and a PhD. Computer Science degree in 1993 and 1998 respectively. Simon is now working as an Associate Professor at the Computer and Information Science Department of the University of Macau. He is also one of the founding members of the Data Analytics and Collaborative Computing Research Group in the Faculty of Science and Technology. Before joining the University of Macau, he worked as an Assistant Professor in the School of Computer Engineering, Nanyang Technological University, Singapore. Prior to his academic career, Simon took up various managerial and technical posts, such as systems engineer, IT consultant and e-commerce director in Melbourne, Hong Kong and Singapore. Some companies that he worked before include Hong Kong Telecom, Singapore Network Services, AES Pro-Data and United Oversea Bank, Singapore. Dr. Fong has published over 289 international conference and peer-reviewed journal papers, mostly in the areas of E-Commerce technology, Business Intelligence and Data-mining.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

*There are details of Research Workshop on Social Media Analysis and Visualisation.

Monday 8th June 2015: Module “Social Media”

Tuesday 9th June 2015: Module “Information Visualisation”

Wednesday 10th June 2015: Module “Visualisation of Structures”

Thursday 11th June 2015: Module “Visualisation of Dynamics”

Friday 12th June 2015: Module “Social Visualisation”

Abstract:
Research Workshop “Social Media Analysis and Visualisation” Social media refers to software that mediates human communication and strengthens online social ties. Prominent examples include Facebook, Twitter, Wikipedia, Flickr, YouTube, FourSquare, Yelp, Delicious, Digg, Second Life, and many others. Technologically these are largely based on Web 2.0 technology, often supplemented with custom mobile apps. Social media have become a powerful economic force and influential in shaping online opinions. This workshop will discuss technical and application dimensions of social media, and explore how the use of social media can be better understood through analysis and visualisation of social media data. Participants will work on case-studies and propose visualisation designs appropriate to the case in question.

Topics covered:

1. Social Media
    - What is social media: definitions and dimensions
    - Motivations: drivers to using social media
    - Size and growth of social media
    - Types of social media: classifications and taxonomies
    - Applications: application areas and criteria

2. Information Visualisation
    - What is information visualisation: definitions and advantages
    - Perception and human visual information processing
    - Visualisation fields
    - Types of data
    - Process of creating visualisations
    - Visualisation primitives
    - Visualisation techniques and interaction

3. Visualisation of Structures
    - Structure: definition
    - Types of structure: communication, artefacts, workspaces
    - Case study I: Communication structures – Twitter,
      Email, Message forums
    - Case study II: Artefact structures – Wikipedia, Blogs, Flickr
    - Case study III: Workspace structures – LiveNet

4. Visualisation of Dynamics
    - Dynamics: definition
    - Aspects of dynamics
    - Traditional representations
    - Case study I: Communication dynamics – Email,
      Message forums
    - Case study II: Artefact dynamics – Wikis, Blogs

5. Social Visualisation
    - Challenges and issues
    - Modes of sharing
    - Examples of social media in visualisation
    - Examples of visualisation in social media

Short bio:
Assistant Professor Robert P. Biuk-Aghai
http://www.cis.umac.mo/~robertb/
http://www.fst.umac.mo/en/staff/robertb.html

He is an Assistant Professor in the Department of Computer and Information Science at the Faculty of Science and Technology of the University of Macau, where he has been working since 1993. Over the years, he has taught a number of subjects, both undergraduate and postgraduate, and supervised several M.Sc. theses and B.Sc. final projects. His main research interest lies in the areas of information visualization and CSCW, and he is one of the founders of the Data Analytics and Collaborative Computing Group (formerly Business Intelligence Group) at his department. He is also the initiator and project lead of the MacauMap (TM) project — a highly popular handheld and web-based tourist map guide system for Macau with about a million downloads since 2003. Moreover, from 2008 to 2012 He was the Coordinator for European Exchange Programs at his faculty.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Prof. Dr. Akira Namatame,
Department of Computer Science,
National Defense Academy of Japan

There are details of this seminar:
1:00 pm. - 1.10 pm. Registration
1.10 pm. – 2.40 pm. Talk 1: Optimizing Complex Networks
2.40 pm. - 2.55 pm. -- Coffee Break –
2.55 pm. - 4.25 pm. Talk 2: Agent-based modeling of supply chain and global logistic networks
On Tuesday 28th April 2015, 1.00 - 4.25 P.M.
at Training 4/1 4th floor, School of Information Technology,
KMUTT, Bangmod, Bangkok.

Abstract:

Talk 1: Optimizing Complex Networks By Prof. Dr. Akira Namatame Abstract: In this talk, I will provide a brief introduction to network science and optimization issues of complex networks. Network science has attracted considerable attention over the last decade. Two principal approaches have contributed to understanding complex networks so far. The first is an assembly mechanism that derives the structure of large-scale networks from processes that describe the piecewise addition of nodes and links according to simple heuristic rules such as preferential attachment. The second approach is via an optimization principle that aims at optimizing network performance through its evolution. The increasing scale and complexity of networked systems has necessitated an approach for network resilience and robustness. Since the network topology has a direct impact on performance such as resilience and robustness, a mechanism of self-control is also an important and immediate requirement. The network topology is the most fundamental network property on which self-control is deployed, and the control over the topology has a direct impact on network performances. The network topology impacts on failure spread in networks. The issue of enhancing network resilience could be analyzed with optimizing the underlying network topology. Traditionally, optimization has a strict mathematical definition, which refers to obtaining the solutions that strictly specialize a well-defined function. Here we adopt a looser definition by extending it to include a tendency of the network to improve its performance as a result of evolution. We define metrics on network robustness and resilience and formulate multi-criteria network optimization problems. We use generic algorithm (GA) to solve these optimization problems and obtain optimal network topologies. Many real-world problems are complex with a very large parameter space. Accordingly, most attempts for finding the best solution are not realistic. Instead, finding a good enough solution or a better solution in an iterative manner becomes alternative.

Talk 2: Agent-based modeling of supply chain and global logistic networks By Prof. Dr. Akira Namatame Abstract: This talk consists of two parts. In the first part, I introduce an agent-based modeling tool that accounts for the decision-making nature of the various elements comprising the supply chain. Agent-based modeling for managing the supply chain views the supply chain as consisting of a set of intelligent agents, each responsible for one or more activities in the supply chain and each interacting with other agents in the planning and executing of their responsibilities. Many firms increasingly depend on a complicated network of global suppliers and partners to deliver products in the right quantity and at the right place and right time in increasingly volatile markets and under persistent cost pressures. Agent-based supply chain modeling also accounts for the time-varying nature of the behavior of certain subsystems according to the changing objectives of the decision makers, based on their knowledge of other decision makers and the environment the supply chain is embedded. Agent based modeling framework has many desirable features (autonomy, intelligence and coordination) for understanding supply chain behavioral dynamics under changing situations. Logistics is the collective term for the range of activities involved in the movement of goods from points of production to final points of sale and consumption. It must insure that the mobility requirements of supply chains are entirely met. Physical distribution includes all the functions of movement and handling of goods, particularly transportation services. The second part of my talk is concerned with the global logistic networks, the world-wide ship networks and airport networks. We can characterize the current global flows of goods and people by analyzing some features of these global networks.

Short bio:
Prof. Dr. Akira Namatame
Department of Computer Science
National Defense Academy of Japan,
Yokosuka, Japan.
URL: www.nda.ac.jp/~nama
Dr. Akira Namatame is Professor of Computer Science Department at National Defense Academy of Japan. His research interests include multi-agent systems, complex networks, supply chain / logistic networks, evolutionary computation, and game theory. In the past ten years, he has given over 35 invited talks, and over 15 tutorial lectures in international conferences and workshops, and academic institutions. He has organized more that 30 international conferences and workshops, and special sessions. He is the editor-in-chief of Springer’s Journal of Economic Interaction and Coordination (JEIC), editor in Modeling and Simulation Society Letter. He has published more than 250 refereed scientific papers, together with eight books on multi-agent systems, agent and complex networks, collective systems and game theory. More detail information can be obtained through.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

There are details of this seminar:

Title: Run-based Exception Prediction for Workflows
By Asst. Prof. Dr. SI Yain Whar, Lawrence
Abstract:
Events such as iteration of activities or lack of available resources can cause temporal exception in business processes. Exception prediction can improve the quality of workflow execution since preventative actions can be taken to reduce the occurrence of exceptions. Thus, it is crucial to provide an accurate and efficient temporal exception prediction capability for workflow management systems. In this seminar, I will describe about a run-based exception prediction algorithm to predict temporal exception in workflows. The proposed algorithm is divided into two phrases, design-time and run time. In design-time, all possible runs are generated from workflow and their estimated execution time and mapping probability are calculated. In run time, temporal exceptions are predicted by analyzing the runs. Simulation experiments are performed to evaluate the proposed approach using five workflow models having different characteristics. Simulation experiments show that our approach is efficient and produces good results in prediction accuracy.


Title: Effect of Segmentation on Financial Time Series Pattern Matching
By Asst. Prof. Dr. SI Yain Whar, Lawrence
Abstract:
In financial time series pattern matching, segmentation is often performed as a pre-processing step to reduce the data points from the input sequence. The segmentation process extracts important data points and produces a time series with reduced data points. Although segmentation methods were evaluated against one or two pattern matching approaches in recent studies, no comparative analysis has been ever performed on all of the well-known methods of time series segmentation methods and pattern matching. In this paper, we evaluate the effectiveness and accuracy of four approaches to financial time series pattern matching when used with four segmentation methods, the perceptually important points, piecewise aggregate approximation, piecewise linear approximation and turning points methods. The pattern matching approaches analysed in this paper include the template-based, rule-based, hybrid, and decision tree approaches. The analysis is performed twice, on a real data set (of Hang Seng Index prices from the Hong Kong stock market) and on a synthetic data set containing positive and negative cases of a technical pattern known as head-and-shoulders.


Title: Generalized analytic signal for envelope detection in 2D QLCT domain
By Assoc. Prof. Dr. KOU Kit Ian, Anna
Abstract:
The quaternion Fourier transform plays a vital role in the representation of multidimensional signals. The quaternion linear canonical transform (QLCT) is a well-known generalization of the quaternion Fourier transform. In this talk, we study some valuable properties of the two-sided QLCT. The 2D analytic signal and Hilbert transform are widely used for various practical applications in image processing. Different approaches to the 2D quaternion Hilbert transforms are introduced that allow the calculation of the associated analytic signals, which can suppress the negative frequency components in the QLCT domains. As an application, plot simulations for edge detection demonstrate the effectiveness of our approach.

Short bio:
Assistant Professor Dr. Lawrence Si Department of Computer and Information Science Faculty of Science and Technology, University of Macau. URL: http://www.fst.umac.mo/en/staff/fstasp.html

Dr. Lawrence Si is an Assistant Professor in the Department of Computer and Information Science at the Faculty of Science and Technology, University of Macau. He obtained his M.Sc. degree in Software Engineering from the University of Macau in 1997. In 2005 he received his Ph.D. degree in Information Technology at the Queensland University of Technology, Australia. His research interests are in the areas of business process management, data mining, and decision support systems. He is a founding member of Data Analytics and Collaborative Computing Group from the Department of Computer and Information Science, University of Macau. He is also an Associate Master of Henry Fok Pearl Jubilee College at the University of Macau.

Associate Professor Dr. Anna, Kit Ian Kou Department of Mathematics Faculty of Science and Technology, University of Macau. URL: http://sheac.rc.umac.mo/?page_id=751

Prof. Anna, Kit Ian Kou is an Associate Professor in the Department of Mathematics at the Faculty of Science and Technology, University of Macau. She obtained her M.Sc. and Ph.D. degree in Mathematics from the University of Macau. Her research interests are in the areas of hypercomplex signal analysis and color image analysis, Quaternion Fourier analysis, Complex and Clifford analysis.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Professor Mark Chignell,
Department of Mechanical and Industrial Engineering
University of Toronto, Canada

There are details of this seminar:
The details of the workshop
In this three-hour event we will look at what serious games are and how to choose applications that are suitable for a serious games approach. We will focus on developing serious games for the elderly, although many of the principles that we discuss should apply to serious games in general. We will do some hands-on activities designing serious games and we will also review some serious games case studies. At the end of this workshop, participants should be able to identify opportunities for building serious games, and they should be able to do the initial “look and feel” design of the user interface for a serious game. The approximate schedule for the workshop is as follows: Hour 1: Introduction to Serious Games, Cognitive Assessment, and Neuro-Rehabilitation. Hour 2: Rapid prototyping design and evaluation of serious game concepts (with hands-on exercises) Hour 3: A serious game case study: Usability engineering and validation of a serious game for cognitive assessment

Short bio:
:http://www.mie.utoronto.ca/faculty/profile.php?id=18
Mark Chignell has been a member of the Mechanical & Industrial Engineering faculty since 1990. Prior to that he was an Assistant Professor in Industrial and Systems Engineering at the University of Southern California from 1984 to 1990. Professor Chignell taught in the Psychology Department at Monash University in Melbourne, Australia from 1980 to 1982. He has a PhD in Psychology (University of Canterbury, New Zealand, 1981), and an MS in Industrial and Systems Engineering (Ohio State, 1984). He is currently the Director of the Knowledge Media Design Institute at the University of Toronto. He has been a visiting scientist at the IBM Centre for Advanced Studies in Toronto since 2002 and he was a visiting scientist at Keio University from 2005 to 2010. Professor Chignell has co-authored books on expert systems and intelligent databases (published by John Wiley & Sons) and he has co-edited three books. He has many journal and conference papers on topics such as hypertext and information retrieval, user interface design and healthcare applications. His research has been funded by the Bell University Laboratories, the National Sciences and Engineering Research Council (NSERC), IBM, the Ontario Centres of Excellence (OCE), the Japan Society for the Promotion of Science, Oki Corporation, Ricoh Corporation, and Apple Computer. He has founded two start-up companies, and currently runs Vocalage Inc., a high tech consulting and product development firm.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Abstract Detail:
The financial industry as we know it today wouldn’t exist without high powered information systems. These systems make markets possible through electronic intermediation and supporting services such as data and analytical support. They are also becoming used increasingly to support or make decisions about taking or controlling some sort of risk – financial, operational, and regulatory. Quantitative trading is an investing technique which makes use of computer algorithms in order to find trading opportunities. Quantitative trading is often associated with the most sophisticated and technically advanced investment houses. This talk is the first in the Financial Engineering talk series. The talk will provide the basic concepts about securities and how they are traded. Beyond this point, the talk will focus on the technology point of view, including the design and building of algorithmic trading models. Common trading platforms will be demonstrated. Possible research topics in this area will be introduced, as well as some guidelines on how anyone can go into starting a career as a quant.

Speaker info: Dr. Carsten Mueller,
Baden-Wuerttemberg Cooperative State University, Germany

Short bio:
Dr. Simon Fong graduated from La Trobe University, Australia, with a 1st Class Honours BEng. Computer Systems degree and a PhD. Computer Science degree in 1993 and 1998 respectively. Simon is now working as an Associate Professor at the Computer and Information Science Department of the University of Macau. He is also one of the founding members of the Data Analytics and Collaborative Computing Research Group in the Faculty of Science and Technology. Prior to joining the University of Macau, he worked as an Assistant Professor in the School of Computer Engineering, Nanyang Technological University, Singapore. Prior to his academic career, Simon took up various managerial and technical posts, such as systems engineer, IT consultant and e-commerce director in Melbourne, Hong Kong and Singapore. Some companies that he worked before include Hong Kong Telecom, Singapore Network Services, AES Pro-Data and United Oversea Bank, Singapore. Dr. Fong has published over 220 international conference and journal papers, mostly in the area of E-Commerce technology, Business Intelligence and Data-mining.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Associate Professor Dr. Simon Fong,
Department of Computer and Information Science, University of Macau
Av. Padre Tomas Pereira, Taipa, Macau SAR
http://www.cis.umac.mo/~ccfong

Abstract:
Modern Bio-inspired computing and metaheuristics algorithms (BiCam) such as PSO, Fireflies, Ants, Bats and Bees algorithms etc. start to demonstrate their power in dealing with tough optimization problems and even NP-hard problems. The research momentum is picking up recently as optimal solutions for combinatorial optimization are possible to be sought by Metaheuristics. The research introduced in this Keynote speech is about a new BiCam algorithm which can outperform the current ones, as well as modifying from some important existing BiCam algorithms for applying in data mining applications; so as to produce optimal solutions in the most efficient way. The benefits offered by BiCam algorithms complement very well the limitations of data mining because data mining models usually face multidimensional combinatorial problems which are typically NP-hard, with very large search space including about finding global optima and overcoming local optima. Data mining models are well known to be prone to suffer from the curse of dimensionality, which also makes them infeasible for exhaustive search or complex analytical methods. Three case studies will be described in this talk: possible integration of BiCam algorithms into some classical data mining algorithms - namely (1) Data Clustering; (2) Feature Selection in Classification, and (3) Highly Non-linear Regression. At the end of the talk, sample source codes will be provided for free, for fellow researchers to apply the integrated bio-inspired optimization and data mining algorithms to application problems of your own choice.

Short bio:
Dr. Simon Fong graduated from La Trobe University, Australia, with a 1st Class Honours BEng. Computer Systems degree and a PhD. Computer Science degree in 1993 and 1998 respectively. Simon is now working as an Associate Professor at the Computer and Information Science Department of the University of Macau. He is also one of the founding members of the Data Analytics and Collaborative Computing Research Group in the Faculty of Science and Technology. Prior to joining the University of Macau, he worked as an Assistant Professor in the School of Computer Engineering, Nanyang Technological University, Singapore. Prior to his academic career, Simon took up various managerial and technical posts, such as systems engineer, IT consultant and e-commerce director in Melbourne, Hong Kong and Singapore. Some companies that he worked before include Hong Kong Telecom, Singapore Network Services, AES Pro-Data and United Oversea Bank, Singapore. Dr. Fong has published over 220 international conference and journal papers, mostly in the area of E-Commerce technology, Business Intelligence and Data-mining.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Professor Dr.Richard Deaves,
The DeGroote School of Business, McMaster University,Canada .

Abstract:
Economics as a field is the study of how individuals (and companies) make optimal decisions when facing constraints. For example, consumers maximize utility subject to a budget constraint. In finance, investors maximize expected utility subject to what risk-return combinations are available. Normative behavior (i.e., that is, how people should make decisions) is often well known. Behavioral economics (and finance) are all about how people actually do behave. In fact, individuals behave suboptimally at times, and in large part this is due to persistent behavioral (i.e., psychologically-based) biases of various kinds.ย In other words, people are less than fully rational. Tiger Woods (and other professional golfers) display a suboptimal behavior that is likely driven by what behavioral economists call loss aversion, which is the human tendency to frame decisions narrowly and overly fixate on short-term losses. In my talk, I explore some patterns in the suboptimal behavior of individuals, particularly in the financial realm, and discuss how these patterns impact financial markets.

Short bio:
The focus of his latest research is the field of behavioral finance. He has also published on such issues as the performance of investment funds; asset market experiments; market efficiency; modeling interest rates; and pricing, efficiency and hedging in futures markets. He has written three books, the latest in 2010 beingBehavioral Finance: Psychology, Decision-making and Markets, published by South-Western, Cengage Learning. Education •PhD, Economics, University of Toronto •Specializations: Financial Economics, Monetary Economics Professional Activities His research has been published extensively, appearing in such prestigious journals as the Journal of Financial and Quantitative Analysis, the Journal of Monetary Economics and the Journal of Banking and Finance. He has also addressed groups of students and executives at other universities in Canada, the U.S., South America, Europe and the Far East. His far-reaching consulting experience has spanned such topics as investor behavior; pension design; hedging in the energy industry; the predictability of interest rates; real options and capital budgeting; and security valuation. He has served as an expert in a variety of regulatory and legal proceedings.
http://www.degroote.mcmaster.ca/profiles/richard-deaves/

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Professor Mark Chignell
Department of Mechanical and Industrial Engineering University of Toronto, Canada.

Abstract:
We are living in a world of “big data” where huge amounts of data are being collected online in areas as diverse as earth science, healthcare, and e-commerce. The key to big data applications is not just collecting large amounts of data, but also creating appropriate linkages between data sets and incorporating that data into usable and useful applications, using meaningful interactions and visualizations. Big data applications will revolutionize healthcare in the future and in this talk we present current research on this topic that is being carried out in my lab. The first project that we focus on is seeking to provide a new kind of clinical decision support system for emergency and intensive care physicians. I will start by reviewing the tasks carried out by these physicians and the requirements that can be inferred from those tasks. I will then review relevant prior research on clinical decision support using a case-based reasoning approach and I will present a wireframe prototype that shows the kind of application functionality and user interface interaction that we are envisioning. After that I address the big data backend that has to be created to support the application functionality. I will report on data mining that we have carried out on the MIMIC II database of 25,000 intensive care patients and the system that we have developed for making predictions and recommendations using a set of abstracted data types (based on cluster analysis). Using prediction of death in the ICU as the criterion I will present results showing that prediction of death using summarized (non-confidential)data is actually more accurate than predictions made using the original raw (confidential data). I will argue that big data applications based on predictions made on summarized data can avoid the privacy problems that are currently slowing the widespread use of healthcare data. The second application considered in this talk is one where the data is not mined from a repository, but is constructed using data collected in real time for cognitive assessment. In this case we envisage a situation where people such as residents in long-term care or patients in a neuro-rehabilitation ward play a game frequently over a period of weeks or months with detailed data on their performance being collected. The resulting data can the be monitored for changes in cognitive status within individuals, e.g., due to the onset of a condition like delirium or to a transition from normal cognition to mild cognitive impairment (MCI), or from MCI to Alzheimer’s Disease. We will describe the Whac-A-Mole game that we have developed jointly with researchers at King Mongkut University, Thonburi and we will then report on an experiment that we carried out to assess the validity of the game as a way of assessing cognitive status. The talk will conclude with a brief review of some of the challenges and opportunities in developing Big Data applications for healthcare.

Short bio:
Professor Mark Chignell:http://www.mie.utoronto.ca/faculty/profile.php?id=18 Mark Chignell has been a member of the Mechanical & Industrial Engineering faculty since 1990. Prior to that he was an Assistant Professor in Industrial and Systems Engineering at the University of Southern California from 1984 to 1990. Professor Chignell taught in the Psychology Department at Monash University in Melbourne, Australia from 1980 to 1982. He has a PhD in Psychology (University of Canterbury, New Zealand, 1981), and an MS in Industrial and Systems Engineering (Ohio State, 1984). He is currently the Director of the Knowledge Media Design Institute at the University of Toronto. He has been a visiting scientist at the IBM Centre for Advanced Studies in Toronto since 2002 and he was a visiting scientist at Keio University from 2005 to 2010. Professor Chignell has co-authored books on expert systems and intelligent databases (published by John Wiley & Sons) and he has co-edited three books. He has many journal and conference papers on topics such as hypertext and information retrieval, user interface design and healthcare applications. His research has been funded by the Bell University Laboratories, the National Sciences and Engineering Research Council (NSERC), IBM, the Ontario Centres of Excellence (OCE), the Japan Society for the Promotion of Science, Oki Corporation, Ricoh Corporation, and Apple Computer. He has founded two start-up companies, and currently runs Vocalage Inc., a high tech consulting and product development firm.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Professor Mark Chignell
Department of Mechanical and Industrial Engineering University of Toronto, Canada.

Abstract:
No one would argue against physics or chemistry being sciences, but engineers often have trouble admitting that psychology is a science. And since engineering is applied science, if something isn’t a science, why should engineers study it? In this talk I will argue that knowledge of the relevant aspects of human psychology is becoming essential for engineers. First I will discuss why engineers need to learn psychology. Then I will review the topics in psychology that engineers need to know about, drawing from the required course (MIE 242, Psychology for Engineers) that I have been teaching to second year industrial engineering students at the University of Toronto since the fall of 2010. The first thing I will do is remove from consideration those aspects of psychology that are either not relevant to engineers, or that are too far removed from “hard science” to form a credible part of the engineering curriculum given the current state of knowledge. Psychoanalysis, abnormal psychology, personality theory, and social psychology will all be discarded from consideration in this discussion. I will then focus on the topics that I cover in MIE 242 and I will justify each of them, explaining why each is relevant to engineers. I will also review the current course text by Smith and Kosslyn, discussing some of its strengths and weaknesses as a text for engineers. After reviewing the lecture and textbook material I will then turn to the seven laboratories and the research paper that is the core of the practicum material covered in the course. I will discuss the research methods and introductory statistics that are needed for the course and I will describe the research paper deliverable where students carry out some independent research. I will then outline a new textbook that I am planning to write for this course and I will give some examples of exam questions that have been used in the course. I will conclude by talking about the expected benefits of the course and how it prepares engineers to do a better job of understanding people and designing for them. I will argue that engineers need to know enough about people to design appropriately, but that the knowledge needed is not just guidelines and heuristics. Instead I will argue that engineers need to have core scientific knowledge about the human brain and cognition that will guide their decision making and that will allow them to carry out appropriate research relating to human perception, motor control, emotion, cognition, language and decision making, when design problems and situations require it. I will conclude by describing how MIE 242 serves as a bridge to later courses in our programme, on engineering psychology and on human factors.

Short bio:
Professor Mark Chignell:http://www.mie.utoronto.ca/faculty/profile.php?id=18 Mark Chignell has been a member of the Mechanical & Industrial Engineering faculty since 1990. Prior to that he was an Assistant Professor in Industrial and Systems Engineering at the University of Southern California from 1984 to 1990. Professor Chignell taught in the Psychology Department at Monash University in Melbourne, Australia from 1980 to 1982. He has a PhD in Psychology (University of Canterbury, New Zealand, 1981), and an MS in Industrial and Systems Engineering (Ohio State, 1984). He is currently the Director of the Knowledge Media Design Institute at the University of Toronto. He has been a visiting scientist at the IBM Centre for Advanced Studies in Toronto since 2002 and he was a visiting scientist at Keio University from 2005 to 2010. Professor Chignell has co-authored books on expert systems and intelligent databases (published by John Wiley & Sons) and he has co-edited three books. He has many journal and conference papers on topics such as hypertext and information retrieval, user interface design and healthcare applications. His research has been funded by the Bell University Laboratories, the National Sciences and Engineering Research Council (NSERC), IBM, the Ontario Centres of Excellence (OCE), the Japan Society for the Promotion of Science, Oki Corporation, Ricoh Corporation, and Apple Computer. He has founded two start-up companies, and currently runs Vocalage Inc., a high tech consulting and product development firm.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Professor Mark Kortschot,
Ph.D., P.Eng., FCAE Chair, Division of Engineering Science University of Toronto, Canada.

Abstract:
There has been a significant change in the perception of entrepreneurship over the past decade, driven by the successes in Silicon Valley, but spreading into the world of product and process design and popularized through crowdsourcing sites such as KickStarter. This has had an impact on engineering education in North America. While the 1980's, many students aspired to careers in large corporations, today, an increasing number of students see an entrepreneurial job as a viable and even preferred alternative. Universities have responded with a renewed emphasis on engineering design education, together with co-curricular support of entrepreneurship for students. In this seminar, Professor Kortschot will discuss the North American culture of entrepreneurship, and will focus on the response of the Engineering Science program at the University of Toronto to the shifting landscape.

Short bio:
Mark T. Kortschot is the Chair of the Division of Engineering Science at the University of Toronto and is a Professor in the Department of Chemical Engineering and Applied Chemistry. He obtained his B.A.Sc. (1984) and M.A.Sc.(1985) degrees in the Faculty of Applied Science and Engineering at the University of Toronto, and a Ph.D. from Cambridge University in 1989. Professor Kortschot is interested in both the design of materials and design with materials. His academic research has focussed on the relationship between microstructure and properties in fibrous materials and he has published ~ 60 refereed journal publications. He is also an active designer, inventor and entrepreneur with special expertise in materials selection, manufacturing, and intellectual property. He is a co-founder of L?g?re Reeds Ltd., which makes premium synthetic saxophone and clarinet reeds used in the world’s leading orchestras, and is the inventor of the Sole Skate?, a compact skateboard chosen as one of TIME’s Top 10 toys of 2011.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Assistant Professor Ryosuke Saga,
Department of Computer Science and Intelligence Systems,
Osaka Prefecture University , Japan.

Abstract:
Information Visualization is important and useful for understanding data. We propose a method to visualize trend for time series text data by using network graph called FACT-Graph. FACT-Graph is basically based on two methods; Class Transition Analysis and Co-occurrence Transition Analysis (CTAs). Additionally, we have carried out comparison visualization by using FACT-Graph based on the feature of CTAs. In this presentation, I introduce FACT-Graph and show related ongoing project about FACT-Graph.

Short bio:
Ryosuke Saga is an Assistant Professor at department of computer science and intelligent systems of Osaka Prefecture University in Japan. He obtained his B.S. in Industrial Engineering in 2003, and M.S. and PhD in Computer Science in 2005 and 2008 from Osaka Prefecture University. He is a senior member of IEEJ (The Institute of Electrical Engineers of Japan) and also works as a member of investigation commissioners of service innovation and that of big data usage in IEEJ. His research interests include data engineering, knowledge management, and visualization.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info:Prof. Dr. Mahmut Parlar,
DeGroote School of Business,
McMaster University, Canada.

Abstract:
U.S. expenditures on search-based advertising exceeded $12 billion in 2010. Advertisers bid for keywords, where bid price determines ad placement, affecting click-through and conversion rates. Advertisers must select keywords, allocating each a proportion of their fixed daily budget. In this paper, we construct a stochastic model for the selection and allocation process. We provide analytical results for the single-keyword problem and examine the multiple-keyword problem numerically. We investigate trade-offs between keywords given varying levels of risk and return. We show the implications of enforcing a probabilistic budget constraint. Our paper provides a critical analysis of the advertiser's problem that may guide future research.

Short bio:
Mahmut Parlar is a Professor of Management Science at McMaster University's DeGroote School of Business. He holds a B.Sc. in mathematics (1973) and an M.Sc. in operations research/statistics (1975) both from the Middle East Technical University, Ankara, Turkey, and a Ph.D. in management sciences (1979) from the University of Waterloo, Canada. Mahmut Parlar's research program is focused in the areas of stochastic modelling, dynamic optimization and game theory applications in supply chain management. His work in these and related areas has been supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) research grants since 1980 and have appeared in the form of more than 100 articles published in journals such as Operations Research, Management Science, Production and Operations Management, Naval Research Logistics, IIE Transactions, Journal of Operations Management, Annals of Operations Research, Transportation Science, Applied Mathematical Finance and others. Professor Parlar is the author of a book entitled "Interactive Operations Research with Maple" published in 2000 by Birkhi user, Boston.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Fredric M. Ham, Ph.D.,
IEEE Fellow, SPIE Fellow, INNS Fellow,
Harris Professor of Electrical Engineering,
Florida Institute of Technology, USA

Details can be found on the website at
http://inns.sit.kmutt.ac.th/plsr.html

Abstract:
Partial Least Squares Regression (PLSR) is a factor analysis-based method that is used to construct predictive models when the number of factors is very large and highly collinear (factors are significantly redundant). Factors are measured variables used to explain, regulate, or predict the behavior of other variables or the responses. Multiple Linear Regression (MLR) methods work well when the number of factors is small, they are not significantly collinear (redundant) and there exists a welldefined relationship to the system responses. When any or all of the above conditions don’t exist MLR will fail. That is when PLSR becomes a very valuable modeling tool. When one is faced with many variables, the relationship between the inputs (independent variables) and the responses (dependent variables) are not well-known, the PLSR statistical calibration modeling approach can be employed. This calibration model when properly constructed can produce highly robust predictive results. PLSR is an excellent method used for prediction when the physical system characteristics, or the details of the process, is not well understood. Through the factor analysis approach, unneeded factors can be removed that can improve the predictive performance of the calibration model. The removal of the unwanted factors can be carried out without detailed knowledge of their characteristics. Thus, unwanted, obscuring components that can exist in the data can be removed in a systematic manner, thus improving the predictive performance of the calibration model. Some applications of PLSR include: statistical calibration model development for parameter estimation, estimation of sinusoids in additive noise, parametric system identification and robust event classification.

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Dr.Hironori Nakajo,
Department of Computer and Information Sciences, Faculty of Engineering,
Graduate School of Tokyo University of Agriculture and Techonology, Japan

If you interest to join this seminar
please reserve a seat at k.Paweena Mongkolpongsiri,
phone no.9815 or
email: paweena@sit.kmutt.ac.th
Thank you.

Speaker info: Assoc. Prof. Dr. Yew-Soon Ong
Director of Centre for Computational Intelligence (C2I) School of Computer Engineering, Nanyang Technological University, Singapore

Abstract of Talk: A novel learning paradigm is presented to automatically identify groups of informative and correlated features from very high dimensions, i.e., millions of dimensions. Specifically, we explicitly incorporate correlation measures as constraints and then propose an efficient embedded feature selection method using recently developed cutting plane strategy. The benefits of the proposed algorithm are two-folds. First, it can identify the optimal discriminative and uncorrelated feature subset to the output labels, denoted here as Support Features, which brings about significant improvements in prediction performance over other state of the art feature selection methods considered in the paper. Second, during the learning process, the underlying group structures of correlated features associated with each support feature, denoted as Affiliated Features, can also be discovered without any additional cost. These affiliated features serve to improve the interpretations on the learning tasks. Extensive empirical studies on both synthetic and very high dimensional real-world datasets verify the validity and efficiency of the proposed method.

Brief Biography : Yew-Soon Ong is currently an Associate Professor and Director of the Center for Computational Intelligence at the School of Computer Engineering, Nanyang Technological University, Singapore. He received a PhD degree on Artificial Intelligence in complex design from the Computational Engineering and Design Center, University of Southampton, UK in 2003. His current research interest in computational intelligence spans across memetic computing, evolutionary design, machine learning, agent-based systems and cloud computing. He is the founding Technical Editor-in-Chief of Memetic Computing Journal, Chief Editor of the Springer book series on studies in adaptation, learning, and optimization, Associate Editor of IEEE Computational Intelligence Magazine, the IEEE Transactions on Systems, Man and Cybernetics - Part B, Information Sciences, Soft Computing, International Journal of System Sciences and others. He has coauthored over 120 refereed publications and his research grants in the last five years amounts to a total of more than 5 million Singapore dollars. These comprises of research funding from both national and international partners that includes the Singapore National Grid Office, A*STAR, Singapore Technologies Dynamics, Boeing Research & Development (USA), Rolls-Royce Singapore and Honda Research Institute Europe (Germany), National Research Foundation and Media Development Authority- Singapore-MIT GAMBIT Lab. Presently, he also Chaired the IEEE Computational Intelligence Society Emergent Technology Technical Committee from 2011-2012 and has served as Guest Editors of several journals including the IEEE Transactions on Evolutionary Computation. Presently, he Chairs the IEEE Computational Intelligence Society Intelligent Systems Technical Committee. His research work on Memetic Algorithm was featured by Thomson Scientific's Essential Science Indicators as one of the most cited emerging area of research in August 2007. Recently, he also received the 2012 IEEE Transactions on Evolutionary Computation Outstanding Paper Award for his work pertaining to the modeling of Probabilistic Memetic Framework. In teaching, he has also received numerous awards including the Nanyang Excellence Award for Teaching in 2008, Most Popular Lecturer Award 2009, and recently invited as Fellow of Renaissance Engineering Programme at Nanyang Technological University.

If you interest to join this seminar
Please contact to Miss Paweena by 1 March at
Email: paweena@sit.kmutt.ac.th

Speaker info: Prof. Dr. Sung-Bae Cho
Softcomputing Laboratory Computer Science Department
Yonsei University, South Korea

If you interest to join this seminar
Please contact to Miss Paweena by 12 February at
Email: paweena@sit.kmutt.ac.th

Tuesday, January 15, 2013 (hosted by D-Lab)

TITLE: A Neurocomputing Approach for Monitoring Plinian Volcanic Eruptions and Other Natural Phenomena Using Infrasound

Fredric M. Ham, Ph.D. Dean, College of Engineering IEEE Fellow, SPIE Fellow, INNS Fellow Harris Professor of Electrical Engineering Florida Institute of Technology 150 West University Boulevard Melbourne, Florida 32901-6975 (USA)