Home UCSC

4012

International Conference on Advances in ICT for Emerging Regions (ICTer) is the successor to the seminal International Information Technology Conference (IITC) held in Sri Lanka since 1998. It provides a platform where research done in ICT is presented by both local and foreign Computer Scientists and IT Professionals. This year we have participants from Germany, Australia, Sweden, Japan, New Zealand, Taiwan, United Kingdom, Singapore and India.

In order to get wider international participation and to promote computing research in the fast emerging regions of the world especially in Asia-Pacific, it was decided to partner with Institute of Electrical and Electronics Engineers (IEEE). The proceeding of the conference are distributed both in print and electronic form. The articles are published throw IEEE explorer and linked to Google scholar and selected papers are published in a special edition of the ICTer Journal with a global audience.

This year the conference will take place at BMICH, Bauddhaloka Mawatha, Colombo 07, Sri Lanka on 12th and 13th December 2013. In addition to the presentation of selected papers, several Keynote Addresses and Invited Speeches by leading personalities in the IT world will be made. The conference will also include pre-conference (11th) and post-conference (14th) high quality tutorials/workshops in areas of current interest in Information and Communication Technology.

Out of 112 Submissions of papers to ICTer2013, the organizing committee has now selected 18 as full paper submissions, 24 short papers and 18 poster papers. There will be 8 keynote speeches from senior professors from United States, Japan, Taiwan, Sweden, Australia and India. There will be approximately 8 workshops and tutorials will be held as pre and post conference activities. Registration for ICTer2013 is now open – http://www.icter.org/conference/register

 

 

 

Workshops and Tutorials

Following is the list of workshops and Tutorials on the 11th and 14th of December 2013. Please use the “W#” system when you register for workshops and tutorials. 

 

W1.   Medical Informatics
(Full Day Workshop on the 11-Dec-2013 from 09.00 to 16.30)

Resource person: Dr. Chamidu Atupelage, Dr. Ruwan Ranaweera, Dr. Mohan Jayathilake, Dr. Roshan Yapa, Dr. Amalka Pinidiyaarachchi and Dr. Anuja Dharmarathne.

During last few decades, medical experts massively collaborate with the engineers to develop advanced techniques and tools for improving the significances of medical diagnosing techniques. Especially, with the development of high-resolution imaging devices and high-spec computers modeling the medical specialists’ knowledge into mathematical methods has become real-time practical problems.  This workshop aims to introduce and hands on cutting-edge medical informatics technologies.

Specially suitable for Medical doctors, bioinformaticians, post-graduate medical informaticians, etc.

 

W2.   e-Waste and Green Computing 

(Full Day Symposium on the 11-Dec-2013 from 09.00 to 16.30)

Resource Persons: Mr. Harsha Wijewardana, Mr. A. M. S. C. M. B. Attanayake, Mr. Kuganathan and Mr. F. Hudah

The Government of Sri Lanka announced not long ago that general populace of Sri Lanka is in possession of more than twenty million mobile phones; in other words the number of phones had surpassed the population of this tiny nation. Although this was hailed as a remarkable achievement keeping pace with other nations in Asia in mobile communication, it also poses threat to our fragile environment as these phones reach their End of Life (EOL) if it is not addressed properly. These tiny digital devices have short life span and some will reach their EOL in less than three years. These devices including computers are made of toxic materials such as lead, copper and mercury. As they reach EOL they become eWaste and haphazard dumping of these will harm our eco system polluting our water and soil. When these toxic materials are in our water and soil, they will also enter into our food chain and many of these toxic materials are carcinogens and cause many other diseases.

Green Computing is defined as the practice of environmentally sustainable Computing or IT. Green computing covers whole life cycle of computers and digital devices which include use of less energy to disposing of servers, computers, network accessories etc.

As mentioned previously, using less energy or less electricity to power computers has become a major part of Green Computing and industrialized world has contributed much of the emission of carbon into the environment as a result of generation of energy using fossil fuel and many other nations such as China, Brazil and India are joining the ranks of high users of fossil fuel as they industrialized at a faster pace. It is said that rise of CO2 has contributed to the raising of global environmental temperature in general; and it is expected with the melting of polar caps, the water levels of our oceans to rise submerging some of our low lying areas in our country; and some tiny island nations may disappear from the world map altogether. Some scientists attribute change in our climate to the rise of temperature worldwide and it will have a huge impact on our food sources threatening worldwide food shortage in a few years.

This workshop will introduce what eWaste and Green Computing is; and it is further expected to impart the best practices which have been adopted worldwide and in Sri Lanka to practice Green Computing with elaboration of worldwide standards. It will also try to introduce Carbon Trading which is practiced many of Sri Lankan Exporters. The organizers of workshop expect that this workshop will become a platform for initiating a dialog in eWaste and Green Computing in Sri Lanka.

W3.   Cloud Computing in Software Development 
(Full Day Workshop on the 11-Dec-2013 from 09.00 to 16.30)
Conducted by the 99x Technologies, Resource Persons: Mr.Samudra Kanankearachchi, Mr.Chatura De Silva, Mr.Geethanga Amarasinghe, Mr.Chathuranga Bandara and Mr.Kalanamith Mannapperuma

Keywords: Cloud Computing, Process distribution,  Multi-tenancy, Cloud Security 

Cloud computing is an emerging paradigm of the service based software industry which subjected to diverse research attempts for last few years. This long-held dream of software as a service is applied in various domains showcasing its strength and versatile applicability. This workshop will cover cloud computing considering, aforementioned great applicability and ability that it has imparted to software development employing various cutting edge technologies. As cloud computing is based on well-defined set of concepts, workshop will also provide better explanation of these concepts for those who like to seek and explore this trendsetting domain and state-of-the-art technologies. Nevertheless, workshop is designed in such a way to provide in-depth knowledge from the fundamental concepts to the hands on experience in cloud based technologies. Though the main focus of this workshop is on cloud computing, this will extend its coverage to related concepts and techniques as well which are associated with cloud computing. Therefore, security, scalability and layering in cloud based environment will also be considered and will be deeply discussed in this workshop. Moving few steps further, this workshop will provide practical exposure of architectural building blocks of cloud environment concerning cloud balancing, cloud bursting architectures and clustering techniques. With this coverage of advance topics in cloud computing, experience sharing sessions will also be held to contribute to the software community. Providing coverage for all above mentioned areas, workshop will be consisted of 8 sessions including a special session allocated to discuss future dimensions and directions of the cloud computing where industry experts will provide the exposure in a more interactive environment.

 

W4.   Low Cost Immersive VR Simulations for Industrial Applications and Serious Games
(Full Day Workshop on the 11-Dec-2013 from 09.00 to 16.30)

Resource Persons:  Prof. N. D. Kodikara, Prof. Rexy Rosa, Mr. Y Prabath Samarasinghe and Mr. Damitha Sandaruwan

The performance of computer hardware has improved tremendously over the past decade and the cost has also gone down.It is now possible to use commodity-off-the-shelf (COTS) hardware to model real world physics in real time with the accuracy required for VR based simulations.In addition, there are free and open source software (FOSS) packages that could be used to build physics simulators and rendering engines to create the virtualenvironments. Accordingly, low cost serious games equal to commercial high end simulators can be developed using off-the-shelf hardware and open source softwarecomponents.This discussion focuses on the design challenges in developing immersive virtual reality solutions for serious games using COTS hardware and FOSS.The term serious game is very broad; it is not  possible to address the design issues related to the broad spectrum of games covered by this term. In this discussion, we limit our focus on serious games usedfor training in handling vehicles, such as ships, flights, and cars in immersive virtual environments.In this workshop the  VR solutions developed by modeling and simulation group of UCSC which were built using the COTShardware and FOSS will be discussed.

 

W5.    Building Future Proof ERP
(Half Day Workshop on the 11-Dec-2013 from 09.00 to 12.30)
Conducted by IFS, Resource Persons: Mr. Amil Kumarasinghe, Mr. Buddika Hiripitiyage and Mr. Sukitha Magallege

Key words:  Big Data, In-memory Databases, Real Time Network Surveillance, Cloud Computing, Mobile Computing, Responsive ERP

Modern day ERPs are feature creeps. They have a wide variety of features to facilitate today’s endless requirements. They are used for all of the internal information systems of an organization and also connected to the external social networks, geospatial data services, mobile applications, third party systems, government tax information services, and the list follows. Every single code line added to the ERP must be added with all of those in mind plus the traditional checklist including performance, usability, clarity, user friendliness, security, scalability and so on.  On the other hand, ERP applications contain lot of data that users access through its user interface. Interactions between users are captured in formal flows such as the progress of a defined work activity. However formal flows capture only a small portion of the actual activity that occurs on an instance of an application. Users access data in patterns in real time and collaborate with each other without acting on an activity in many daily business actions. These kinds of access patterns can provide rich insight into emergent business requirements and problems if they are processed, categorized and presented in real time.  Handling of big data to derive decisions is possible by bringing together real time network traffic surveillance, in-memory databases and fast set based time segmentation of data in order to provide business insight to decision makers in real time. The workshop will discuss technical aspects of how to build future proof ERP while accomplishing the emerging market needs and simultaneously dealing with big data. Industry Experts will offer two unique interactive sessions that will cover the aforementioned aspects.

 

W6.  Video Forgery and Motion Editing
(Full Day Workshop on the 14-Dec-2013 from 09.00 to 16.30)
Resource Person: Prof. Timothy K. Shih National Central University, Taiwan

Video Forgery is a technique for generating fake video by altering, combining, or creating new video contents. We change the behavior of actors in a video. For instance, the outcome of a 100-meter race in the Olympic Game can be falsified. We track objects and segment motions using a modified mean shift mechanism. The resulting video layers can be played in different speeds and at different reference points with respect to the original video. In order to obtain a smooth movement of target objects, a motion interpolation mechanism is proposed based on reference stick figures (i.e., a structure of human skeleton) and video inpainting mechanism. The video inpainting mechanism is performed in a quasi-3D space via guided 3D patch matching. Interpolated target objects and background layers are fused. It is hard to tell whether a falsified video is the original. In addition, in this talk, we demonstrate a new technique to allow users to change the dynamic texture used in a video background for special effect production. For instance, the dynamic texture of fire, smoke, water, cloud, and others can be edited through a series of automatic algorithms. Motion estimations of global and local textures are used. Video blending techniques are used in conjunction with a color balancing technique. The editing procedure will search for suitable patches in irregular shape blocks, to reproduce a realistic dynamic background, such as large waterfall, fire scene, or smoky background. The technique is suitable for making science fiction movies. We demonstrate the original and the falsified videos in our website at http://www.csie.ncu.edu.tw/~tshih. Although video falsifying may create a moral problem, our intension is to create special effects in movie industry.

 

W7.   Mining Unstructured Text

(Full Day Tutorial on the 14-Dec-2013 from 09.00 to 16.30)

Resource Persons: Dr. A. Ruvan Weerasinghe and Mr. VirajWelgama

Many real world applications require us to try and figure out the meaning of words and phrases in unstructured texts. While complex machine readable dictionaries may help, they require man years of work to compile and so are not available for many languages. This tutorial will cover text mining techniques that can be used to make sense of words and phrases occurring in natural text using an open source toolkit (NLTK). It will give the participant a grasp of how to set about processing large unstructured texts in order to find out information contained in them.  While many examples used in the tutorial will be in English, attention will also be given to issues surrounding the processing of Sinhala and Tamil data.
Organized by the Language Technology Research Laboratory (LTRL), University of Colombo School of Computing (UCSC)

 

W8. Computational Biology

(Full Day Totorial on the 14-Dec-2013 from 09.00 to 16.30)

Resource Persons: Prof. Mahesan Niranjan and Mrs. Rupika Wijesinghe

This tutorial will review computational algorithms that are used in extracting useful information from complex biological datasets. We will start with an introduction to biology from an information perspective and outline the computational challenges. We will look at algorithms for sequence analysis such as dynamic programming and hidden Markov models; inference from high throughput experimental data via cluster analysis, classification and matrix factorization; and present an overview of parameter inference for systems biology models.

We will also use this opportunity to discuss how research activity in the topics of bioinformatics and computational biology may be triggered and sustained in the Sri Lankan environment with impact on biological, medical and agricultural sectors.

 

W9. One head, many hats: Cross disciplinary technological interventions

(Half Day Workshop on the 14-Dec-2013 from 09.00 to 12.30)

Resource Persons: Dr. Santosh Vijaykumar, Dr. Owen Noel Newton Fernando and Mr. Vajira Sampath Rathnayake

The past decade has witnessed a rapid profusion of information and communication technology (ICT) based interventions in developing countries across domains such as public health, agriculture and education. The very nature of such interventions commands collaborations between people from a range of sectors (academia, industry, policymaking, etc.) sectors and disciplinary backgrounds (clinical sciences, social sciences, technology studies, etc.). While well intentioned, such collaborations can pose constant challenges in terms of optimizing the richness, and balancing the contributions of, each discipline. The purpose of this workshop is to offer ICTer practitioners and researchers an in-depth view of collaborative processes that involve a range of disciplines and brainstorm ideas on overcoming the key challenges in such situations. In doing so, a conceptual differentiation of three approaches – interdisciplinary, multidisciplinary and transdisciplinary – will be discussed. This will be followed by real-life case studies in executing such approaches based on projects at the Center of Social Media Innovations for Communities (COSMIC) at Singapore’s Nanyang Technological University. The workshop will culminate with a team project comprising delegates from different disciplinary backgrounds who will construct a solution to a chronic developing world problem.

 

W10. Document Image Analysis and Recognition

(Half Day Totorial on the 14-Dec-2013 from 09.00 to 12.30)

Resource Persons: Professor Umapada Pal

Document Image Analysis (DIA) is the process that performs the overall interpretation of document images.  Along with the existing applications like reading aid for the blind, postal automation, bank check automation, nowadays document analysis techniques has been used in many new areas. For example, currently companies are interested in implementing digital mailrooms to improve the efficiency of paper-intensive workflows and to reduce the burden of information processing of incoming mails, faxes, forms, invoices, reports, employee records, etc. Multi-script document recognition, Historical document transcription, Word spotting, Video document analysis and recognition, etc. are other challenging areas. This tutorial talk will give overview in all important aspects of document image analysis and recognition areas to the participants. Also for better understanding the subjects, demonstrations of some DIA systems will be given to the audience. Finally, some open and new challenging problems will be discussed.

 

 

Keynote Speakers.

Video Forgery and Motion Editing

Professor Timothy K. ShihNational Central University, Taiwan

Video Forgery is a technique for generating fake video by altering, combining, or creating new video contents. We change the behavior of actors in a video. For instance, the outcome of a 100-meter race in the Olympic Game can be falsified. We track objects and segment motions using a modified mean shift mechanism. The resulting video layers can be played in different speeds and at different reference points with respect to the original video. In order to obtain a smooth movement of target objects, a motion interpolation mechanism is proposed based on reference stick figures (i.e., a structure of human skeleton) and video inpainting mechanism. The video inpainting mechanism is performed in a quasi-3D space via guided 3D patch matching. Interpolated target objects and background layers are fused. It is hard to tell whether a falsified video is the original. In addition, in this talk, we demonstrate a new technique to allow users to change the dynamic texture used in a video background for special effect production. For instance, the dynamic texture of fire, smoke, water, cloud, and others can be edited through a series of automatic algorithms. Motion estimations of global and local textures are used. Video blending techniques are used in conjunction with a color balancing technique. The editing procedure will search for suitable patches in irregular shape blocks, to reproduce a realistic dynamic background, such as large waterfall, fire scene, or smoky background. The technique is suitable for making science fiction movies. We demonstrate the original and the falsified videos in our website at http://www.csie.ncu.edu.tw/~tshih. Although video falsifying may create a moral problem, our intension is to create special effects in movie industry.


Multimodality + Multimedia + Sensors = Pleasant Interfaces

Professor Masahito HirakawaInterdisciplinary Graduate School of Science and Engineering, Shimane University, Nishikawatsu 1060, Matsue 690-8504, Japan

Knowing the computer is a machine to help people achieve their tasks with less effort, its functions should be fully usable. Otherwise, the machine, or the computer, doesn’t make sense even if a lot of powerful functions are provided. This is why user interface development is a key technology in these days. While the notion of invisible interface has been recognized as a goal of its development, we say a pleasant interface is much preferable, which is beyond the invisible interface so that interfaces of future computers should make people fun and comfortable as well as usable without any special effort in advance of their use, that is, invisibleness. Meanwhile, some may claim that mind uploading is an ultimate user interface that is referred to as the process of transferring the totality or considerable majority of the mental contents from a particular human brain into a computer [1]. Our approach of pleasant interfaces is quite different from mind uploading in the sense that we are interested in humans alive.
We discuss trials which have been carried out in our laboratory as a step toward development of pleasant interfaces, which include multi-dip aqua interface, spatial auditory system, and footstep-based body motion analysis system.

[1] B. Goertzel and M. Ikle, “Special Issue on Mind Uploading,” International Journal of Machine Consciousness, Vol.4, No.1, pp.1-3, 2012; DOI: 10.1142/S1793843012020015.


Design-based Mobile Learning Research: Results and reflections on communication and sustainability

Prof. Robert RambergStockholm University, Sweden.

In the last decades design-based research has grown in application within the learning sciences. Key to design-based research within Technology Enhanced Learning (TEL), is that researchers, users and practitioner’s work together to produce a meaningful change in contexts of practice. Much research within mobile learning has however come under criticism for being technology-driven with a large portion of studies lacking explicit educational foundations (eg. Kukulska-Hulme et al., 2011, Traxler & Kukulska-Hulme, 2005). I.e., lacking at least one of the key characteristics of design-based research in general and design-based research within TEL in particular. Elaborated views of mobile learning have been articulated and these have constituted a significant step in the evolution of mobile learning, characterized by a shift of focus, from an imprecise and inadequate foregrounding of technology, towards a conceptualization of mobile learning that emphasizes social practices mediated by mobile technology. Crucial to design-based research in general and perhaps particularly to design-based research within TEL and mobile learning is that results from research are usable also to others than the research community. I.e.  to practitioners and other stakeholders involved in designing for a particular learning context or activity. Further, design solutions that have been collaboratively developed with stakeholders need to be usable also by others outside of the immediate collaboration. How else would these results become usable to a wider audience of practitioners?

Building on a design-based research approach, in the past five years a number of studies pertaining to mobile learning and design of out-door and in-door learning activities have been carried out in collaboration with a primary school in a suburb to Stockholm, Sweden. Results from these studies have among other things shown the importance of scaffolding and support across out-door and in-door learning activities. In my talk, I will present and discuss our design-based research process and results from studies conducted on out-door learning activities (for instance Nouri et al., 2011,and Nouri, 2011). I will also discuss and present different ways of trying to conceptualize and communicate results from the design-based research process in terms of guidelines for design and evaluation (for instance Eliasson et al., 2011, Eliasson et al., 2012, Eliasson et al., submitted).


Enhancing Health Communication using Digital Media: Trends and Experiences

Assoc Prof May Oo LwinNanyang Technological University, Singapore.

For decades, health authorities have utilized traditional media to reach various publics. Public health communication has depended primarily on dissemination of messages which originate from health organization to communities and individual public. These communication strategies utilized traditional media such as television, radios and mailers which lack individual reach. Recent developments in digital technology offer unprecedented communication opportunities via mobile and social media. Through strategies such as tailoring and crowdsourcing, health messages can now be targeted to individuals or groups, and further shared among their larger social networks. This talk will identify discuss key trends in the utilization of digital technology for heath communication to address public health issues. Case studies related to innovative digital solutions are presented. Each of these studies are conceptually bound (comprise alert, care and education) but address different health issues through a distinct integration of social media functions.


Architecture for Digital Knowledge Ecosystems

Prof. Athula Ginige, Professor of Information Technology – School of Computing, Engineering & Mathematics, University of Western Sydney – Australia

Ability to access required knowledge in a timely manner has been a major contributory factor for enhancing the productivity and empowerment of an individual, organisation, country and the society as a whole. To achieve this first the knowledge needs to be captured, organized and stored in repositories at a suitable granularity with appropriate meta-data and associations. The knowledge that needs to be captured could already exit in some other form or need to capture in real time as events are happening.
Development of the printing press greatly enhanced the unidirectional flow of knowledge; ie from author to thousands of readers. Electronic broadcast mediums; radio and Television and Web 1:0 further enhanced this unidirectional flow of knowledge. Web 2.0 technologies enabled every user to create and share content; text, images, photos, audio and video with an individual, group or with everybody. This bidirectional flow of information empowered the passive information consumers to become information producers. This resulted in new applications such as Facebook, twitter, YouTube, flickr etc. The number of users using these applications is growing very fast. This empowerment of users is now enabling the creation of Digital Knowledge ecosystems. In these systems users that consume the information also contributes to generating new knowledge through user actions. Such systems require suitable information architectures, effective information capture and access methods, new algorithms and approaches to discover new knowledge from user inputs and ways to identify user context to provide relevant information.
We are currently working on few projects to provide information to users in the right context especially using mobile devices. In these projects we had to address the above mentioned research challenges. Based on these project scenarios and solutions that we found for the research challengeswe are now developing a generic architecture for digital knowledge ecosystems which I will present in this talk.


Document Image Analysis: Past, Present and Future

Professor Umapada PalComputer Vision and Pattern Recognition Unit, Indian Statistical Institute, 203 B T Road, Kolkata-108, India

Document Image Analysis (DIA) is the process that performs the overall interpretation of document images.  It is one of the most fascinating and challenging areas of Pattern Recognition with various practical applications like: Reading aid for the blind, Postal mail sorting, Bank  check processing, Forensic data analysis, Seal or logo based document indexing etc.  Origin of this area can be found in 1870 when Carey invented a ratina scanner and at present many sophisticated techniques of this area can be found in the literatures.  In this talk, we will briefly discuss the state of the art techniques staring from the beginning to present days including current trends and recent advances of DIA. Also some open and new challenging problems of DIA area will be discussed towards its future research for the benefit of the audience. Finally, for better understanding of the different DIA techniques, demonstrations of some DIA systems will be shown to the audience.


Statistical Relational Learning – Different formalisms and representations

Dr. Sushil KulkarniJai Hind College, University Of Mumbai, Mumbai, India.

Traditionally machine learning deals with homogeneous data taken from an entity and assumed to be independently and identically distributed (iid). Each entity corresponds to a single relation contains number of attributes. However in real life, data are assumed to be heterogeneous, unstructured, contains noise and uncertainty. Real life data taken from set of entities are related to each other. Moreover, it can be stored in different multiple relations, which are complex to understand and attributes of these relations are dependent on each other. The need to model such dependencies between entities has led to the emergence of a subarea in machine learning called Statistical Relational Learning (SRL). In SRL, it is assumed to forget i.i.d. assumption for data, which is made almost throughout in machine learning research. Combination of statistical learning and relational learning give SRL. Statistical learning deals with data uncertainty and relational learning address complex relational structures. SRL is a powerful tool to represent, model and learn complex relational structures. There are various methods available in SRL for representation or formalism, learning and inference.

In this talk, we first introduce graphical structures of SRL.  We examine different formalisms and representations developed in SRL. For specific study, we first investigate logical and relational extensions of Baysian networks that include Bayesian logic programs and probabilistic relational models. Lastly we upgraded Markov networks to get Markov logic network.

 


From Bioinformatics to Systems Biology

Prof. Mahesan NiranjanUniversity of Southampton, United Kingdom.

In recent years, we have seen rapid growth in the amount of biological data available to us. Rapid advances in instrumentation coupled with increasing ability to archive and distribute such data over the World Wide Web, and the systematic efforts going into collecting meta-data have contributed to this.  Such data relate to the nucleotide sequences of thousands of organisms as well as individual humans, amino acid sequences of proteins, macromolecular structures, high throughput experimental measurements of gene and protein expression, and curated annotations of cellular functions curated from literature. It is widely acknowledged that extracting useful information from such datasets will involve extensive mathematical and computational modelling. Whereas past effort in the area of bioinformatics has primarily focused on the storage and distribution of data, modelling to understand cellular functions at the systems level is seeing increasing research activity. In this talk, I will review some basic principles and challenges faced in current research activity in the area. I will also describe in detail three problems of interest to my group: morphogene propagation in development, estimation of parameters for systems biology models and the regulation of cellular protein concentration.
 

Registration Fees
Category Dates Local (Rs) International (US$)
Full Package 11-14 Dec 20,000 200
Two-Day Conference Participation 12-13 Dec 15,000 150
One-Day Conference Participation 12 or 13 Dec 10,000 100
Full-Day Tutorial/Workshop 11 or 14 Dec 5,000 50
Half-Day Tutorial/Workshop 11 or 14 Dec 3,000 30
Conference Dinner Ticket 12th Dec 5,000 50
Conference Tour Ticket 15th Dec 8,000 80

 

You can get the registration form and submit here.