• Home
• News
• People
• Research
• Projects
• Publications
• Grants
• Seminars
• 2009
• 2008
• 2007
• 2006
• 2005
• Media
• Students
• Smart Services CRC
• Undergraduate Projects
• Gallery
• CHAI Software
• Group Resources
• Admin

CHAI researchers new Smart Services Projects

Read More

Can Computers Teach You To Think and Care? Revisiting The Modeling Debates with an Eye to the Future (Joint CoCo Seminar)
Susanne P. Lajoie

Wed 11 November 11:00AM

Seminar Details

CHAI Seminar Schedule 2009

Here is the schedule for the Computer Human Adapted Interaction Research Group seminars. They range from formal presentations of mature work by members of the group or visitors to less formal reports of work-in-progress. The regular meeting time is 11:00am Wednesday in the School of IT Building Seminar Room 123.

Subscribe to the CHAI Seminar Calendar

• RSS Feed
• iCal
• HTML Google Calendar

Weekly Seminars

Special Seminars

:: Towards ubiquitous task management

Speaker:
Christina Yum

Description:
At present, people employ a number of strategies to define and manage their tasks over both physical and digital workspaces. Between our personal and professional lives, we are required to divide our attention between numerous demands and responsibilities. We are at a point in time where how we manage these is changing. As computing moves into the cloud and smartphones become more usable and powerful, opportunities to improve our task management practice are presented. There is great potential for computer-based task management tools, such as online to-do lists, to improve the productivity of individuals and the co-ordination amongst teams, groups and families.

We present a Taxonomy of Ubiquitous Task Management as an important direction towards unifying the existing work and directing future developments. We present a field study which contributes to the understanding of how computer-based task management tools can support task management at a desktop and when mobile, and the usefulness of our Taxonomy to analyse the qualitative data produced.

Speaker Bio:
Christina is a BIT student completing her Honours degree at CHAI under supervision of Professor Judy Kay.

:: Moving Beyond Appliance Computing on Tabletop Systems

Speaker:
Chris Ackad

Description:
Collaborative tabletops, Single Display Groupware devices, enable people to collaborate at an interactive table. To date, such tabletop computing research has been dominated by an appliance model, exploring new interaction approaches within a single application. Tabletops have created the need to explore new ideas for interaction in this design space. Considerable research has explored aspects of this design space ranging from computer human interaction techniques, facilitating collaboration between people on the tabletop to user interface design. Tabletops have the potential to be integrated into a ubiquitous environment to provide a seamless interaction with other technologies. However, little work has been conducted in providing core functionality to the tabletop that is common to desktop computers and other devices.

We are exploring ways to move beyond this, supporting the user in changing between arbitrary applications, making use of different file-sets and altering the facilities available at the tabletop. Such actions are available on desktops, but the particular constraints and affordances of tabletops provide different design constraints for interface elements that can support this functionality.

Speaker Bio:
Christopher Ackad is a fourth year Bachelor of IT student who is currently in his honour's year under the supervision of Professor Judy Kay.

:: INDICATE: A High-Level Language and Architecture for specifying tabletop gestures

Speaker:
Paul Sztajer

Description:
As tabletop hardware becomes more prevalent in society, the need to understand gestures and their properties, affordances and constraints becomes increasingly important. Currently, a major barrier to exploring the interaction space of tabletop gestures is the difficulty involved in creating, editing and adapting gestures to new hardware devices. In response to these issues, we present Indicate: a high-level language which defines gestures and an accompanying architecture to interpret this language and find gestures on tables with arbitrary hardware setups. We evaluate the system's feasibility and value through the use of a prototype, and determine its suitability as an aid to interface prototyping.

Speaker Bio:
Paul Sztajer is in his 5th and final year Software Engineering/Science degree at Sydney University. Currently he is completing his treatise under the supervision of Judy Kay, Markus Rittenbruch and Yash Shrivastava.

:: Honours Presentation Practice Talk

Speaker:
Glen Pink

Description:
To optimally personalise services for users on mobile devices, as much information as possible about context and users must be utilised. This includes data that would be considered private to the user: a user would not want any of this data transferred off their personal device. Hence personalisation should be done on the device. As contexts are constantly changing, and new contexts are experienced, not only must content be updated but the way in which personalisation occurs must also be updated. However, allowing code to be transferred onto mobile devices and be executed creates potential for malicious code to do great damage to the device. It is vital that dynamic code is prevented from doing any damage. This talk presents a dynamic code loading framework that is reasonably safe, usable, and useful; that provides the ability to create applications that are more personalised and context-aware.

Speaker Bio:
Glen is currently completing his final year of a BIT as an Honours year at CHAI.

:: Data Mining for Individualised Hints in e-Learning

Speaker:
Anna Dominguez

Description:
Unlike in traditional classrooms, students using an e-Learning system often do not have direct access to help from a teacher. While mediums such as email allow for help to be provided, there is a need for immediate feedback to be given to students to aid their learning. Intelligent Tutoring Systems (ITSs) can provide such aid, but it is pre-defined, which makes ITSs slow and costly to build. In this presentation we describe the building of a hinting system that uses Data Mining on live data to provide dynamic and individualised help to students of the NCSS Challenge Website, an already existing e-Learning system, without the overhead of building an ITS. We also present our findings from a large-scale experiment carried out to evaluate the effectiveness of the system.

Speaker Bio:
Anna Katrina Dominguez is a fourth year student completing Honours as part of a BIT degree.

:: PhD Probation Review Practice Talk

Speaker:
Greg Darke

Description:
Greg Darke will give a trial run through of his presentation for his PhD Probation Review.

Speaker Bio:
Greg Darke is a PhD student in the CHAI research group. His thesis looks to creating a framework to support flexible privacy management for client side mobile personalisation.

He recently received 1st class honours for his thesis on creating a new file access mechanism that enables people to achieve a new, faster, natural means to access, or, re-find, their files than is currently possible.

:: Curriculum mapping for modelling long term learning by degrees

Speaker:
Richard Gluga

Description:
The presentation begins with a discussion of the complex operational environments and complex educational data structures within a typical University. The problem explored is the lack of a centrally managed system for mapping detailed information for University degree programs and Units of Study, and thus the lack of end-to-end capability/attribute mapping. A background study of existing Learner Management Systems reveals a strong focus on Units of Study as the top level containers, with very little treatment given to non-linear multi-year degree program structures. Likewise, current capability/attribute mapping efforts are usually localized to a group of related units or a single discipline. There are few examples of capability/attribute mapping systems that operate at a program level, or in a cross-discipline context. This level of mapping is required to support accreditation reviews and to aid students in making informed enrollment decisions.

The rest of the presentation will discuss a two-phased system implementation to address this problem. Namely, a cross-discipline, cross-Faculty curriculum management system, and the application of end-to-end capability/attribute mapping. Phase 1 of the solution is nearing production deployment to three Faculties within the University of Sydney, while Phase 2 will focus on capitalizing on the capability mappings to create open learner models for students to explore.

Speaker Bio:
Richard Gluga is a M.Phil. student with the Computer Human Adapted Interaction (CHAI) Research Group at the University of Sydney. In 2006 he completed a Bachelor of Software Engineering and Science (Computer Science) degree (also at the University of Sydney), and graduated with first class honours. His Engineering thesis was in the area of template-driven educational Content Management Systems. Between 2007 and 2009, Richard worked for IBM as Enterprise Integration Consultant, after which he returned to Sydney University to pursue a Research Masters whilst also performing some design and development for the Faculty of Engineering.

:: Understanding File Access Mechanisms for Embedded Ubicomp Collaboration Interfaces

Speaker:
Anthony Collins

Description:
Anthony will give a practice talk for a paper that he will present at UbiComp 2009, the 11th conference on Ubiquitious Computing, held in October in Orlando, Florida, USA.

Emerging ubicomp environments will include a range of new ways for people to interact with their digital artefacts: at interactive tables, wall displays and making use of carried projectors. These will complement people's increasingly sophisticated carried devices, such as smartphones, as well as more conventional portable computing devices such as laptops. Unlike a carried device, these embedded interaction displays will provide large screens that are effective for people to collaborate on tasks like planning and to share information and discuss it. If this is to be possible, an essential prerequisite is an effective mechanism for people to access their files. At this stage, it is unclear how to support collaborative file access for Single Display Groupware (SDG).

This paper explores the nature of interfaces to support people in accessing their files at tabletop displays embedded in the environment. To do this, we designed a study comparing people's interaction with two very different classes of file system access interface: Focus, explicitly designed for tabletops, and the familiar hierarchical Windows Explorer. In our within-subjects double-crossover study, participants collaborated on 4 planning tasks. Based on video, logs, questionnaires and interviews, we conclude that both classes of interface have a place. Notably, Focus contributed to improved collaboration and more efficient use of the workspace than with Explorer. Our results inform a set of recommendations for future interfaces enabling this important class of interaction -- supporting access to files for collaboration at tabletop devices embedded in an ubicomp environment.

Speaker Bio:
Anthony Collins is a PhD candidate with the Computer Human Adapted Interaction (CHAI) Research Group at the University of Sydney. In 2006 he completed a Bachelor of Computer Science and Technology (Advanced) degree (also at the University of Sydney), and graduated with first class honours. His honours thesis was in the area of tabletop interaction, and explored the issue of supporting collaborative file access around a multi-user tabletop interface. In 2007 he started his PhD under the supervision of Professor Judy Kay, also in the area of collaborative tabletop information retrieval.

:: Curator: Design Environment for Curating Tabletop Museum Experiences

Speaker:
Benjamin Sprengart

Description:
In this presentation, Benjamin Sprengart will show the achievements of his Bachelor thesis "Curator: Design Environment for Curating Tabletop Museum Experiences".

The application developed in Benjamin's thesis, Curator, is a design tool for curating tabletop museum experiences. The program enables users with basic computer knowledge to easily and quickly create data-sets for the Cruiser/Focus tabletop system. A short demo of this program is shown in this presentation as well as the results of two user studies which helped to evaluate Curator.

After the presentation, an example data-set running on Focus will be shown.

Speaker Bio:
Benjamin Sprengart is an undergraduate student at the University of Applied Sciences Kaiserslautern, location Zweibruecken. He is completing a Bachelor of Science specialised Media Informatics. Currently Benjamin is writing his final thesis in the CHAI research group.

:: My experience at CHAI, the project and Australien and Germany career to study

Speaker:
Sonke Gissel

Description:
Last 5 Months Sonke spend at CHAI to work on an update for TADA-Ed - the CHAI developed tool for educational data mining. The presentation will cover some of the problems I have faced during some stages and my attempts to solve it.

As this is my first time in Australia I will also cover some of the interesting differences I come across during my stay here.

Speaker Bio:
Sonke Gissel is a Bachelor of Science student at BHT Berlin and is having an internship at CHAI under supervision of Dr. Kalina Yacef.

:: Mining Student's Learning Behaviour from Educational Data

Speaker:
Rajibussalim

Description:
Within the last two decades, researchers began to realise the potential of useful knowledge and information that can be discovered from educational data. However, simple statistical processes are more likely unable to do the task to extract the rich information from a large amount of data from a database. The process requires a more powerful technique to be implemented such as data mining. Reflect is one educational system used for a number of years at the University of Sydney as an additional source of learning and practice to the classroom teaching. Students have been using the system for years and a large amount of students' interaction data have been collected. This research investigates how some data mining techniques could be used to learn student's data interaction from Reflect in order to search for previously unknown, potentially useful pedagogical knowledge hidden in the data.

Speaker Bio:
Rajibussalim is a Master of Science (Research) student working under supervision of Dr. Kalina Yacef and Prof. Judy Kay.

:: Google Wave Developer Day (Sydney)

Speaker:
Greg Darke

Description:
An overview of the Wave Developer Day in Sydney, including the Hackathon. This presentation will show some of the more interesting demonstrations from the day, a quick overview of the API and the current state of the Gadget and Robot written by Greg.

:: Trip Report from UMAP - User Modelling, Adaptation and Personalisation, Trento, Italy

Speaker:
Judy Kay and Bob Kummerfeld

:: Pervasive Computing

Speaker:
Judy Kay

Description:
This seminar will provide an overview of INFO5047 Pervasive Computing. The goal is to ensure that all members of the group can contribute suggestions and the possibility of adding links to our research.

Speaker Bio:
Judy Kay is Professor of Computer Science and a principal in CHAI. Her research aims to support a highly personalised way for people to interact with the computers that they carry, those embedded within the environment as well as desktop computers. Critical to this vision is that the user should always feel a sense of control over the personalisation. She creates new tools for interaction in support of lifelong learning, collaboration and personalised smart services.

:: WWDC 2009 Trip Report

Speaker:
James Bunton

Description:
James will give a trip talk covering the most exciting new technologies that Apple announced and demoed at the recent World Wide Developer's Conference in San Francisco.

Speaker Bio:
James Bunton is a PhD student in the School of IT studying in the CHAI Research Group.

:: Enhance learning of programming with less work - utilize visualization with automatic assessment!

Speaker:
Mikko-Jussi Laakso

Description:
Learning to program is one of the key requirements in computer science education. However, novice programmers often find it a difficult task to master. There isn't probably any single solution to solve this problem, but it is still important to develop and research new means and techniques to further improve the learning experience and outcome.

In this seminar, the engagement taxonomy, the capabilities of automatic assessment and a visualisation tool (ViLLE) will be presented and discussed in terms of the benefits to the teacher and also for student learning.

ViLLE, a dynamic program visualisation tool developed at University of Turku, Finland. ViLLE is a language-independent visualization tool aiming at providing a more abstract view of programming, and it can be utilized both in lectures and for independent learning. It has a built-in syntax editor, with which the user can add new languages to the tool or modify the syntax of the built-in languages, including e.g. Java, C++ etc. The visualizations can be viewed in any of the (user or pre-) defined languages. To emphasize the language independency, ViLLE has a parallel view in which the execution of a program and the program code itself can be viewed simultaneously in two languages.

While the execution progresses, the user can observe program outputs and changes in variable values. In addition, we can engage the learner with pop-up questions with automated assessment capabilities.

Speaker Bio:
Mikko-Jussi Laakso is currently working as a researcher at University of Turku and as a researcher at Helsinki University of Technology. His research interest covers program and algorithm visualization, learning environments, computer aided and automatic assessment in computer science education. Currently he has published 5 international journal papers and 8 international conference papers and 6 other publications.

:: Moving Beyond Appliance Computing on Tabletop Systems (Research Methods Practice Talk)

Speaker:
Chris Ackad

Description:
Tabletop computing is a fairly new form of interactive computing where users interact with a tabletop computer through the use of touch or pen interfaces. The goal of this project is to develop an application launcher and switcher for the Cruiser system that allows users to easily launch and manage applications installed on the tabletop computer.

Very little work has been done in exploring how best to support functionality offered by elements which are essential to a desktop operating system in the tabletop environment. The next stage of tabletop development is to move away from the perception that a tabletop computer is a single use application to that of a universal operating system. Currently there are many different applications for the Cruiser tabletop computer, one of which is Focus, as well as various other applications that have been developed at the University. At the moment there is no seamless way of launching an application or switching between applications. The project work will involve the development of a usable interface that allows users to easily launch and switch between applications nor is there a way to alter the dataset used by an application, as well as the required backend system, and finally to evaluate the usability of its interface in realistic scenarios.

Speaker Bio:
Christopher Ackad is a fourth year Bachelor of IT student who is currently in his honour's year under the supervision of Professor Judy Kay.

:: Large-scale, long-term learner models supporting flexible curriculum definition (Research Methods Practice Talk)

Speaker:
Richard Gluga

Description:
A University Degree is bound by many internal and external syllabus requirements (graduate attributes, generic skills, professional skills, accreditation competencies). These requirements come from many different sources (University, Faculty, School & external bodies like ACM, ASC, IEEE, EA, etc.). They are defined in different terminology, but are often semantically related. The core and elective subjects part of a degree must satisfy all of these requirement sets, but the semantic mappings and relationships are missing in current systems. We intend to apply open learner modeling techniques and Web 2.0 technologies to develop a curriculum mapping system that addresses these problems and allows for flexible user introspection of complex curriculum information.

Speaker Bio:
Richard Gluga is a M.Phil. student with the Computer Human Adapted Interaction (CHAI) Research Group at the University of Sydney. In 2006 he completed a Bachelor of Software Engineering and Science (Computer Science) degree (also at the University of Sydney), and graduated with first class honours. His Engineering thesis was in the area of template-driven educational Content Management Systems. Between 2007 and 2009, Richard worked for IBM as Enterprise Integration Consultant, after which he returned to Sydney University to pursue a Research Masters whilst also performing some design and development for the Faculty of Engineering.

:: A holistic understanding of personal task management and the role of mobile devices (Research Methods Practice Talk)

Speaker:
Christina Yum

Description:
At present, individuals and teams employ numerous strategies to manage their tasks over both physical and digital mediums. There is much potential for computer-based "to-do list" tools to maintain links between tasks, resources, events, contacts and other associated digital information to improve the productivity of individuals and the co-ordination amongst team members. A field trial where people make use of a set of tools for planning their tasks over multiple machines and using an iPhone will also be completed to learn how different devices are used in personal task management. These studies will give a better understand of how these tools could better support the management and completion of tasks both in a team and individual setting. This project is important as task list creation and management is extremely diverse amongst individuals and teams.

Speaker Bio:
Christina is a Bachelor of Information Technology student completing an Honours year with the CHAI research group. She joined the group to work on the Keep In Touch trial as a vacation scholar in December 2008 after completing her undergraduate BIT degree at the University of Technology, Sydney.

:: CHI 2009 Trip Report - Part 2 (General Papers)

Speaker:
Anthony Collins

Description:
Anthony will give a trip report of CHI 2009, the SIGCHI conference on human factors in computing systems, held in Boston Massachusetts, USA. In this final instalment, he will give an overview of the general HCI papers presented this year.

Speaker Bio:
Anthony Collins is a PhD student with the Computer Human Adapted Interaction (CHAI) Research Group at the University of Sydney. In 2006 he completed a Bachelor of Computer Science and Technology (Advanced) degree (also at the University of Sydney), and graduated with first class honours. His honours thesis was in the area of tabletop interaction, and explored the issue of supporting collaborative file access around a multi-user tabletop interface. In 2007 he started his PhD under the supervision of Professor Judy Kay, also in the area of collaborative tabletop information retrieval.

:: CHI 2009 Trip Report - Part 1 (Tabletop Papers)

Speaker:
Anthony Collins

Description:
Anthony will give a trip report of CHI 2009, the SIGCHI conference on human factors in computing systems, held in Boston Massachusetts, USA. In part 1 (of 2), he will give an overview of the papers related to tabletop interaction. He will also discuss the papers from the Multitouch and Surface Computing workshop that he attended.

Speaker Bio:
Anthony Collins is a PhD student with the Computer Human Adapted Interaction (CHAI) Research Group at the University of Sydney. In 2006 he completed a Bachelor of Computer Science and Technology (Advanced) degree (also at the University of Sydney), and graduated with first class honours. His honours thesis was in the area of tabletop interaction, and explored the issue of supporting collaborative file access around a multi-user tabletop interface. In 2007 he started his PhD under the supervision of Professor Judy Kay, also in the area of collaborative tabletop information retrieval.

:: Large-scale, long-term learner models supporting flexible curriculum definition

Speaker:
Richard Gluga

Description:
A University Degree is bound by many internal and external syllabus requirements (graduate attributes, generic skills, professional skills, accreditation competencies). These requirements come from many different sources (University, Faculty, School & external bodies like ACM, ASC, IEEE, EA, etc.). They are defined in different terminology, but are often semantically related. The core and elective subjects part of a degree must satisfy all of these requirement sets, but the semantic mappings and relationships are missing in current systems. We intend to apply open learner modeling techniques and Web 2.0 technologies to develop a curriculum mapping system that addresses these problems and allows for flexible user introspection of complex curriculum information.

Speaker Bio:
Richard Gluga is a M.Phil. student with the Computer Human Adapted Interaction (CHAI) Research Group at the University of Sydney. In 2006 he completed a Bachelor of Software Engineering and Science (Computer Science) degree (also at the University of Sydney), and graduated with first class honours. His Engineering thesis was in the area of template-driven educational Content Management Systems. Between 2007 and 2009, Richard worked for IBM as Enterprise Integration Consultant, after which he returned to Sydney University to pursue a Research Masters whilst also performing some design and development for the Faculty of Engineering.

:: ExploreME: Building a Software Framework for Tabletop Interactive Displays in Museums

Speaker:
Benjamin Sprengart

Description:
Use of tabletop displays with new interaction primitives is undoubted. That is why many museums provide additional information on exhibits using this method of presentation to extend their attractiveness.

However, it is a real challenge for exhibition designers to create a tabletop exhibition. E.g. contents of the virtual exhibition may be changed due to new attainments or because the physical exhibition has expired. Exhibition designers are no programmers - so there is a need to give them the possibility to easily compose a virtual exhibition.

ExploreME is a framework for creating interactive tabletop displays for museums. It offers the possibility to provide additional information on exhibits as well as points of interest in a museum. Based on these documents the framework makes it easy for the exhibition designer to select an index to associate related documents for inviting the visitor to explore the exhibits, their background and related digital artifacts.

Speaker Bio:
Benjamin Sprengart is an undergraduate student at the University of Applied Sciences Kaiserslautern, location Zweibruecken. He is completing a Bachelor of Science specialised Media Informatics. Currently Benjamin is writing his final thesis in the CHAI research group.

:: Data mining in e-learning systems to assist future learners

Speaker:
Anna Dominguez

Description:
The use of tutoring software for educational purposes has become increasingly common in recent years. Some advantages of using such software are fairly obvious; for example, students' work can be marked automatically in many cases, and unlike in a classroom situation, students can work at their own pace. Aside from this, however, the use of these systems presents many interesting opportunities. Content and structure do not have to be static, and as such the educational systems can be adapted to individual users to improve their learning experience. This project looks at the collection and analysis of data from such a system, specifically in the context of the NCSS Challenge Website. We investigate how data mining carried out on records of students' interaction with the website and their answers to questions can find patterns and trends, and how such patterns can help future users. We look to create a more personalised learning environment for students, and explore how the trends in the data can be used to better tailor tutoring facilities to each specific student.

Speaker Bio:
I am a 4th year BIT student at the University of Sydney. I'm currently completing Honours at CHAI as part of the final year of my degree.

:: Indicate: A gestural language for tabletop developers

Speaker:
Paul Sztajer

Description:
The development of novel interfaces involving the use of gestural inputs is a trend in software research today. With increasing functionality and power, the need for prototyping gestures has become paramount. However, researchers developing such interfaces are inevitably forced to focus much on their work on the development of algorithms to detect such gestures, as existing systems are generally limited to a specific piece of hardware or type of gesture. A need exists for a system that interprets a high-level specification of a large number of gestures and exhibit specified behaviour when a gesture is detected on a wide variety of hardware devices.

This project concerns the development of Indicate: a language for describing tabletop gestures, and a tool which translates gestures into actions for existing software frameworks.

Speaker Bio:
Paul is a 5th year Software Engineering/Science student at the University of Sydney. He is currently completing his final year engineering project with CHAI and the Braccetto project (NICTA) under the supervision of Judy Kay, Markus Rittenbruch and Yash Shrivastava.

:: Execution of Potentially Malicious Code on Mobile Devices

Speaker:
Glen Pink

Description:
As smart phone use becomes increasingly prevalent, so does the need for applications for these devices. Allowing development of software for these devices by third parties creates the potential for the engineering and distribution of malware. Hence distributing and running these applications requires the source of the of these applications to be trusted. There are currently several approaches for attempting to create trust including manual code checking, but these have disadvantages and potential for error. We look at approaches for minimising the danger of running arbitrary code on mobile devices, motivated by producing a secure python interpreter for use on mobile devices.

Speaker Bio:
Glen is currently undertaking a his final year of a BIT as an Honours year at CHAI. He has previously worked on the Locator project at CHAI and is now working in mobile security.

:: Moving Beyond Appliance Computing on Tabletop Systems

Speaker:
Chris Ackad

Description:
Tabletop computing is a fairly new form of interactive computing where users interact with a tabletop computer through the use of touch or pen interfaces.

The goal of this project is to develop an application launcher and switcher for the Cruiser system that allows users to easily launch and manage applications installed on the tabletop computer.

Very little work has been done in exploring how best to support functionality offered by elements which are essential to a desktop operating system in the tabletop environment. The next stage of tabletop development is to move away from the perception that a tabletop computer is a single use application to that of a universal operating system. Currently there are many different applications for the Cruiser tabletop computer, one of which is Focus, as well as various other applications that have been developed at the University. At the moment there is no seamless way of launching an application or switching between applications. The project work will involve the development of a usable interface that allows users to easily launch and switch between applications nor is there a way to alter the dataset used by an application, as well as the required backend system, and finally to evaluate the usability of its interface in realistic scenarios.

Speaker Bio:
Christopher Ackad is a fourth year Bachelor of IT student who is currently in his honour's year under the supervision of Professor Judy Kay.

:: A holistic view of personal task management

Speaker:
Christina Yum

Description:
At present, individuals employ a number of strategies to manage, track and support completion of their tasks over both physical and virtual workspaces. Each task involves intricate connections between associated information and resources, and usually has dependency on other people. There is much potential for a computer-based task management tool, with the capability to infer and maintain links between tasks, resources, contacts and other associated information, to improve the productivity of individuals and the co-ordination amongst team members. Despite this, there are many tools designed for this purpose, both in the academic and commercial world that have failed to attract widespread acceptance and use. This project will take a holistic approach to understanding this class of tools through the creation of a taxonomy to classify and distinguish between these tools. Design features as well as relevant human factors and the team dynamics that affect task list management will also be assessed in a user study to look at how a particular tools is used in practice.

Speaker Bio:
Christina is a Bachelor of Information Technology student completing an Honours year with the CHAI research group. She joined the group to work on the Keep In Touch trial as a vacation scholar in December 2008 after completing her undergraduate BIT degree at the University of Technology, Sydney.

:: Explanation Subsystems in Learner Modeling Systems

Speaker:
Preet Kukreti

Description:
Long term learner modeling systems gather information about students over time and use domain-specific rules to show a student their progress or suggest further actions. Existing research shows that is desirable for learner modeling systems to be 'open', allowing the students to not only scrutinize their own learner model but to also see how these learner models are evaluated by the system. Explanation subsystems support scrutability in learner modeling systems and provide an interface to users that enables such scrutinisation.

We will implement an explanation subsystem for an existing learner modeling system called Reflect - a system designed to aid students in learning computer programming. The design and implementation will explore and build upon past work on scrutability in interfaces, scrutable modeling frameworks, and explanation in personalised systems. Evaluation will consist of user studies to analyse the effectiveness of explanation on scrutability in the learner modeling system.

Speaker Bio:
This is my final year of a Bachelor of Software Engineering at the University of Sydney. Presently I am writing my treatise at CHAI as a part of my course requirements.

:: Creating a scrutable model of an individual's social network as a basis for improving Personal Information Management (PIM)

Speaker:
Calvin Ikhwan

Description:
Personal information encapsulates many facets of a person's life including personal friends, family, acquaintances, as well as business partners/clients. The project in this treatise deal with a person's interactions with these people and the respective spaces they are in. As email is becoming an increasingly important part of everyday life, its increasing share of communication traffic will be able to provide a reasonable source of evidence for user modelling. What are being modelled then are the everyday interactions and the sequence of events that are captured through email, be it ordinary or extraordinary. This treatise aims to investigate the use personal information (email and its metadata) to build and maintain a scrutable model of a person and use this to enhance Personal Information Management by focusing on the person's e-mail/social network.

We will write a program examine e-mail and its metadata to extract any evidence of a person's closeness to another. A framework for building context-aware system, the PersonisAD will then be used to integrate the diverse types of evidences to develop a model of a person's social network. In order to ensure control of personal information, the system will be scrutable, allowing the person being modeled to determine the most accurate model of their social network. This model can then be possibly be utilised to improve PIM. A user-study of the system will be done at the end to evaluate the usefulness of the model.

Speaker Bio:
I am currently completing my degree in Bachelor of Engineering (E-Commerce). This treatise is done as part of my course requirements.

:: Low-Cost Multi-Touch Tabletop & Microsoft Surface SDK

Speaker:
Benjamin Sprengart

Description:
Benjamin will present the work of his former projects at University of Applied Sciences Kaiserslautern. He will first introduce himself. After that he will introduce into two techniques of multi-touch and show how his group finally build a table based on Diffused Illumination. The focus of the second part of Benjamin's presentation will be on a short introduction to Microsoft's Surface and the included SDK.

Speaker Bio:
Benjamin Sprengart is an undergraduate student at the University of Applied Sciences Kaiserslautern, location Zweibruecken. He is completing a Bachelor of Science specialised Media Informatics. Currently Benjamin is writing his final thesis in the CHAI research group.

:: Heuristics to Support Design of New Software for Interaction at Tabletops

Speaker:
Anthony Collins

Description:
Anthony will give a practice talk for his presentation at the CHI 2009 Workshop on Multitouch and Surface Computing, to be held in Boston, Massachusetts on April 4, 2009.

Heuristic Evaluation is a 'discount' usability test that can support improved design decisions early in the development cycle. It is particularly a technique to support software design for tabletop interaction because the field is so new and there is potential to explore many new ideas for interaction. This makes Heuristic Evaluation particularly valuable. Many sets of heuristics have been proposed in the past, both for interfaces in general (not just computing interfaces), and ones that are more specialised. For tabletop, and other horizontal interactive interfaces, current sets of heuristics fall short. In this talk, I will discuss and build from previous sets of relevant heuristics, to formulate a new set of heuristics for software design for tabletop interaction.

Speaker Bio:
Anthony Collins is a PhD student with the Computer Human Adapted Interaction (CHAI) Research Group at the University of Sydney. In 2006 he completed a Bachelor of Computer Science and Technology (Advanced) degree (also at the University of Sydney), and graduated with first class honours. His honours thesis was in the area of tabletop interaction, and explored the issue of supporting collaborative file access around a multi-user tabletop interface. In 2007 he started his PhD under the supervision of Professor Judy Kay, also in the area of collaborative tabletop information retrieval.

:: Using Linguistically Motivated Features in Document Retrieval for Question Answering

Speaker:
Luiz Pizzato

Description:
In this talk I will present the work conducted for my thesis where I investigated the impact of using linguistic features in the Information Retrieval (IR) stage of Question Answering (QA) systems. We hypothesise that techniques that are commonly used in the final answer extraction stage can improve the overall results of a QA system when adopted in the earlier IR stage. In particular, we study the use of the following information: i) named entities in a pseudo-relevance feedback process; and ii) semantic relations between words of questions and text sentences.

The study of the use of named entities is inspired by the common practice of filtering out sentences that do not contain the expected answer type. We consequently introduce a pseudo-relevance feedback that inserts entities of the correct answer type in the original query. Our experiments show that this technique leads to a query drift and the final results do not improve with respect to a query without additional feedback.

To study the use of relational information, we design an IR framework that is more efficient (in both speed and memory consumption) than standard approaches based on relational databases and on the concatenation of word pairs at the indexing stage. The resulting framework allows a multi-layer index that uses an extension to the standard vector space model as a ranking strategy. The resulting ranking strategy improves precision, without compromising the overall recall, by including linguistic word relations.

We present the Question Prediction Language Model (QPLM), a model of relational information that borrows concepts from Semantic Role Labelling (SRL) but is designed for the fast generation of annotation and its use for indexing and retrieval. The results are of quality comparable to SRL and indicate that linguistic information encoded in the form of semantic relations does enhance the retrieval quality of text and the final accuracy of QA systems.

Speaker Bio:
Luiz received his Bachelor of Computer Science in 2000 from the Pontifical Catholic University of Porto Alegre (PUCRS), Brazil. In the same year, he joined the Hewlett Packard/PUCRS Research Centre in High Performance Computing. In 2003, Luiz received a Master of Computer Science from PUCRS for his thesis involving information retrieval (IR) and thesauri information. During his Masters degree he also developed the Folha-RIcol corpus, which has been used by different researchers to evaluate IR systems for the Brazilian Portuguese language. In 2003, Luiz integrated his Masters research with the SINO search engine to enable the online search of legal decisions made by the Portuguese Attorney General. In 2008 at Macquarie University, Luiz submitted his PhD thesis which focused on using language information in the IR stages of the question answering task. Luiz has also produced over a dozen peer-reviewed publications.

:: Trip Report for Army Research Institute, Adaptive Training Technology Workshop, Charleston, South Carolina, March 3-5, 2009

Speaker:
Judy Kay

Description:
"Advances in cognitive science combined with ever-increasing information processing power make it feasible to tailor computer-aided instruction to the needs of individual users. The goal of the workshop is to bring together experts to present their pertinent work and help determine future directions for research and development in the area of adaptive training technology. Our long term goal is to foster more efficient and effective training, based in sound student and pedagogical models." [From the abstract for the workshop]

Speaker Bio:
Judy Kay is Professor of Computer Science at the School of Information Technologies (formerly, Basser Department of Computer Science) at the University of Sydney. She is a principal in the CHAI: Computer Human Adapted Interaction Research Group, leading research in advanced technologies for human computer interaction, supporting personalisation, pervasive and mobile interaction. Her vision is to support a highly personalised way for people to interact with the computers that they carry, those embedded within the environment as well as desktop computers. CHAI researchers are exploring ways to ensure that people can readily make use of these as part of their own personalised augmented memory. We aim to put a personalisation layer over these devices so that each person can make the most effective use of them for key aspects of their lives. Critical to this vision is that the user should always feel a sense of control over the personalisation. She creates new tools for interaction in support of lifelong learning, collaboration and personalised smart services.

:: Introduction to Context Aware Applications for the Andriod Platform

Speaker:
Ben Taylor

:: Educational Interfaces for the Cruiser Tabletop

Speaker:
Sam Harding

:: Content Repurposing for the Mobile Web

Speaker:
Lynden Shields

Description:
The trend of the World Wide Web has been slowly but surely confirming Tim Berners-Lee's fear: spliting of the Web. It is not uncommon to see a separate version of website designed especially for, say, the iPhone. This created a whole different set of standards and formats for the so-called "mobile Web", making it difficult to have a consistent Web presence across a spectrum of mediums and devices. This talk goes over my research into solutions that others have come up with and why they are not yet ideal, the demonstrator made for the purpose of illustrating what I believe to be the ideal solution, as well as identifying future work that might result in the ideal solution.

Speaker Bio:
Currently between second and third year of studying for a Bachelor of Computer Science and Technology Advanced. My summer scholarship is in the area of making it easier to have a consistent Web presence for both the Desktop Web and the Mobile Web.

:: Overview of the Smart Services CRC Education and New Media Project

Speaker:
Peter Reimann

Description:
The overall project aim is to develop tools and services that add value to digital collections as provided in educational repositories, such as learning objects and other assets. The project conducts research not only into how the resource providers (such as a government agency) a collection) can provide services to its educational clientele (i.e., teachers and students), such as integration with lesson and course planning, collaboration services (catering to team learning), assessment services, etc, but also how end users' activities and products can be integrated with the "collection" in order to add value to (the educational use of) the collection. Using web 2.0 approaches and technologies, the research will probe into issues such as how to motivate and manage user contributions, how to create and maintain relations with online user communities (such as teacher communities), how to export and import content to/from 'social spaces' (such as facebook, myspace, SecondLife), how to integrate content from Learning Objects/Learning Designs repositories, etc.

In the first year, the research is particularly concerned with (a) how to make content available across a variety of immersive learning environments (including Second Life), (b) how to add annotation and tagging services to immersive learning environments, and (c) how to support teachers (and instructional designers) in designing for learning experiences in immersive environments.

Speaker Bio:
Peter is currently employed as Professor of Education in the Faculty of Education and Social Work, University of Sydney. Peter studied Psychology at the University of Freiburg (Germany) and received his PhD from the same university. He worked for a number of years as assistant professor in Cognitive Psychology in Freiburg, and between 1996 and 2003 was Professor of Educational Psychology at the University of Heidelberg (Germany). He also spent two years as a researcher at the Learning Research and Development Centre, Univ. of Pittsburgh (USA). Peter's primary research areas are learning research with a focus on educational computing, multimedia-based and knowledge-based learning environments, e-learning, and the development of evaluation and assessment methods for the effectiveness of computer-based technologies. Current research activities comprise, among other issues, the analysis of individual and group problem solving/learning processes and possible support by means of ICT, and analysis of the use of mobile IT in informal learning settings (outdoors, in museums, etc.). He's also interested in e-research methods for learning research, in particular educational data- and text mining. In addition to basic research, Peter has been involved in a number of European Commission-funded R&D projects and worked frequently as consultant for the European Commission's educational technology research programs. He has also done consultancy for numerours industry partners, and more recently for the Singapore Ministry of Education as well as The Learning Federation. Since 2003, he has been co-director of the CoCo research centre at the University of Sydney, which has grown to 20+ researchers and PhD students, and also, since 2005, has directed the ICT unit of the Faculty of Education and Social Work. To learn more about the CoCo research centre and Peter's projects, please visit the CoCo web-site.

:: B2C communication using context-aware consumer shopping tools on mobile devices

Speaker:
Sam Thorogood

Description:
This talk is a very brief progress report regarding work on an iPhone application targeted for Sensis. The growth of mobile computing presents many avenues for direct and context-aware communication between business and consumer. This information, for our purposes, includes a consumer shopping list and notifications of current sales.

Speaker Bio:
Sam is currently undertaking a summer scholarship here at the University after completing his honours with Judy Kay in 2008. He intends to take some time off after the scholarship and return in May to begin his PhD.

:: ADCS 2008 Trip Report

Speaker:
James Bunton

Description:
James will give a trip report of the Australasian Document Computing Symposium held in Hobart in December 2008.

Speaker Bio:
James Bunton recently completed honours at the School of IT, majoring in Principles of Computer Science and Software Engineering. He is soon to begin a PhD in the CHAI Research Group.

:: Linux.conf.au 2009 Trip Report

Speaker:
Greg Darke

Description:
Greg Darke will give a trip report of Linux.conf.au (LCA for short) held at the University of Tasmania.

Speaker Bio:
Greg Darke recently completed his honours thesis at the School of IT, majoring in Software Engineering and Networking. He is currently doing a vacation scholarship in the CHAI research group on the Reflect learning system. He will soon be starting a PhD with the same group.

:: Educational Interfaces for the Cruiser Tabletop

Speaker:
Sam Harding

Description:
In modern day society, the world of gaming, both computer and non-digital, provides both a large learning basis and a wide platform for human interaction. By exploring the ways in which we interact through the basic principles of games, we can create higher levels of interfaces for developing IT hardware. This project is looking primarily at card games, and the simulation of real time environments to heighten the involvement of users of the cruiser tabletop. The project is approaching the gaming aspect not from the traditional 'computer screen' way as seen commonly around the world, but in a new way which is based around the unique environment of a tabletop.

Speaker Bio:
I am currently between my second and third year of a Bachelor of Advanced Sciences and a Bachelor of Arts. I enjoy my involvement in co-curricular projects, such as the Summer Vacation Scholarships, as these not only provide a large and diverse learning process, but also are very enjoyable and are very interesting. The school of IT provides many opportunities for students to explore paths which can lead to great discoveries, and it's through these paths i have found my true passion for computers.

:: B2C communication using context-aware consumer shopping tools on mobile devices

Speaker:
Sam Thorogood

Description:
The recent growth of mobile computing presents many possibilities for revolutionising traditional advertising media, especially in terms of direct information exchange between business and consumer. This information - as well as the traditional static advertisement - might include consumer shopping lists or current sales. I will speak briefly on the possibilities in this space and discuss my work in implementing a demonstrator product on the Apple iPhone.

Speaker Bio:
Sam is currently undertaking a summer scholarship here at the University after completing his honours with Judy Kay in 2008. He intends to take some time off after the scholarship and return in May to begin his PhD.

:: Introduction to Context Aware Applications for the Andriod Platform

Speaker:
Ben Taylor

Description:
the emergence of multiple new, smart mobile devices come opportunities for many new applications in Pervasive and Ubiquitous Computing. I will look at previous projects which have explored applications for mobile devices and some of the directions that this new technology is taking us. I will also take a look at the Android phone, its development environment and some of the things I have done in it so far. Finally I will introduce the demonstrator of this technology that I have and will be developing over this summer period.

Speaker Bio:
Currently between second and third year, studying for a Bachelor of Information Technology. My summer scholarship involves creating a demonstrator application for the Android OS to show how Pervasive Computing practices can be applied to real world scenarios.

:: Data Mining on Rich User Profiles for Matching

Speaker:
Anna Dominguez

Description:
The use of the Internet for social networking and continuous contact has become widespread in recent years, and most of the Web-based services that facilitate this require users to answer lengthy questionnaires to create detailed profiles. As such, there is now an extremely large amount of data available on users, ranging from purely demographic details to information on hobbies and interests. This talk describes how such profiles can be analysed using data mining techniques to find ideal matches between two people in a romantic context and how this specifically relates to the Australian dating website RSVP.com. This talk also reviews literature on Stereotypes and Lifestyle, two particularly promising approaches to grouping people for cases like RSVP, and will then present results from a preliminary investigation of RSVP data from the last 5 years.

Speaker Bio:
I am a BIT student at the University of Sydney, about to go into Honours in fourth year. At present I am just about to complete my Summer Scholarship at CHAI in which I've been involved in the Personalization project for the Smart Services CRC.

:: Multipage documents on the tabletop

Speaker:
Emma Fitzgerald

Description:
Although a number of interfaces for surface computing have been developed, to date the problem of viewing and navigating multipage documents on a tabletop has not been investigated. This talk will discuss issues related to user interfaces in a tabletop environment and two approaches to viewing multipage documents: a single page view and an N-page view. The existing Cruiser framework is used as a basis for the development of this interface.

Speaker Bio:
Emma Fitzgerald has just completed a Bachelor of Computer Engineering/Bachelor of Science and will be starting her PhD in March. She is currently undertaking a summer scholarship with CHAI.

:: Content Repurposing for the Mobile Web

Speaker:
Lynden Shields

Description:
The trend of the World Wide Web has been slowly but surely confirming Tim Berners-Lee's fear: spliting of the Web. It is not uncommon to see a separate version of website designed especially for, say, the iPhone. This created a whole different set of standards and formats for the so-called "mobile Web", making it difficult to have a consistent Web presence across a spectrum of mediums and devices. My work investigates others' attempts at finding solutions for this problem, and finding ways of extending them, especially in the area of maximising reusable content for different devices with minimum effort on the part of the web developer.

Speaker Bio:
Currently between second and third year of studying for a Bachelor of Computer Science and Technology Advanced. My summer scholarship is in the area of making it easier to have a consistent Web presence for both the Desktop Web and the Mobile Web.

:: A usability study of a communication system for older users

Speaker:
Christina Yum

Description:
Keep in Touch (KiT) was designed as an inter-generational communication system allowing older people with limited computing experience to connect with their family and friends. A trial of the system is being conducted with the aim of evaluating the current hardware, software and functionality. The trial will provide insight into the social benefits a communication device, such as KiT, could provide to older people. In addition, a review of current literature highlights the importance of a simple, unassuming user interface and tailored introduction and training strategies to assist adoption of the device amongst these users. The interface was further critiqued through a cognitive walkthrough exercise uncovering a number of problems and providing design recommendations for further development.

Speaker Bio:
Christina Yum is a Bachelor of Information Technology student beginning an Honours year at the University of Sydney in 2009. She has a broad understanding of both IT and Business subjects and has worked casually as a junior research assistant at the University since 2006. She also has experience in the industry after working for a semester at American Express in 2006 and Westpac in 2008.

:: Can Computers Teach You To Think and Care? Revisiting The Modeling Debates with an Eye to the Future (Joint CoCo Seminar)

Speaker:
Susanne P. Lajoie

Description:
Researchers in the artificial intelligence and education community ask provocative questions that move the field forward in improving the quality of teaching, training and learning across disciplines. As a response to Self's [1] call for computers that care I will address this call in the context of computers as cognitive tools. In the 90's we [2] posed the question "to model or not to model?" and considered 3 camps of researchers, those who modeled learning and implemented such models into intelligent tutoring systems that could diagnose and adapt to individual differences. The second camp were the non-modellers who thought it impossible to model human learning and chose not to have the computer diagnose errors but rather envisioned the use of technology as a cognitive tool that situated experiences for learners in authentic contexts where technology supports the social experience that serves to scaffold learners. Finally, there was a middle camp that combined cognitive apprenticeship, constructivist learning, and cognitive tools with computer-based student modeling. This last camp adheres to the belief that computers can and should serve part of the cognitive mentorship function without giving over control of the learning and assessment process to those using the system.

The second volume of computers as cognitive tools [3] was intended to break the distance between the three camps and was subtitled "No More Walls." In particular, it focused on learning paradigm shifts that were guiding the design of computers as cognitive tools. Three guiding paradigms were highlighted: information processing, constructivism, and situativity, and reviewed with respect to how designers incorporated theories into the design of computers as cognitive tools. Researchers demonstrated the value in modeling both individual knowledge construction and learning in social situations through the use of technology. Computer and human tutors, as well as peers were considered as assisting in the modeling and the new guiding question was: who or what should do the modeling? New cognitive tools were designed with multiple forms of knowledge representation and demonstrations were provided of how such representations could be used for assessment purposes.

A decade later, researchers are facing different questions. Perhaps the most complex question pertains to modeling both the affective and cognitive aspects of learning and finding ways to engage learners in a world that requires adaptivity to new information that is presented at a rapid pace with demands for multiprocessing and quick decisions. In addressing the affective component of learning, we need to remain focused on how affect influences decision-making and how individuals can be engaged as they interact with technologies [4,5,1,6]. Computers as cognitive tools need to model the intersection of heart and mind. I will address a range of modeling approaches to designing computers as cognitive tools that examine the role of affect, emotion, and culture in the context of medical decision making in contexts that require either individual decisions or joint team decisions.

References

[1] Self, J. (1999). The defining characteristics of intelligent tutoring systems research: ITSs care, precisely. International Journal of Artificial Intelligence in Education. 10(3-4), 350-364.

[2] Derry, S. J., & Lajoie, S. P. (1993). A middle camp for (un)intelligent computing. In [3] S. P. Lajoie & S. J. Derry (Eds.), Computers as cognitive tools (pp.1 -11). Hillsdale, NJ: Lawrence Erlbaum Associates.

[4] Lajoie, S. P. (Ed.). (2000). Computers as cognitive tools (vol. 2): No more walls (pp. 1-430). Mahwah, NJ: Erlbaum.

[5] Halverson, R., & Collins, A. (2006). How information technologies weaken the identification of learning with schooling. Review and Practice in Technology Enhanced Learning, 1(2), 145-155.

[6] Lepper, M. (1988). Motivational considerations in the study of instruction. Cognition and Instruction, 5(4), 289-309.

[7] Snow, R. (1989). Toward assessment of cognitive and conative structures in learning. Educational Researcher, 18(9), 8-14.

Speaker Bio:
Professor Susanne Lajoie received her Doctorate from Stanford University in 1986. She is a James McGill Research Professor and Chair of the Department of Educational and Counselling Psychology at McGill University. She is a Fellow of the American Psychological Association, appointed for her outstanding contributions to the field of Psychology as well as an Inaugural Fellow of the American Educational Research Association. Dr. Lajoie is a recipient of the McGill Carrie Derick Award for graduate supervision and teaching. Dr. Lajoie has engaged in a wide array of innovative research and scholarly activities where she designs technology rich learning environments for educational and professional practices. She uses a cognitive approach to identify learning trajectories that help novice learners become more skilled in the areas of science, statistics, and medicine. She has designed effective computer based learning environments in these domains based on her research findings. She has been invited to present her research worldwide including Australia, France, Germany, Hong Kong, Korea, Singapore, Spain, Sweden, Taiwan, Mexico, the UK and the Ukraine. She has numerous publications including 2 volumes on Computers as Cognitive tools published by Erlbaum. These volumes have highlighted the necessity for theory-driven design of technologies for education and training.

Past Seminars

• Seminars in 2008
• Seminars in 2007
• Seminars in 2006
• Seminars in 2005