Judy Kay - JITT - Just-In-Time Training projects
 Home    Bio   Research   Teaching
JITT - Just-In-Time Training

This project focuses on the problem of information overload for newcomers in an organisation. It explores the potential of workflow tools to drive temporal management of a just-in-time workplace training system, which will deliver personalised and structured presentation of organisational documents. The goal is to offer people

  • just the right information,
  • at just the right time,
  • with an appropriate modality
  • and presentation form
for the user's preferences, the importance of the information and the user's current context. The issues addressed in this project are important to most large enterprises.

This work brings together researchers in three areas: workflow (Joseph Davis), personalisation (Judy Kay) and tutoring strategies (Kalina Yacef). Our long term vision for this work is ambitious, requiring integration of research from several areas:

  • Semantic indexing of the documents so that their access can be made flexible;
  • Workflow management and information synchronisation;
  • Decision rationale maintenance, work that has largely been done in the context of requirements definition but is applicable in this work;
  • Personalisation of several aspects including the selection of documents, personalisation of the actual documents, prioritising the documents;
  • Information and knowledge sharing;
  • Information scheduling and teaching strategies which provide the core of such a system, combining the other elements.
JITT 1
Joseph Davis, Sam Holden, Judy Kay, Bob Kummerfeld, Ken Lin, Josiah Poon, Aaron Quigley, Greg Saunders and Kalina Yacef.
This project is funded by SITCRC, July 2002 - June 2003. It serves an important role in exploring the combination of workflow, personalisation and tutoring strategies for the selection and ordering of documents within an organisation.
Demonstration System for JITT - Just-In-Time Training
Admin interface

Publications:
 

Davis, J, J Kay, B Kummerfeld, K Lin, J Poon, A Quigley, G Saunders, K Yacef, (2003) Workflow, user modeling and tutoring strategies for just-in-time document delivery, Alem L and A McLean, (eds) Proceedings of Technologies for Electronic Documents for Supporting Learning, Volume X of the AIED2003 Supplementary Proceedings , 664-673.

 

Davis, J, J Kay, K Lin, J Poon, A Quigley, K Yacef, (2002) Workflow Based Just-in-time Training, Kay, J and J Thom, (eds) Proceedings ADCS2002, Australian Document Computing Symposium, 16 December 2002, 145-148. online proceedings

 

Kay, J and K Yacef, (2001) Personalisation in a just-in-time organisational training system, Proceedings of ADCS'2001, Australian Document Computing Symposium, 83-86. online proceedings


Acknowledgement: This work is currently funded by a grant from SITCRC, Smart Internet Technology Cooperative Research Centre.
Context-aware JITT
Bob Kummerfeld, Joseph Davis, Judy Kay, Josiah Poon, Aaron Quigley, Greg Saunders and Kalina Yacef.
This is the next stage of the work. The projects at the right all map into this future direction for the work.
Current projects available:
An important next step in this work involves integration of semantic indexing of documents. This would mean that even when the user asks for information using terms that were not envisaged by the author of the document, the system will be able to find them. There are many ways to tackle this problem. This project should link with the existing work on ontologies, especially Mecurio.
A specialised direction for this work is towards a meeting assistant. This work links with the new `smart meeting room', which will be rich in technology to support meetings. The goal of this project is to assist the newcomer during a meeting: as the participants proceed through the agenda items, the JITT software tracks the issues relevant to each agenda item and presents each person with a personalised collection of organisational information relevant to the currrent agenda item and their own knowledge. This project offers many interesting user interface issues to explore.
The student JITT. This project shifts from employees to students and their need to learn about the administrative rules relevant to them. Like new employees in a large organisation, students cannot learn all the relevant administrative rules when they first enter the University. This is true for all three major groups of students: undergraduates, coursework postgraduates and research postgraduates. This project explores adaptation of the ideas described in the previous JITT papers, but now applied to the multitude of things that students should know. This project offers opportunities to also explore a range of modalities that are favoured by this population.
Scrutable sterotypic models of organisational knowledge. A stereotype allows a very limited amount of information about a person to serve as a basis for inferences about a large user model for them. The good thing about stereotypes is that the user does not have to give the system much information. The bad thing is that stereotypes, even if statistically valid, are broad generalisations about people and parts are invariably wrong for any one person. The goal of this project is to build models of the organisational knowledge in an organisation and to construct a collection of stereotypes for this. Then the project involves using VlUM and other visualisation tools to enable people to check out theses stereotypic user models. For an honours project, the models should be modest in size, with around 100 concepts. For higher degree students, they should involve larger models and this would then involve some quite challenging and interesting issues in structuring the user model.
Note the long list of people involved in the JITT work. Many of them will be able to supervise JITT projects in their own areas of expertise.
Top of Page