Joining Forces

The ICCBR-workshops "Human Centered Case-Based Reasoning" and "Cognitive Approaches in Memory-Based Reasoning" are joining forces: there is considerable overlap in topics we want to address, and we decided to have a single workshop under the banner "Human-Centered and Cognitive approaches to Case-Based Reasoning".

Workshop Focus

The problem-solving landscape has evolved with the widespread availability of data on the web, the large scale network connectivity through the internet and the increasing processing power available from modern computers. At the same time, intelligent systems are not longer considered to be black boxes that provide a full solution to a problem on their own; instead, problem solving is seen as a more interactive process.

Domains no longer have stable knowledge sources being applied to fixed problems but tend to be typified by flexible, evolving tasks being solved with rapidly changing, distributed knowledge sources. In humans, success at problem-solving lies in their ability to integrate experience-based reasoning with other problem-solving strategies. Likewise, can CBR integrate new strategies into the problem-solving process?

The traditional split of an experience into a problem and solution component may no longer apply. Humans interacting with a system may not be looking for a single best solution; they might also want a rich navigation experience as they learn more about the problem topic, and refine their world model to reflect this. Do evolving problem-solving systems require new flexible representations to support their reasoning?

Learning in CBR often has the narrow focus of retaining new experiences for future retrieval. In humans, learning has a broader scope that encompasses curiosity, an understanding of our competence, strategies to acquire new knowledge, and the means of embedding acquired knowledge. Has this proactive view of learning been given sufficient attention by the CBR community?

It is also of interest to examine how human knowledge modelling and construction can best be supported through technology. Looking at the human side of the interactive system, we have to consider issues of communication and, in particular, explanations. Problem descriptions, solutions, or input and output in general, can be incomplete and changing. As a consequence, there has to be communication between human and software agents. Communication requires mutual understanding that can be essentially supported by explanations.

Many of these issues are already being addressed in other disciplines, however, the question remains as to how findings from human-computer interaction, the social sciences and psychology may be integrated with the design of CBR systems. This integration ranges from psychologically plausible knowledge models over a focus on usability of fielded systems to the delivery of an attractive end user experience.

We would like this forum to encourage CBR researchers to take a fresh look at their cognitive roots and what they have been doing. This workshop also aims to bring together researchers and practitioners exploring human aspects of the design, implementation, and use of CBR-systems, from a broad range of areas, to share their problems and methodologies across different research and application areas. The workshop will examine methods, mechanisms, and techniques to keep the human in the centre of attention during the whole life-cycle of an intelligent system, from initial problem description through to knowledge acquisition and modelling and on to interactive use and maintenance.

Workshop Format

The workshop will last half a day and will be organised into two main parts.

The first part will consist of lightning presentations of the accepted papers. The goal of this sessions is to introduce the work of all participants.

The second part will consist of two panel discussions, each dedicated to one specific issue. The suggested issues are "Human Centred CBR" and "Cognitive Approaches to Memory-Based Reasoning", but these are subject to change dependent on the interests of the attendees and the nature of submissions. The goal of these panels is to discuss the various approaches to each of these basic issues, to identify the critical problems in need of attention and to identify the most promising future research directions.

The panels will be comprised of authors of accepted position papers. Each panelist is given the task to describe his take on the issue in form of a short, 5 minute kick-off talk. The panel will then engage in open discussion with the floor.

The workshop will be concluded with an open discussion summarising the most challenging issues and the most important lessons learned.

Last modified: Thursday, 2011-09-08 11:09 UTC.