Overview

Case-based Reasoning systems rely on their underlying knowledge that is organized in knowledge containers and required for each phase of the case-based problem-solving process. The knowledge acquisition is a core task during the development of Case-Based Reasoning systems. Once a Case-Based Reasoning system is running, the knowledge within the knowledge containers has to be maintained in order to keep the system up-to-date. The generation and usage of knowledge affects all areas of Case-Based Reasoning: the generation of vocabulary knowledge, the specification of similarity measures or the adaptation of retrieved solutions.

The workshop focuses on methods for supporting the automated generation/acquisition of knowledge for Case-Based Reasoning systems. We are looking for approaches that contribute to any one of the knowledge containers or the phases of the case-based problem-solving process. An important and special case is knowledge about similarity, because cases are selected based on their similarity to a current problem description. Since similarity measures usually contain significant domain knowledge, knowledge generation plays a major role in the development of similarity measures.

Naturally, the knowledge generated is uncertain. Representing and processing this uncertainty (which corresponds to a kind of meta-knowledge, i.e., knowledge about the knowledge) in an adequate way is an important issue that should also be addressed in the workshop. Especially in knowledge areas with complex, approximate, imprecise cases and heterogeneous domains, the domain knowledge is usually uncertain and incomplete. We are interested in examining how reliable the acquired knowledge is, or what confidence we have in it as well as the types of uncertainty that need to be determined and dealt with using appropriate methods and techniques.

The organizers welcome contributions on the topic of:

* Assigning confidence
* Automatic case generation adaptation knowledge generation
* Maintenance knowledge
* Probabilistic reasoning and Bayesian methods,
* Fuzzy sets, possibility theory, evidence theory,
* Rough sets and information theory,
* Machine learning and data mining algorithms
* Case and knowledge representation, acquisition, and modeling,
* Maintenance and management of CBR systems,
* Case indexing and retrieval,
* Similarity assessment and adaptation
* Flexible similarity measures,
* Similarity measures for complex, imprecise and heterogeneous case domains,
* Maintenance of corporate memories,
* Instance-based and case-based learning,
* Confidence,
* CBR applications.

Submissions

We encourage submissions of papers that report on advances in these core areas. In addition to full papers we also encourage submissions presenting more preliminary results and discussing open problems, for example, dealing with insights or important open problems for future research derived from the construction and use of applications. Correspondingly, two types of contributions will be solicited, namely short communications (short talks) and full papers (long talks).

We also encourage authors to submit papers complementing possible submissions to the main ICCBR conference, for example, papers presenting preliminary extensions or explicitly focusing on unsolved problems. In this case, we only ask to inform us about the existence of a related conference submission and its title.

We invite paper submissions including descriptions of works in progress, research contributions, and position statements. Submissions should attempt to address issues relating to knowledge discovery and similarity in case-based reasoning. Workshop papers should be submitted in Springer LNCS format, which is the format required for the final camera ready copy, with a maximum of 10 pages.

Submissions should be made through the workshop conference management system: EasyChair: https://www.easychair.org/login.cgi?conf=iccbr2010.
For further information do not hesitate to contact the workshop organizers.