Call for Papers

Special Issue in Knowledge-Based Systems

on

Enhancing experience reuse and lesson learning

Knowledge-based systems (KBS) intend to provide end-users with knowledge access and reasoning in specific subject domains. One of the main objectives of KBS is to share expert knowledge and to enhance knowledge reuse and learning. From a technical point of view, such systems generally incorporate, at least, a knowledge base, an inference engine and a user interface.

However, the elaboration of knowledge models issued from expert interviews and/or from document repositories is known to be a difficult and expensive task. For the last few years, alternative approaches for knowledge-based systems elaboration have been explored to overcome this difficulty. They focus on the representation and reuse of experiences, which can be defined as knowledge in action, knowledge used during the achievement of a particular task. These approaches favor contextualized, shared and reusable, lessons-learned oriented knowledge engineering.

Contextualization: experiences closely relate to actual activities. Contextual information (a context consists of a set of facts and a description of an environment within which those facts are believed to be true) can easily be modeled and linked to relevant knowledge that may be either used or produced by activities;

Sharing: an important issue of experience-based systems is to facilitate the sharing of experiences over time (from one project to another for instance) and over space (from an organization to another for instance). Since actors contribute to activities with their viewpoints and with specialized skills and knowledge it is important to share experiences and to promote collaborative reuse;

Reuse: the reuse of experiences is a key inference mechanism for practical experience-based systems. These mechanisms should enable to find appropriate experiences that have been already capitalized and to integrate them in current activities and roles;

Lessons-learning: fragments of knowledge collected over time and space should be analyzed to produce more general knowledge such as best practices and lessons-learned.

Experience feedback processes are based on the reuse and generalization of knowledge monitored during the activities as it is illustrated in the following figure.

From a practical point of view, effective experience reuse and lessons-learned are increasingly important assets of enterprises and represent sources of competitive advantages in various domains such as systems design and engineering, quality management, maintenance, dependability engineering, risk management, diagnosis, planning and project management …

From a technical point of view, different underlying models and techniques have been elaborated and used to improve knowledge contextualization and facilitate its reuse such as case-based and trace-based reasoning, experience feedback and lessons-learned systems.

This special issue aims at:

• focusing on the formalisms and tools that are relevant to model experiences and facilitate their reuse,
• providing an overview on theoretical and empirical research related to experience-based systems and lessons-learned systems,
• gathering research works from multiple disciplines (information system, artificial intelligence, industrial engineering, medical services or else…) with applications and to compare worldwide approaches and practices.

However, this special issue will not be focus on research on processing of statistical data (monitored process data, ), data mining or data warehousing since these approaches are not directly based on the formalization of experience produced by human activity.

Call for paper diffusion: 15 October 2012

Submission deadline: 15 February 2013

First notification: 15 May 2013

Re-submission: 15 September 2013

Second notification and final acceptance: 15 December 2013

Camera-ready: January 2014

• Contributions are invited in the form of original high-quality research and review papers (preferably no more than 20 double line spaced manuscript pages, including tables and figures), following the formatting style for Elsevier.
  
  
• A submission that has already been published in conference proceedings has to be submitted as more than 35% update in comparison to the published version.
  
  
• The title page should not include name, affiliation, and e-mail address of the authors.
  
  
• All paper has to be submitted through the journal electronic submission EES, ees.elsevier.com/knosys

Full recommendations are available in KBS or Elsevier websites:

http://www.journals.elsevier.com/knowledge-based-systems/http://www.elsevier.com/wps/find/journaldescription.cws_home/525448/authorinstructions

http://www.elsevier.com/wps/find/journaldescription.cws_home/525448/authorinstructions

Prof. Eric Bonjour, Université de Lorraine, France, eric.bonjouruniv-lorrainefr

Prof. Laurent Geneste, ENIT, France, laurent.genesteenitfr

Prof. Ralph Bergmann, University of Trier, Germany, bergmannuni-trierde