A CBR approach to Knowledge Management

Deepak Khemani
IIT Madras


Abstract

The sharing of experience has always contributed to the collective knowledge of societies. Traditionally experience has been shared in the form of fables and stories passed on from person to person. In well defined domains the role of apprenticeship has been instrumental in knowledge sharing. In the information age computers have started playing a role as facilitators in knowledge sharing. We look at an approach to knowledge management (KM) based on case based reasoning (CBR). Case based reasoning is a methodology that enables the storage and reuse of experiences. Cases can accrue through the decisions of expert personnel, and the case base forms a collective memory accessible to the organization. We look at an application that has been designed to deliver faster and cheaper analysis of soil properties for farmers. It has been hypothesized that a circular paper chromatogram made from a soil sample contains important information of the mineral content in the soil. Since chromatogram preparation is cheaper than chemical analysis the goal is to predict the nutrients from the chromatogram image features in future rather than direct chemical analysis. We build a case base of chromatogram images along with chemically obtained soil properties, and compute the soil properties of a new sample by retrieving the most similar chromatograms. We also briefly touch upon a similar application from the manufacturing industry.