Journal article
From texts, images, and data to attribute based case representation
01 Jul 2013
Abstract
In this article we study complex case representations in Case-Based Reasoning. To some degree this is a survey paper. But in addition it gives a unified approach to solving the problems connected with representations mentioned in the title in a way that has not been considered so far. The most popular form to represent cases use attribute-based representations. They allow an easy formulation of similarity measures and retrieval functions. However, in practical applications, case problems and solutions are in the first place given in other ways, e.g. by using texts, images, sensor data or speech data. On this level it is hard to apply reasoning and in particular CBR. This is due to the difficulty to determine similarity measures and retrieval functions. In order to overcome this we introduce a general level structure that allows to bridge the gap between bit-oriented low level and the attribute-oriented high level that is accessible to humans as well as CBR systems. The approach is put in the form of a process model.
Metrics
6 File views/ downloads
2 Record Views
Details
- Title
- From texts, images, and data to attribute based case representation
- Creators
- Michael Richter (Author) - University of CalgaryRosina O. Weber (Author) - Drexel University
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- College of Information Science and Technology (1995-2013); Information Science (Informatics)
- Identifiers
- 991014632427404721