Multiple Classification Ripple Down Rules : Evaluation and Possibilities
نویسنده
چکیده
Ripple Down Rules (RDR) is a knowledge acquisition method which constrains the interactions between the expert and a shell to acquire only correct knowledge. Although RDR works well, it is only suitable for the problem of providing a single classification for a set of data. Multiple Classification Ripple Down Rules (MCRDR) is an extension of RDR which allows multiple independent classifications. The approach has been evaluated in simulation studies where the human expert is replaced by a simulated expert. MCRDR may provide a basis for building a general problem solver for a range of problems beyond classification.
منابع مشابه
Extending Ripple-Down Rules
Ripple-Down Rules (RDR) has had considerable success in providing simple incremental knowledge acquisition in classification domains. It has been extended to multiple classification, configuration, search and more recently to resource allocation tasks. Based on the experience of applying RDR to a resource allocation task, this paper proposes a generalisation of RDR to enable it to apply to a wi...
متن کاملEvaluation of the FastFIX Prototype 5Cs CARD System
The 5Cs architecture offers a hybrid Card And Rule-Driven (CARD) system that supports the Collaborative generation and refinement of a relational structure of Cases, ConditionNodes, Classifications, and Conclusions (hence 5Cs). It stretches the Multiple Classification Ripple Down Rules (MCRDR) algorithm and data structure to encompass collaborative classification, classification merging, and cl...
متن کاملThe automatic compression of multiple classification ripple down rule knowledge based systems: preliminary experiments
Ripple Down Rules (RDR) have a longstanding (and successful) history in the field of biomedical engineering. RDR are a knowledge acquisition and representation technique that allow knowledge to be rapidly acquired and maintained by the domain expert. A key feature of RDR, and the reason why maintenance is easily managed, is that rules are never modified or deleted but they are locally patched. ...
متن کاملIntermediate Concept Discovery in Ripple-Down Rule Knowledge Bases
In this paper we investigate how Ripple Down Rules (RDR) knowledge-based systems (KBS) can be reorganized and intermediate concepts discovered. The experiment is based on a real-world multiple-classification RDR knowledge base (KB) with 3710 rules and 2211 cornerstone cases used for interpreting lipid results in chemical pathology. RDR knowledge acquisition can start with a minimal ontology and...
متن کاملAchieving Rapid Knowledge Acquisition in a High- Volume Call Centre
Ripple Down Rules (RDR) has been applied to a number of domains. In this paper we consider a new application area that presents a number of new challenges. Our application is a high-volume call centre that provides a service / help desk function in a complex problem domain. We propose that the combined use of multiple classification ripple-down-rules (MCRDR) together with a web-enabled hyperlin...
متن کامل