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 locally patched. That is, new rules are exceptions to previous rules and the new rule is validated within the context of previously seen cases. One drawback of local patching is that knowledge can be repeated in different locations of the knowledge base. This paper describes some work done on removing repeated knowledge. The experiments reported were performed on a pathology knowledge base but the algorithm is applicable to any multiple classification RDR knowledge based system. The results support the findings of others that exception structures are compact representations with few opportunities to reduce further. This also suggests that experts tend to provide overly general rules in the first instance which they modify by adding specialisations in the form of exception rules as new cases are seen.
[1]
Stephen Muggleton,et al.
Duce, An Oracle-based Approach to Constructive Induction
,
1987,
IJCAI.
[2]
Debbie Richards,et al.
The Reuse of Ripple Down Rule Knowledge Bases : Using Machine Learning to Remove Repetition
,
1996
.
[3]
Heikki Mannila,et al.
Learning rules with local exceptions
,
1994
.
[4]
G Edwards,et al.
Peirs: A pathologist‐maintained expert system for the interpretation of chemical pathology reports
,
1993,
Pathology.
[5]
Paul Compton,et al.
Local Patching Produces Compact Knowledge Bases
,
1994,
EKAW.
[6]
Rani Siromoney,et al.
Variations and local exceptions in inductive logic programming
,
1993,
Machine Intelligence 14.