Mining negative rules using GRD

1
Mining negative rules using GRD

• D. R. Thiruvady and G. I. Webb
• PAKDD 2004

2
Outline

• Introduction
• OPUS
• GRD
• Conclusion

3
Introduction

• Association rule
• A gt B (A is antecedent ,B is consequent)
• Negative Rules
• Either antecedent or consequent or both are
  negated

4
Optimized Pruning for Unordered Search (OPUS)
5
Generate Rule Discovery (GRD)

• Extends OPUS by remove the requirement that
  consequent be single variable
• K number of rules replace minimum support

6
GRD

• Four measures with respect to a rule X?Y

7
GRD

• Symbol

8
Properties
9
Properties
10
Properties
11
Properties
12
GRD

• Symbol
• CurrentLHSInit ?
• AvailableLHSInit Antecedent
• AvailableRHSInit Consequent
• Function
• Insolution(a?c)rule a?c in solution
• Proven(X)pruning rules provided to the algorithm
  prove the proposition X

13
GRD algorithm
Prune 1
Prune 2
Prune 3,4,5
14
GRD algorithm2
Prune 6
15
Pruning
16
Pruning
17
Pruning
18
Update Constraints
19
Negative with GRD
20
Conclusion

• Disadvantage
• Search space too large