Spatial Association Rules
- All the methods discussed previously only mine characteristic rules. We sometimes need to find relationships (spatial and non-spatial) between attributes that belong to different classes.
A spatial association rule is of the form X -->Y (c%), where X and Y are sets of spatial or nonspatial predicates and c% is the confidence of the rule.
- is_a(X,City) and has(X, beach)-->close_to(X, sea) (Spatial RHS)
- Is_a(X, student) --> goes_to(X, school) -- not a spatial association rule.
Concepts of minimum support and confidence are used to generate rules. Different values at each level .