Professional Documents
Culture Documents
Amit Basu - Data Mining III
Amit Basu - Data Mining III
Pivoting
diapers beer
Rule form:
When lift > 1 then the rule is better at predicting the result
than guessing
,
_
,
_
d d
d
k
k d
j
j
k d
k
d
R
If d=6, R = 602 rules
The Problem of Lots of Data
Itemsets that qualify are called large itemsets, and all others
small itemsets.
Apriori principle:
Progressively
identifies large
itemsets of different
sizes
Dissociation rules
Shopper characteristics
Store characteristics
Seasonal factors
Text Mining
Spatial data
GIS
Temporal data
Time series
Behavioral patterns
Web Mining
Web usage
Web content
Mining Image Data
Neural networks
Supervised learning
Discovering patterns
Unsupervised learning
Clustering
Mining Spatial Data
Distance-based clustering
Feature extraction
Association rules
Search engines
Metacrawlers
Dynamic personalization
Issues and Trends
Property
Accuracy