Efficient Evaluation of Queries with Mining Predicates by Chaudhuri, Narasayya, and Sarawagi

1
Efficient Evaluation of Queries with Mining
Predicatesby Chaudhuri, Narasayya, and Sarawagi

• CSci 8701 Group G07
• Charles Braxmeier

2
Problem Statement

• Find more efficient ways to execute queries where
  one or more of the predicates are the results of
  data mining decisions
• Example Query Find fans who went to a Minnesota
  hockey game last year who may be football fans as
  well

3
Contributions of the Paper

• Great detail about different types of mining
  models (clustering, decision trees, etc.)
• Discussion regarding the different ways mining
  predicate(s) can be joined within a query
• Analysis on the experiments done to test theories
  regarding query optimization based on the
  structure of mining model

4
Key Concepts

• Upper Envelope Predicate
• Tightness of the Querys Predicates
• Mining Model
• Decision Tree
• Naïve Bayes Classifiers
• Bottom-up
• Top-Down

5
Key Concepts (contd.)

• Mining Model (continued)
• Clustering
• Centroid-based
• Model-based
• Boundary-based

6
Validation Methodology

• Experimentation based on the theories posed
  regarding query reorganization
• Twenty (20) different data sets used. Data sets
  vary based on
• Data set size
• Number of dimensions in data set
• Size of data set used to train the mining model

7
Validation Methodology (contd.)

• Analysis of Experiment Results
• 65 of query access paths affected by
  re-arranging the query based on the upper
  envelope predicate
• Average run-time decreased by 65 by re-arranging
  the query based on the upper envelope predicate
• More variance in run-time decrease than access
  paths affected

8
Assumptions

• Clustering can be evaluated via Bayes classifiers
• Therefore, not too much background info on
  clustering and how its experiments were different
  than Bayes experiments
• Continuous data sets are split into discrete data
  sets to assist in mining predictions
• Not necessarily realistic
• Example, latitude / longitude

9
Possible Revisions to Paper

• Spend more time on analysis of experiments and
  results, rather than the background info
• Background information took up approximately 60
  of the paper

10
Questions?