Sample-Wise Enumeration Methods for Mining Microarray Datasets

A Microarray Dataset

Challenge I

Challenge II

Column/Item Enumeration Lattice

Column/Item Enumeration Lattice

General Framework for Column/Item Enumeration

A Multidimensional View

Sample/Row Enumeration Algorihtms

Existing Row/Sample Enumeration Algorithms

Concepts of CARPENTER

Row Enumeration

Pruning Method 1

Pruning method 2

Pruning Method 3: Minimum Support

From CARPENTER to FARMER

Interesting Rule Groups

Pruning by Interestingness Measure

Ordering of Rows: All Class C before ~C

Pruning Method: Minimum Confidence

Pruning method: Minimum chi-square

Finding Lower Bound, MineLB

Implementation

Experimental studies

Example results--Prostate

Example results--Prostate

Naive Classification Approach

Classification results

Summary of Experiments

COBBLER: Combining Column and Row Enumeration

Single Enumeration Tree

Dynamic Enumeration Tree

Dynamic Enumeration Tree

Switching Condition

Switching Condition

Switching Condition

Length and Row ratio

Extension of our work by other groups (with or without citation)

Extension of our work by other groups (with or without citation) II

Extension of our work by other groups (with or without citation) III

Extension of our work by other groups (with or without citation) IV

Extension of our work by other groups (with or without citation) V

Extension of our work by other groups (with or without citation) VI

Extension of our work by other groups (with or without citation) VII

Extension of our work (Conclusion)

Future Work: Generalize Framework for Row Enumeration Algorithms?

Conclusions

"Thank you!!!"