This is a summary of a set of 1 experiments using a LONI pipeline workflow file that performs 3000 independent jobs, each one with the CBDA-SL and the knockoff filter feature mining strategies. Each experiments has a total of 9000 jobs and is uniquely identified by 6 input arguments: # of jobs [M], % of missing values [misValperc], min [Kcol_min] and max [Kcol_max] % for FSR-Feature Sampling Range, min [Nrow_min] and max [Nrow_max] % for SSR-Subject Sampling Range.
This document has the final results, by experiment. See https://drive.google.com/file/d/0B5sz_T_1CNJQWmlsRTZEcjBEOEk/view?ths=true for some general documentation of the CBDA-SL project and github https://github.com/SOCR/CBDA for some of the code.
Features selected by both the knockoff filter and the CBDA-SL algorithms are shown as spikes in the histograms shown below. I list the top features selected, set to 15 here.
## [1] "C:/Users/simeonem/Documents/CBDA-SL/ExperimentsNov2016/NULL9000/NEW"
## [1] EXPERIMENT 1
## M misValperc Kcol_min Kcol_max Nrow_min Nrow_max
## 9000 0 5 15 60 80
## [1] 1 100 200 400 600 800 1000 1200 1400 1500
## [1] "TABLE with CBDA-SL & KNOCKOFF FILTER RESULTS"
## CBDA Frequency Density
## 393 13 0.1838755
## 49 11 0.1555870
## 607 11 0.1555870
## 840 11 0.1555870
## 913 11 0.1555870
## 928 11 0.1555870
## 1104 11 0.1555870
## 1145 11 0.1555870
## 1307 11 0.1555870
## 1491 11 0.1555870
## 125 10 0.1414427
## 174 10 0.1414427
## 203 10 0.1414427
## 260 10 0.1414427
## 349 10 0.1414427
## [1] "EXPERIMENT" "1"
## [1] "Nonzero Features"
## [1] 1 100 200 400 600 800 1000 1200 1400 1500
##
##
##
##
##
##
## [1] EXPERIMENT 2
## M misValperc Kcol_min Kcol_max Nrow_min Nrow_max
## 9000 0 15 30 60 80
## [1] 1 100 200 400 600 800 1000 1200 1400 1500
## [1] "TABLE with CBDA-SL & KNOCKOFF FILTER RESULTS"
## CBDA Frequency Density
## 519 23 0.1387465
## 453 21 0.1266815
## 549 21 0.1266815
## 986 21 0.1266815
## 1257 21 0.1266815
## 652 20 0.1206491
## 10 19 0.1146166
## 174 19 0.1146166
## 867 19 0.1146166
## 970 19 0.1146166
## 1169 19 0.1146166
## 1422 19 0.1146166
## 168 18 0.1085842
## 176 18 0.1085842
## 319 18 0.1085842
## [1] "EXPERIMENT" "2"
## [1] "Nonzero Features"
## [1] 1 100 200 400 600 800 1000 1200 1400 1500
The features listed above are then used to run a final analysis applying both the CBDA-SL and the knockoff filter. The ONLY features used for analysis are the ones listed above. A final summary of the accuracy of the overall procedure is determined by using the CDBA-SL object on the subset of subjects held off for prediction. The predictions are then used to generate the confusion matrix. We basically combine the CBDA-SL & Knockoff Filter algorithms to first select the top features during the first round. Then, the second stage uses the top features to run a final predictive modeling step that can ultimately be tested for accuracy, sensitivity,…..