Optimization of Signal Significance by Bagging Decision Trees
- Creators
- Narsky, I.
- Others:
- Lyons, Louis
- Karagöz Ünel, Müge
Abstract
An algorithm for optimization of signal significance or any other classification figure of merit (FOM) suited for analysis of HEP data is described. This algorithm trains decision trees on many bootstrap replicas of training data with each tree required to optimize the signal significance or any other chosen FOM. New data are then classified by a simple majority vote of the built trees. The performance of the algorithm has been studied using a search for the radiative leptonic decay B → γlν at BABAR and shown to be superior to that of all other attempted classifiers including such powerful methods as boosted decision trees. In the B → γeν channel, the described algorithm increases the expected signal significance from 2.4σ obtained by an original method designed for the B → γlν analysis to 3.0σ.
Additional Information
© 2006 Imperial College Press. Work partially supported by Department of Energy under Grant DE-FG03-92-ER40701. Thanks to Frank Porter for comments on a draft of this note.Attached Files
Accepted Version - 0507157.pdf
Files
Name | Size | Download all |
---|---|---|
md5:4ee9e38316ecc664798ef7cd42c68d6c
|
177.8 kB | Preview Download |
Additional details
- Eprint ID
- 98926
- Resolver ID
- CaltechAUTHORS:20190930-110457242
- DE-FG03-92-ER40701
- Department of Energy (DOE)
- Created
-
2019-10-07Created from EPrint's datestamp field
- Updated
-
2021-11-16Created from EPrint's last_modified field