U.S. Department of EnergyOffice of Scientific and Technical Information
Separating Signal From Background Using Ensembles of Rules
Abstract
Machine learning has emerged as a important tool for separating signal events from associated background in high energy particle physics experiments. This paper describes a new machine learning method based on ensembles of rules. Each rule consists of a conjuction of a small number of simple statements (''cuts'') concerning the values of individual input variables. These rule ensembles produce predictive accuracy comparable to the best methods. However their principal advantage lies in interpretation. Because of its simple form, each rule is easy to understand, as is its influence on the predictive model. Similarly, the degree of relevance of each of the respective input variables can be assessed. Graphical representations are presented that can be used to ascertain the dependence of the model jointly on the variables used for prediction.
- Authors:
- Publication Date:
- Research Org.:
- SLAC National Accelerator Lab., Menlo Park, CA (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 895817
- Report Number(s):
- SLAC-PUB-12247
TRN: US200703%%521
- DOE Contract Number:
- AC02-76SF00515
- Resource Type:
- Conference
- Resource Relation:
- Conference: Prepared for PHYSTATO5: Statistical Problems in Particle Physics, Astrophysics and Cosmology, Oxford, England, United Kingdom, 12-15 Sep 2005
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSICS; ACCURACY; ASTROPHYSICS; COSMOLOGY; FORECASTING; LEARNING; PHYSICS; Instrumentation,INST
Citation Formats
Friedman, J H, and /SLAC /Stanford U., Phys. Dept. Separating Signal From Background Using Ensembles of Rules. United States: N. p., 2006.
Web.
Friedman, J H, & /SLAC /Stanford U., Phys. Dept. Separating Signal From Background Using Ensembles of Rules. United States.
Friedman, J H, and /SLAC /Stanford U., Phys. Dept. 2006.
"Separating Signal From Background Using Ensembles of Rules". United States. https://www.osti.gov/servlets/purl/895817.
@article{osti_895817,
title = {Separating Signal From Background Using Ensembles of Rules},
author = {Friedman, J H and /SLAC /Stanford U., Phys. Dept.},
abstractNote = {Machine learning has emerged as a important tool for separating signal events from associated background in high energy particle physics experiments. This paper describes a new machine learning method based on ensembles of rules. Each rule consists of a conjuction of a small number of simple statements (''cuts'') concerning the values of individual input variables. These rule ensembles produce predictive accuracy comparable to the best methods. However their principal advantage lies in interpretation. Because of its simple form, each rule is easy to understand, as is its influence on the predictive model. Similarly, the degree of relevance of each of the respective input variables can be assessed. Graphical representations are presented that can be used to ascertain the dependence of the model jointly on the variables used for prediction.},
doi = {},
url = {https://www.osti.gov/biblio/895817},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Fri Dec 01 00:00:00 EST 2006},
month = {Fri Dec 01 00:00:00 EST 2006}
}