- author = "Gabriel Kronberger",
- title = "Symbolic Regression for Knowledge Discovery -- Bloat, Overfitting, and Variable Interaction Networks",
- publisher = "Trauner Verlag+Buchservice GmbH",
- year = "2011",
- number = "64",
- series = "Johannes Kepler University, Linz, Reihe C",
- keywords = "genetic algorithms, genetic programming",
- isbn13 = "978-3-85499-875-4",
-
URL = "
https://www.amazon.co.uk/Symbolic-Regression-Knowledge-Discovery-Overfitting/dp/3854998759",
-
URL = "https://pure.fh-ooe.at/de/publications/symbolic-regression-for-knowledge-discovery-bloat-overfitting-and",
- broken = "http://www.trauner.at/buchdetail.aspx?artnr=20134641&unterkat=30",
-
abstract = "This work describes an approach for data analysis
based on symbolic regression and genetic programming,
that produces an overall view of the dependencies of
all variables of a system. The identified dependencies
are represented in form of a variable interaction
network.
In the first part of this work, this approach is described in detail. Important issues are the prevention of bloat and overfitting, the simplification of models, and the identification of relevant input variables. In this context, different methods for bloat control are presented and compared. In addition, a novel way to detect and reduce over fitting is presented and analysed.
The second part of this work demonstrates how comprehensive symbolic regression can be applied for analysis of real-world systems. Variable interaction networks for a blast furnace process and an industrial chemical process are presented and discussed. Additionally, the same approach is also applied on an economic data set to identify macro-economic dependencies.",
- notes = "Described in SIGEVOlution 2011 5(4) page 25 doi:10.1145/2078245.2078249 https://evolution.sigevo.org/issues/SIGEVOlution0504.pdf",
- size = "214 pages. See also \cite{Kronberger:thesis}",
