Data Mining Using Unguided Symbolic Regression on a Blast Furnace Dataset

Created by W.Langdon from gp-bibliography.bib Revision:1.4549

  author =       "Michael Kommenda and Gabriel Kronberger and 
                 Christoph Feilmayr and Michael Affenzeller",
  title =        "Data Mining Using Unguided Symbolic Regression on a
                 Blast Furnace Dataset",
  booktitle =    "Applications of Evolutionary Computing,
                 EvoApplications 2011: {EvoCOMPLEX}, {EvoGAMES},
                 {EvoIASP}, {EvoINTELLIGENCE}, {EvoNUM}, {EvoSTOC}",
  year =         "2011",
  month =        "27-29 " # apr,
  editor =       "Cecilia {Di Chio} and Stefano Cagnoni and 
                 Carlos Cotta and Marc Ebner and Aniko Ekart and 
                 Anna I Esparcia-Alcazar and Juan J. Merelo and 
                 Ferrante Neri and Mike Preuss and Hendrik Richter and 
                 Julian Togelius and Georgios N. Yannakakis",
  series =       "LNCS",
  volume =       "6624",
  publisher =    "Springer Verlag",
  address =      "Turin, Italy",
  publisher_address = "Berlin",
  pages =        "274--283",
  organisation = "EvoStar",
  keywords =     "genetic algorithms, genetic programming, Variable
                 Selection, Data Mining, Blast Furnace, steel",
  isbn13 =       "978-3-642-20524-8",
  DOI =          "doi:10.1007/978-3-642-20525-5_28",
  abstract =     "In this paper a data mining approach for variable
                 selection and knowledge extraction from datasets is
                 presented. The approach is based on unguided symbolic
                 regression (every variable present in the dataset is
                 treated as the target variable in multiple regression
                 runs) and a novel variable relevance metric for genetic
                 programming. The relevance of each input variable is
                 calculated and a model approximating the target
                 variable is created. The genetic programming
                 configurations with different target variables are
                 executed multiple times to reduce stochastic effects
                 and the aggregated results are displayed as a variable
                 interaction network. This interaction network
                 highlights important system components and implicit
                 relations between the variables. The whole approach is
                 tested on a blast furnace dataset, because of the
                 complexity of the blast furnace and the many
                 interrelations between the variables. Finally the
                 achieved results are discussed with respect to existing
                 knowledge about the blast furnace process.",
  notes =        "Part of \cite{DiChio:2011:evo_a} EvoApplications2011
                 held inconjunction with EuroGP'2011, EvoCOP2011 and

Genetic Programming entries for Michael Kommenda Gabriel Kronberger Christoph Feilmayr Michael Affenzeller