Data Mining using Unguided Symbolic Regression on a Blast Furnace Dataset
This is an incremental method for domain experts in industrial processes like blast furnace operation, focusing on data mining for variable interaction analysis.
The paper tackles the problem of variable selection and knowledge extraction from complex datasets by using unguided symbolic regression and a novel variable relevance metric for genetic programming, applied to a blast furnace dataset to identify important system components and implicit relations.
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.