Evolving Polynomials of the Inputs for Decision Tree Building
Chris J. Hinde1, Anoud I. Bani-Hani1,
Thomas.W. Jackson2, and
Yen P. Cheung3
1. Loughborough University/Computer Science, Loughborough, UK
2. Loughborough University/Information Science, Loughborough, UK
3. Monash University/Clayton School of IT, Wellington, Australia
2. Loughborough University/Information Science, Loughborough, UK
3. Monash University/Clayton School of IT, Wellington, Australia
Abstract—The aim of this research is to extend the discrimination of a decision tree builder by adding polynomials of the base inputs to the inputs. The polynomials used to extend the inputs are evolved using the quality of the decision trees resulting from the extended inputs as a fitness function. Our approach generates a decision tree using the base inputs and compares it with a decision tree built using the extended input space. Results show substantial improvements. Rough set reducts are also employed and show no reduction in discrimination through the transformed space.
Index Terms—decision tree building, polynomials
Cite: Chris J. Hinde, Anoud I. Bani-Hani, Thomas.W. Jackson, and Yen P. Cheung, "Evolving Polynomials of the Inputs for Decision Tree Building," Journal of Emerging Technologies in Web Intelligence, Vol. 4, No. 2, pp. 198-203, May 2012. doi:10.4304/jetwi.4.2.198-203
Index Terms—decision tree building, polynomials
Cite: Chris J. Hinde, Anoud I. Bani-Hani, Thomas.W. Jackson, and Yen P. Cheung, "Evolving Polynomials of the Inputs for Decision Tree Building," Journal of Emerging Technologies in Web Intelligence, Vol. 4, No. 2, pp. 198-203, May 2012. doi:10.4304/jetwi.4.2.198-203
Array