Comparison of Chi-square based algorithms for discretization of continuous chicken egg quality traits

Cebeci, Zeynel and Yildiz, Figen (2017) Comparison of Chi-square based algorithms for discretization of continuous chicken egg quality traits. AGRÁRINFORMATIKA / JOURNAL OF AGRICULTURAL INFORMATICS, 8 (1). pp. 13-22. ISSN 2061-862X

Preview

Text document (8).pdf - Published Version Available under License Creative Commons Attribution. Download (889kB) | Preview

Abstract

Discretization isa data pre-processing task fortransforming continuous variables into discrete ones. In this study,four Chi-square based supervised discretization algorithms (ChiMerge, Chi2, Extended Chi2 and Modified Chi2) were compared fordiscretization ofthe fourteencontinuous variables in a chicken egg quality traits dataset. We found that all of the algorithms had similar performances in term of training model accuracies obtained with C5.0 classification tree algorithmwhereasChiMerge and Chi2 were better than the remainingalgorithms in term of training error rates.The numbersof intervals obtained withChi2 tended to belargewhile they were very smallin Extended Chi2 and Modified Chi2. The numbersof intervals from ChiMerge increased as the significance level increases whereas they werethesame at all the levels of significance for the remainingalgorithms. Consequently, it was revealed that ChiMerge at the significance levelsof 0.05 and 0.10 was more efficientthan the others and could be a better choice in discretization of the egg quality traits.

Actions (login required)

Edit Item