Classifying cultural heritage images by using decision tree classifiers in WEKA

Abstract

This paper presents the first step toward looking for an advanced solution of image classification using decision trees in the Weka software. The aim of the paper is to evaluate the ability of different decision tree classifiers for cultural heritage image classification involving a small sample, based on three types of extracted image features: (1) Fuzzy and texture histogram, (2) edge histogram, and (3) DCT coefficients. The used decision tree algorithms involve J48, Hoeffding Tree, Random Tree, and Random Forest. The results indicate that the Random Forest algorithm performs best in classifying a small sample of cultural heritage images, while the Random Tree performs worst with the lowest classification accuracy.