A comparative study of algorithms constructing decision trees
Halima Elaidi, Zahra Benabbou, Hassan Abbar · 2018
Decision tree is a very popular and efficient supervised technique of machine learning. It is used and developed in order to detect meaningful rules of classification from data. Decision trees employ the top-down-divide-and-conquer strategy to get smaller and more homogeneous subsets as the construction of the tree progresses. ID3 and C4.5 are two classical algorithms of the decision trees theory; this paper discusses in detail these algorithms and presents a comparative study of them by giving the advantages and disadvantages of each one of them.