AnalisisKinerja Algoritma C 4.5 dengan Pruning Menggunakan Decision Tree
Total View This Week0
Total View This Week0
Institusion
Universitas Sumatera Utara
Author
Sellyana, Ari (STUDENT ID : 137038024)
(LECTURER ID : 0001075703)
(LECTURER ID : 0026106209)
(LECTURER ID : 0001075703)
(LECTURER ID : 0026106209)
Subject
Classification
Datestamp
2022-12-14 08:49:30
Abstract :
Classification is a method that works to find a model that distinguishes data classes, so classification can predict the class label of an object that is not yet known. One method often used is the classification algorithm C 4.5, where the classification process carried out by the establishment of the tree (decision tree) based on the training data that has been provided. After the formation of the tree, the tree will be simplified by way of pruning trees (pruning), was conducted to identify and eliminate some of the branches of the tree that is aimed at getting more efficient models and reduce errors in classification, which is due to specialization in the training data. This study analyzes the degree of accuracy without pruning algorithm C 4.5 and C 4.5 with pruning algorithm that in a decision tree to classify data. The results of this study states the formation of decision tree algorithm C 4.5 without pruning has a speed of 1,823 seconds and algorithm C 4.5 with pruning 0599 seconds. So it was found that the formation of decision tree algorithm C 4.5 with pruning has- a higher speed than without pruning the difference between 1,224 seconds. As for the results of the level of accuracy of the data, without pruning process is better than the use of pruning, 95.54% to C 4.5 without pruning and 91.96% for C 4.5 with pruning, with the difference between 3:58%
Classification is a method that works to find a model that distinguishes data classes, so classification can predict the class label of an object that is not yet known. One method often used is the classification algorithm C 4.5, where the classification process carried out by the establishment of the tree (decision tree) based on the training data that has been provided. After the formation of the tree, the tree will be simplified by way of pruning trees (pruning), was conducted to identify and eliminate some of the branches of the tree that is aimed at getting more efficient models and reduce errors in classification, which is due to specialization in the training data. This study analyzes the degree of accuracy without pruning algorithm C 4.5 and C 4.5 with pruning algorithm that in a decision tree to classify data. The results of this study states the formation of decision tree algorithm C 4.5 without pruning has a speed of 1,823 seconds and algorithm C 4.5 with pruning 0599 seconds. So it was found that the formation of decision tree algorithm C 4.5 with pruning has- a higher speed than without pruning the difference between 1,224 seconds. As for the results of the level of accuracy of the data, without pruning process is better than the use of pruning, 95.54% to C 4.5 without pruning and 91.96% for C 4.5 with pruning, with the difference between 3:58%
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
