ANALISIS RANDOM FOREST PADA KLASIFIKASI CART KETIDAKTEPATAN WAKTU KELULUSAN MAHASISWA UNIVERSITAS TERBUKA
ANALISIS RANDOM FOREST PADA KLASIFIKASI CART KETIDAKTEPATAN WAKTU KELULUSAN MAHASISWA UNIVERSITAS TERBUKA
复制标题
分析 随机森林 PADA KLASIFIKASI CART KETIDAKTEPATAN WAKTU KELULUSAN MAHASISWA UNIVERSITAS TERBUKA
DOI：
10.30598/barekengvol13iss3pp177-184ar910
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Putu Suniantara
中科院分区：
文献类型：
--
作者：
Jurnal Ilmu Matematika;D. Terapan;Universitas Terbuka;Gede Suwardika;I. Ketut;Putu Suniantara
分类回归树(CART)是广泛应用于各个领域的分类方法之一。该方法被认为能够处理各种数据条件。然而，CART方法在分类树预测方面存在弱点，对学习数据的变化不太稳定，这会导致分类树预测结果的重大变化。为了改进CART分类树的预测能力，提出了一种集成随机森林方法，该方法结合多棵分类树来提高稳定性和确定分类预测。本研究旨在利用随机森林来提高CART预测的稳定性和准确性。本研究使用的案例是开放大学学生毕业过程中的不准确分类。分析结果表明，随机森林能够提高学生毕业不准确分类的准确率，达到收敛，分类预测达到93.23%。
Classification and Regression Tree (CART) is one of the classification methods that are popularly used in various fields. The method is considered capable of dealing with various data conditions. However, the CART method has weaknesses in the classification tree prediction, which is less stable in changes in learning data which will cause major changes in the results of the classification tree prediction. Improving the predictions of the CART classification tree, an ensemble random forest method was developed that combines many classification trees to improve stability and determine classification predictions. This study aims to improve CART predictive stability and accuracy with Random Forest. The case used in this study is the classification of inaccuracies in Open University student graduation. The results of the analysis show that random forest is able to increase the accuracy of the classification of the inaccuracy of student graduation that reaches convergence with the prediction of classification reaching 93.23%.