15-12-2025
جامعة الملك عبدالعزيز
KING ABDULAZIZ UNIVERSITY
Faculty of Computing and Information Technology in Rabigh
Document Details
Document Type
:
Article In Journal
Document Title
:
Random Projection Random Discretization Ensembles - Ensembles of Linear Multivariate Decision Trees
الإسقاط العشوائي وعشوائية تفريد الفرق- فرق أشجار القرار الخطية متعددة المتغيرات
Subject
:
Computer Science
Document Language
:
English
Abstract
:
: In this paper, we present a novel ensemble method random projection random discretization ensembles(RPRDE) to create ensembles of linear multivariate decision trees by using a univariate decision tree algorithm. The present method combines the better computational complexity of a univariate decision tree algorithm with the better representational power of linear multivariate decision trees. We develop random discretization (RD) method that creates random discretized features from continuous features. Random projection (RP) is used to create new features that are linear combinations of original features. A new dataset is created by augmenting discretized features (created by using RD) with features created by using RP. Each decision tree of a RPRD ensemble is trained on one dataset from the pool of these datasets by using a univariate decision tree algorithm. As these multivariate decision trees (because of features created by RP) have more representational power than univariate decision trees, we expect accurate decision trees in the ensemble. Diverse training datasets ensure diverse decision trees in the ensemble. We study the performance of RPRDE against other popular ensemble techniques using C4.5 tree as the base classifier. RPRDE matches or outperforms other popular ensemble methods. Experiments results also suggest that the proposed method is quite robust to the class noise
ISSN
:
1041-4347
Journal Name
:
IEEE Transactions on Knowledge and Data Engineering (TKDE)
Volume
:
26
Issue Number
:
5
Publishing Year
:
1435 AH
2014 AD
Article Type
:
Article
Added Date
:
Monday, December 8, 2014
Researchers
Researcher Name (Arabic)
Researcher Name (English)
Researcher Type
Dr Grade
Email
امير احمد
Ahmad, Amir
Investigator
Doctorate
amirahmad01@gmail.com
كيفن براون
Brown, Gavin
Researcher
Doctorate
gbrown@cs.man.ac.uk
Files
File Name
Type
Description
37603.pdf
pdf
Back To Researches Page