خط مشی دسترسیدرباره ما
ثبت نامثبت نام
راهنماراهنما
فارسی
ورودورود
صفحه اصلیصفحه اصلی
جستجوی مدارک
تمام متن
منابع دیجیتالی
رکورد قبلیرکورد بعدی
Document Type : Latin Dissertation
Language of Document : English
Record Number : 151372
Doc. No : ET23164
Main Entry : Ahmed Esmat Mahmoud Ibrahim
Title Proper : Learning Continuous Functions Using Decision trees Learning Algorithm
Note : This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract : C4.5 as an implementation of a decision tree learning algorithm called ID3. C4.5accounts for discrete classes and does not consider continuous output. The purpose ofthis work is to suggest two approaches to modify the C4.5 implementation to accountfor continuous output. These two approaches are called: Multi-Class C4.5 and Multi-Binary trees C4.5. Multi-Class C4.5 approach groups continuous output values intochunks and averages them. A discrete class replaces the class of the examples in eachgroup. The average of the continuous classes in each group is associated to the discreteclass of the group. Multi-Binary trees C4.5 approach divides the examples into chunksand builds as many decision trees as the number of chunks. The end boundary of eachchunk becomes the class of the chunk. Each tree is built to indicate that the class ofexamples is less than or equal to the end boundary. Based on experiments conductedon six well-known domains, these two....-....,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject : Electericl tess
: برق
electronic file name : TL46400.pdf
Title and statement of responsibility and : Learning Continuous Functions Using Decision trees Learning Algorithm [Thesis]
 
 
 
(در صورت عدم وضوح تصویر اینجا را کلیک نمایید)