|
Title:
|
COMPARING PREDICTIONS OF MACHINE SPEEDUPS USING MICRO-ARCHITECTURE INDEPENDENT CHARACTERISTICS |
|
Author(s):
|
C.l. Curotto |
|
ISBN:
|
978-972-8924-88-1 |
|
Editors:
|
Ajith P. Abraham |
|
Year:
|
2009 |
|
Edition:
|
Single |
|
Keywords:
|
Data mining, decision trees, K nearest neighbour, neural networks, benchmarking |
|
Type:
|
Short Paper |
|
First Page:
|
158 |
|
Last Page:
|
162 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
In order to predict machine speedups from a benchmarks dataset, different data mining (DM) models (DMMs) are built
using micro-architecture independent characteristics. The objective is to compare some previously reported results with
those produced by several algorithms including: Microsoft Decision Trees (MSDT); Microsoft Neural Networks
(MSNN); Waikato Environment for Knowledge Analysis (WEKA) M5P and Clus Predictive Clustering System (Clus). |
|
|
|
|
|
|
