|
Title:
|
A NEW APPROACH TO IMPROVE MULTI-DIMENSIONAL STOCK DATA REDUCTION |
|
Author(s):
|
Jian Jiang , Zhe Zhang , Huaiqing Wang , Xiaoyan Liu , Xuhao Luo , Lin Wang |
|
ISBN:
|
978-972-8924-40-9 |
|
Editors:
|
Jörg Roth, Jairo Gutiérrez and Ajith P. Abraham (series editors: Piet Kommers, Pedro Isaías and Nian-Shing Chen) |
|
Year:
|
2007 |
|
Edition:
|
Single |
|
Keywords:
|
Pre-processing, Dimension reduction, Multi-dimensional stock data, Principal component analysis (PCA), Perceptually
Important Algorithm (PIP), Data mining |
|
Type:
|
Short Paper |
|
First Page:
|
198 |
|
Last Page:
|
202 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
With the increase of economic globalization and evolution of information technology, high-dimensional stock data
reduction has become an essential part as pre-processing technique for data compression and effective future data mining
process. In this paper, we study the effect of dimension reduction technique, which is commonly used for
correlated multi-dimensional data. We use PCA as one of the representatives of the reduction techniques.
And we improve the results of Principle Component Analysis (PCA) by using proper pre-processing
approach based on Perceptually Important Point (PIP) algorithm. By using our approach, we can improve the
efficiency of dimension reduction to stock data. Encouraging experiment is reported from the tests that our
approach can provide a much higher reservation ratio for the reduced multi-dimensional stock data. |
|
|
|
|
|
|
