Authors:
Sang-Hong Lee
1
;
Dong-Kun Shin
2
and
Joon S. Lim
3
Affiliations:
1
Division of Software, Kyungwon University, Korea, Republic of
;
2
Sahmyook University, Korea, Republic of
;
3
Kyungwon University, Korea, Republic of
Keyword(s):
Fuzzy Neural Networks, Feature Selection, Principal Component Analysis, KOSPI
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Computational Intelligence
;
Decision Support Systems
;
Enterprise Software Technologies
;
Evolutionary Computing
;
Expert Systems
;
Health Information Systems
;
Knowledge Discovery and Information Retrieval
;
Knowledge Engineering and Ontology Development
;
Knowledge-Based Systems
;
Machine Learning
;
Soft Computing
;
Software Engineering
;
Symbolic Systems
Abstract:
This paper proposes stock forecasting using a principal component analysis (PCA) and a non-overlap area distribution measurement method based on a neural network with weighted fuzzy membership functions (NEWFM). The non-overlap area distribution measurement method selects the minimum number of four input features with the highest performance result from 12 initial input features by removing the worst input features one by one. PCA is a vector space transform often used for reducing multidimensional data sets to lower dimensions for analysis. The seven dimensional data sets with the highest performance result are extracted by PCA. The highest performance results in a non-overlap area distribution measurement method and PCA are 58.35% as the same results.