|
|
Bimonthly Since 1986 |
ISSN 1004-9037
|
|
|
|
|
Publication Details |
Edited by: Editorial Board of Journal of Data Acquisition and Processing
P.O. Box 2704, Beijing 100190, P.R. China
Sponsored by: Institute of Computing Technology, CAS & China Computer Federation
Undertaken by: Institute of Computing Technology, CAS
Published by: SCIENCE PRESS, BEIJING, CHINA
Distributed by:
China: All Local Post Offices
|
|
|
|
|
|
|
|
|
|
Abstract
Rough sets are efficient in approximately represent the whole space associated with data. Rough neural networks are the phenomena based on rough sets on top of neural networks. In the existing systems, the rough set learning is compatible with one decision table. However, in many real-world applications, there is need for multiple decision tables. To address this problem, this paper proposes a novel approach with rough sets based on deep learning architecture. In other words, a deep rough set theory is used to define a deep learning model with multiple decision tables a to realize a decision support system. The proposed deep learning architecture considers each decision table’s local properties to arrive at an approximate decision globally. We also proposed an algorithm known as Deep Neural Network on Rough Sets (DNNRS) to realize the proposed deep learning architecture. The ability of the proposed model to handle multiple decisions tables with soft computing and machine learning paves way for solving real world problems. The proposed rough sets based deep learning model has capability that mimics human brain thinking which is suitable in many decision support systems. Empirical study has revealed the significance of the proposed model.
Keyword
Rough Sets, Deep Learning, Rough Neural Networks, Machine Learning, Soft Computing
PDF Download (click here)
|
|
|
|
|