Fuzzy associative learning of feature dependency for time series forecasting
Neuro-fuzzy system (NFS) has successfully been widely applied in solving problems across diverse fields, such as signal detection, fault detection, and forecasting. In recent years, many forecasting problems require the processing and learning of large number of dynamic data streams. Existing system...
Saved in:
Main Authors: | , , , |
---|---|
Other Authors: | |
Format: | Conference or Workshop Item |
Language: | English |
Published: |
2013
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/97871 http://hdl.handle.net/10220/12421 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-97871 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-978712020-05-28T07:17:46Z Fuzzy associative learning of feature dependency for time series forecasting Ng, See Kiong Cheu, Eng Yeow Sim, Kelvin Quek, Chai School of Computer Engineering International Joint Conference on Neural Networks (2012 : Brisbane, Australia) DRNTU::Engineering::Computer science and engineering Neuro-fuzzy system (NFS) has successfully been widely applied in solving problems across diverse fields, such as signal detection, fault detection, and forecasting. In recent years, many forecasting problems require the processing and learning of large number of dynamic data streams. Existing systems are inadequate in handling this type of complex problem. This paper presents a learning system that incorporates an evolving correlation-based feature selector to handle the high dimensionality of the data streams, and an evolving NFS to sequentially model and extract fuzzy knowledge about these data streams. The proposed system requires no prior knowledge of the data, reads the stream of data in a single pass, and accounts for the time-varying characteristics of the data. These three features allow the system to handle large and dynamic data. The effectiveness of the proposed system is validated on both synthetic and real-world problems. The experiments illustrate the viability of the proposed learning technique, and exemplifies how it can outperform existing NFS. Experiment on real-world stock price forecasting shows a remarkable reduction of error rate by 15.4%. 2013-07-29T03:21:11Z 2019-12-06T19:47:32Z 2013-07-29T03:21:11Z 2019-12-06T19:47:32Z 2012 2012 Conference Paper Cheu, E. Y., Sim, K., Ng, S. K., & Quek, C. (2012). Fuzzy associative learning of feature dependency for time series forecasting. The 2012 International Joint Conference on Neural Networks (IJCNN). https://hdl.handle.net/10356/97871 http://hdl.handle.net/10220/12421 10.1109/IJCNN.2012.6252542 en © 2012 IEEE. |
institution |
Nanyang Technological University |
building |
NTU Library |
country |
Singapore |
collection |
DR-NTU |
language |
English |
topic |
DRNTU::Engineering::Computer science and engineering |
spellingShingle |
DRNTU::Engineering::Computer science and engineering Ng, See Kiong Cheu, Eng Yeow Sim, Kelvin Quek, Chai Fuzzy associative learning of feature dependency for time series forecasting |
description |
Neuro-fuzzy system (NFS) has successfully been widely applied in solving problems across diverse fields, such as signal detection, fault detection, and forecasting. In recent years, many forecasting problems require the processing and learning of large number of dynamic data streams. Existing systems are inadequate in handling this type of complex problem. This paper presents a learning system that incorporates an evolving correlation-based feature selector to handle the high dimensionality of the data streams, and an evolving NFS to sequentially model and extract fuzzy knowledge about these data streams. The proposed system requires no prior knowledge of the data, reads the stream of data in a single pass, and accounts for the time-varying characteristics of the data. These three features allow the system to handle large and dynamic data. The effectiveness of the proposed system is validated on both synthetic and real-world problems. The experiments illustrate the viability of the proposed learning technique, and exemplifies how it can outperform existing NFS. Experiment on real-world stock price forecasting shows a remarkable reduction of error rate by 15.4%. |
author2 |
School of Computer Engineering |
author_facet |
School of Computer Engineering Ng, See Kiong Cheu, Eng Yeow Sim, Kelvin Quek, Chai |
format |
Conference or Workshop Item |
author |
Ng, See Kiong Cheu, Eng Yeow Sim, Kelvin Quek, Chai |
author_sort |
Ng, See Kiong |
title |
Fuzzy associative learning of feature dependency for time series forecasting |
title_short |
Fuzzy associative learning of feature dependency for time series forecasting |
title_full |
Fuzzy associative learning of feature dependency for time series forecasting |
title_fullStr |
Fuzzy associative learning of feature dependency for time series forecasting |
title_full_unstemmed |
Fuzzy associative learning of feature dependency for time series forecasting |
title_sort |
fuzzy associative learning of feature dependency for time series forecasting |
publishDate |
2013 |
url |
https://hdl.handle.net/10356/97871 http://hdl.handle.net/10220/12421 |
_version_ |
1681056868551622656 |