Online Sliding Window Based Self-Organising Fuzzy Neural Network for Cognitive Reasoning

G Leng, Anjan Ray, TM McGinnity, SA Coleman, LP Maguire

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose an online sliding window based selforganisingfuzzy neural network (SOFNN) as the corecomponent of a cognitive reasoning system for a smart homeenvironment. The network has the ability to configure itsneuronal structure through adding and pruning of neuronswhile exploring the relationships between the inputs and thedesired reasoning outputs, thus enabling continuous learningand reasoning to provide meaningful cognitive understandingof the environment. Initially, the network is trained withenvironmentally realistic synthesised data thus demonstratingits adaptation capabilities. The network is then validated usingunseen data. In the simulation, we have studied the networkstructures and responses for three different scenarios with andwithout online sliding window based approaches and theresults obtained show the effectiveness of the proposed method.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages114-119
Number of pages6
Publication statusPublished - 28 May 2013
EventCOGNITIVE 2013 : The Fifth International Conference on Advanced Cognitive Technologies and Applications - Valencia, Spain
Duration: 28 May 2013 → …

Conference

ConferenceCOGNITIVE 2013 : The Fifth International Conference on Advanced Cognitive Technologies and Applications
Period28/05/13 → …

Fingerprint

Fuzzy neural networks
Neural networks

Cite this

@inproceedings{ebc7f7538aee41938de8d345f65a7af0,
title = "Online Sliding Window Based Self-Organising Fuzzy Neural Network for Cognitive Reasoning",
abstract = "We propose an online sliding window based selforganisingfuzzy neural network (SOFNN) as the corecomponent of a cognitive reasoning system for a smart homeenvironment. The network has the ability to configure itsneuronal structure through adding and pruning of neuronswhile exploring the relationships between the inputs and thedesired reasoning outputs, thus enabling continuous learningand reasoning to provide meaningful cognitive understandingof the environment. Initially, the network is trained withenvironmentally realistic synthesised data thus demonstratingits adaptation capabilities. The network is then validated usingunseen data. In the simulation, we have studied the networkstructures and responses for three different scenarios with andwithout online sliding window based approaches and theresults obtained show the effectiveness of the proposed method.",
author = "G Leng and Anjan Ray and TM McGinnity and SA Coleman and LP Maguire",
year = "2013",
month = "5",
day = "28",
language = "English",
isbn = "978-1-61208-273-8",
pages = "114--119",
booktitle = "Unknown Host Publication",

}

Leng, G, Ray, A, McGinnity, TM, Coleman, SA & Maguire, LP 2013, Online Sliding Window Based Self-Organising Fuzzy Neural Network for Cognitive Reasoning. in Unknown Host Publication. pp. 114-119, COGNITIVE 2013 : The Fifth International Conference on Advanced Cognitive Technologies and Applications, 28/05/13.

Online Sliding Window Based Self-Organising Fuzzy Neural Network for Cognitive Reasoning. / Leng, G; Ray, Anjan; McGinnity, TM; Coleman, SA; Maguire, LP.

Unknown Host Publication. 2013. p. 114-119.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Online Sliding Window Based Self-Organising Fuzzy Neural Network for Cognitive Reasoning

AU - Leng, G

AU - Ray, Anjan

AU - McGinnity, TM

AU - Coleman, SA

AU - Maguire, LP

PY - 2013/5/28

Y1 - 2013/5/28

N2 - We propose an online sliding window based selforganisingfuzzy neural network (SOFNN) as the corecomponent of a cognitive reasoning system for a smart homeenvironment. The network has the ability to configure itsneuronal structure through adding and pruning of neuronswhile exploring the relationships between the inputs and thedesired reasoning outputs, thus enabling continuous learningand reasoning to provide meaningful cognitive understandingof the environment. Initially, the network is trained withenvironmentally realistic synthesised data thus demonstratingits adaptation capabilities. The network is then validated usingunseen data. In the simulation, we have studied the networkstructures and responses for three different scenarios with andwithout online sliding window based approaches and theresults obtained show the effectiveness of the proposed method.

AB - We propose an online sliding window based selforganisingfuzzy neural network (SOFNN) as the corecomponent of a cognitive reasoning system for a smart homeenvironment. The network has the ability to configure itsneuronal structure through adding and pruning of neuronswhile exploring the relationships between the inputs and thedesired reasoning outputs, thus enabling continuous learningand reasoning to provide meaningful cognitive understandingof the environment. Initially, the network is trained withenvironmentally realistic synthesised data thus demonstratingits adaptation capabilities. The network is then validated usingunseen data. In the simulation, we have studied the networkstructures and responses for three different scenarios with andwithout online sliding window based approaches and theresults obtained show the effectiveness of the proposed method.

M3 - Conference contribution

SN - 978-1-61208-273-8

SP - 114

EP - 119

BT - Unknown Host Publication

ER -