A new dynamic approach for non-singleton fuzzification in noisy time-series prediction

Non-singleton fuzzification is used to model uncertain (e.g. noisy) inputs within fuzzy logic systems. In the standard approach, assuming the fuzzification type is known, the observed [noisy] input is usually considered to be the core of the input fuzzy set, usually being the centre of its membershi...

Full description

Bibliographic Details
Main Authors: Pourabdollah, Amir, John, Robert, Garibaldi, Jonathan M.
Format: Conference or Workshop Item
Published: 2017
Subjects:
Online Access:https://eprints.nottingham.ac.uk/45209/
_version_ 1848797090095824896
author Pourabdollah, Amir
John, Robert
Garibaldi, Jonathan M.
author_facet Pourabdollah, Amir
John, Robert
Garibaldi, Jonathan M.
author_sort Pourabdollah, Amir
building Nottingham Research Data Repository
collection Online Access
description Non-singleton fuzzification is used to model uncertain (e.g. noisy) inputs within fuzzy logic systems. In the standard approach, assuming the fuzzification type is known, the observed [noisy] input is usually considered to be the core of the input fuzzy set, usually being the centre of its membership function. This paper proposes a new fuzzification method (not type), in which the core of an input fuzzy set is not necessarily located at the observed input, rather it is dynamically adjusted based on statistical methods. Using the weighted moving average, a few past samples are aggregated to roughly estimate where the input fuzzy set should be located. While the added complexity is not huge, applying this method to the well-known Mackey-Glass and Lorenz time-series prediction problems, show significant error reduction when the input is corrupted by different noise levels.
first_indexed 2025-11-14T19:58:21Z
format Conference or Workshop Item
id nottingham-45209
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:58:21Z
publishDate 2017
recordtype eprints
repository_type Digital Repository
spelling nottingham-452092020-05-04T19:01:59Z https://eprints.nottingham.ac.uk/45209/ A new dynamic approach for non-singleton fuzzification in noisy time-series prediction Pourabdollah, Amir John, Robert Garibaldi, Jonathan M. Non-singleton fuzzification is used to model uncertain (e.g. noisy) inputs within fuzzy logic systems. In the standard approach, assuming the fuzzification type is known, the observed [noisy] input is usually considered to be the core of the input fuzzy set, usually being the centre of its membership function. This paper proposes a new fuzzification method (not type), in which the core of an input fuzzy set is not necessarily located at the observed input, rather it is dynamically adjusted based on statistical methods. Using the weighted moving average, a few past samples are aggregated to roughly estimate where the input fuzzy set should be located. While the added complexity is not huge, applying this method to the well-known Mackey-Glass and Lorenz time-series prediction problems, show significant error reduction when the input is corrupted by different noise levels. 2017-08-24 Conference or Workshop Item PeerReviewed Pourabdollah, Amir, John, Robert and Garibaldi, Jonathan M. (2017) A new dynamic approach for non-singleton fuzzification in noisy time-series prediction. In: 2017 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), 9-12 July 2017, Naples, Italy. Noise measurement Standards Fuzzy sets Fuzzy logic Uncertainty Time series analysis Estimation http://ieeexplore.ieee.org/abstract/document/8015575/
spellingShingle Noise measurement
Standards
Fuzzy sets
Fuzzy logic
Uncertainty
Time series analysis
Estimation
Pourabdollah, Amir
John, Robert
Garibaldi, Jonathan M.
A new dynamic approach for non-singleton fuzzification in noisy time-series prediction
title A new dynamic approach for non-singleton fuzzification in noisy time-series prediction
title_full A new dynamic approach for non-singleton fuzzification in noisy time-series prediction
title_fullStr A new dynamic approach for non-singleton fuzzification in noisy time-series prediction
title_full_unstemmed A new dynamic approach for non-singleton fuzzification in noisy time-series prediction
title_short A new dynamic approach for non-singleton fuzzification in noisy time-series prediction
title_sort new dynamic approach for non-singleton fuzzification in noisy time-series prediction
topic Noise measurement
Standards
Fuzzy sets
Fuzzy logic
Uncertainty
Time series analysis
Estimation
url https://eprints.nottingham.ac.uk/45209/
https://eprints.nottingham.ac.uk/45209/