Simulation Study on How Input Data Affects Time-Series Classification Model Results

This paper discusses the results of a study investigating how input data characteristics affect the performance of time-series classification models. In this experiment, we used 82 synthetically generated time-series datasets, created based on predefined functions with added noise. These datasets va...

Full description

Saved in:
Bibliographic Details
Main Authors: Maria Sadowska, Krzysztof Gajowniczek
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Entropy
Subjects:
Online Access:https://www.mdpi.com/1099-4300/27/6/624
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849472190382080000
author Maria Sadowska
Krzysztof Gajowniczek
author_facet Maria Sadowska
Krzysztof Gajowniczek
author_sort Maria Sadowska
collection DOAJ
description This paper discusses the results of a study investigating how input data characteristics affect the performance of time-series classification models. In this experiment, we used 82 synthetically generated time-series datasets, created based on predefined functions with added noise. These datasets varied in structure, including differences in the number of classes and noise levels, while maintaining a consistent length and total number of observations. This design allowed us to systematically assess the influence of dataset characteristics on classification outcomes. Seven classification models were evaluated and their performance was compared using accuracy metrics, training time and memory requirements. According to the evaluation, the CNN Classifier achieved the best results, demonstrating the highest robustness to an increasing number of classes and noise. In contrast, the least effective model was the Catch22 Classifier. Overall, the performed research leads to the conclusion that as the number of classes and the level of noise in the data increase, all classification models become less effective, achieving lower accuracy metrics.
format Article
id doaj-art-58e2b387202c4ebb9e9943d40d24aec2
institution Kabale University
issn 1099-4300
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Entropy
spelling doaj-art-58e2b387202c4ebb9e9943d40d24aec22025-08-20T03:24:36ZengMDPI AGEntropy1099-43002025-06-0127662410.3390/e27060624Simulation Study on How Input Data Affects Time-Series Classification Model ResultsMaria Sadowska0Krzysztof Gajowniczek1Institute of Information Technology, Warsaw University of Life Sciences-SGGW, 02-787 Warszawa, PolandInstitute of Information Technology, Warsaw University of Life Sciences-SGGW, 02-787 Warszawa, PolandThis paper discusses the results of a study investigating how input data characteristics affect the performance of time-series classification models. In this experiment, we used 82 synthetically generated time-series datasets, created based on predefined functions with added noise. These datasets varied in structure, including differences in the number of classes and noise levels, while maintaining a consistent length and total number of observations. This design allowed us to systematically assess the influence of dataset characteristics on classification outcomes. Seven classification models were evaluated and their performance was compared using accuracy metrics, training time and memory requirements. According to the evaluation, the CNN Classifier achieved the best results, demonstrating the highest robustness to an increasing number of classes and noise. In contrast, the least effective model was the Catch22 Classifier. Overall, the performed research leads to the conclusion that as the number of classes and the level of noise in the data increase, all classification models become less effective, achieving lower accuracy metrics.https://www.mdpi.com/1099-4300/27/6/624time seriesclassificationsynthetic data
spellingShingle Maria Sadowska
Krzysztof Gajowniczek
Simulation Study on How Input Data Affects Time-Series Classification Model Results
Entropy
time series
classification
synthetic data
title Simulation Study on How Input Data Affects Time-Series Classification Model Results
title_full Simulation Study on How Input Data Affects Time-Series Classification Model Results
title_fullStr Simulation Study on How Input Data Affects Time-Series Classification Model Results
title_full_unstemmed Simulation Study on How Input Data Affects Time-Series Classification Model Results
title_short Simulation Study on How Input Data Affects Time-Series Classification Model Results
title_sort simulation study on how input data affects time series classification model results
topic time series
classification
synthetic data
url https://www.mdpi.com/1099-4300/27/6/624
work_keys_str_mv AT mariasadowska simulationstudyonhowinputdataaffectstimeseriesclassificationmodelresults
AT krzysztofgajowniczek simulationstudyonhowinputdataaffectstimeseriesclassificationmodelresults