Cargando…
Overview of Machine Learning Process Modelling
Much research has been conducted in the area of machine learning algorithms; however, the question of a general description of an artificial learner’s (empirical) performance has mainly remained unanswered. A general, restrictions-free theory on its performance has not been developed yet. In this st...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8469427/ https://www.ncbi.nlm.nih.gov/pubmed/34573748 http://dx.doi.org/10.3390/e23091123 |
_version_ | 1784573928464711680 |
---|---|
author | Brumen, Boštjan Černezel, Aleš Bošnjak, Leon |
author_facet | Brumen, Boštjan Černezel, Aleš Bošnjak, Leon |
author_sort | Brumen, Boštjan |
collection | PubMed |
description | Much research has been conducted in the area of machine learning algorithms; however, the question of a general description of an artificial learner’s (empirical) performance has mainly remained unanswered. A general, restrictions-free theory on its performance has not been developed yet. In this study, we investigate which function most appropriately describes learning curves produced by several machine learning algorithms, and how well these curves can predict the future performance of an algorithm. Decision trees, neural networks, Naïve Bayes, and Support Vector Machines were applied to 130 datasets from publicly available repositories. Three different functions (power, logarithmic, and exponential) were fit to the measured outputs. Using rigorous statistical methods and two measures for the goodness-of-fit, the power law model proved to be the most appropriate model for describing the learning curve produced by the algorithms in terms of goodness-of-fit and prediction capabilities. The presented study, first of its kind in scale and rigour, provides results (and methods) that can be used to assess the performance of novel or existing artificial learners and forecast their ‘capacity to learn’ based on the amount of available or desired data. |
format | Online Article Text |
id | pubmed-8469427 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-84694272021-09-27 Overview of Machine Learning Process Modelling Brumen, Boštjan Černezel, Aleš Bošnjak, Leon Entropy (Basel) Article Much research has been conducted in the area of machine learning algorithms; however, the question of a general description of an artificial learner’s (empirical) performance has mainly remained unanswered. A general, restrictions-free theory on its performance has not been developed yet. In this study, we investigate which function most appropriately describes learning curves produced by several machine learning algorithms, and how well these curves can predict the future performance of an algorithm. Decision trees, neural networks, Naïve Bayes, and Support Vector Machines were applied to 130 datasets from publicly available repositories. Three different functions (power, logarithmic, and exponential) were fit to the measured outputs. Using rigorous statistical methods and two measures for the goodness-of-fit, the power law model proved to be the most appropriate model for describing the learning curve produced by the algorithms in terms of goodness-of-fit and prediction capabilities. The presented study, first of its kind in scale and rigour, provides results (and methods) that can be used to assess the performance of novel or existing artificial learners and forecast their ‘capacity to learn’ based on the amount of available or desired data. MDPI 2021-08-28 /pmc/articles/PMC8469427/ /pubmed/34573748 http://dx.doi.org/10.3390/e23091123 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Brumen, Boštjan Černezel, Aleš Bošnjak, Leon Overview of Machine Learning Process Modelling |
title | Overview of Machine Learning Process Modelling |
title_full | Overview of Machine Learning Process Modelling |
title_fullStr | Overview of Machine Learning Process Modelling |
title_full_unstemmed | Overview of Machine Learning Process Modelling |
title_short | Overview of Machine Learning Process Modelling |
title_sort | overview of machine learning process modelling |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8469427/ https://www.ncbi.nlm.nih.gov/pubmed/34573748 http://dx.doi.org/10.3390/e23091123 |
work_keys_str_mv | AT brumenbostjan overviewofmachinelearningprocessmodelling AT cernezelales overviewofmachinelearningprocessmodelling AT bosnjakleon overviewofmachinelearningprocessmodelling |