Random Forest Similarity Maps: A Scalable Visual Representation for Global and Local Interpretation

Machine Learning prediction algorithms have made significant contributions in today’s world, leading to increased usage in various domains. However, as ML algorithms surge, the need for transparent and interpretable models becomes essential. Visual representations have shown to be instrumental in ad...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Dipankar Mazumdar, Mário Popolin Neto, Fernando V. Paulovich
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	Random Forest classification model visualization explainable artificial intelligence (XAI) dimensionality reduction Electronics TK7800-8360
Acceso en línea:	https://doaj.org/article/7a08344bc0154928a55d6d7975855372
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:7a08344bc0154928a55d6d7975855372
record_format	dspace
spelling	oai:doaj.org-article:7a08344bc0154928a55d6d79758553722021-11-25T17:25:25ZRandom Forest Similarity Maps: A Scalable Visual Representation for Global and Local Interpretation10.3390/electronics102228622079-9292https://doaj.org/article/7a08344bc0154928a55d6d79758553722021-11-01T00:00:00Zhttps://www.mdpi.com/2079-9292/10/22/2862https://doaj.org/toc/2079-9292Machine Learning prediction algorithms have made significant contributions in today’s world, leading to increased usage in various domains. However, as ML algorithms surge, the need for transparent and interpretable models becomes essential. Visual representations have shown to be instrumental in addressing such an issue, allowing users to grasp models’ inner workings. Despite their popularity, visualization techniques still present visual scalability limitations, mainly when applied to analyze popular and complex models, such as Random Forests (RF). In this work, we propose <i>Random Forest Similarity Map (RFMap)</i>, a scalable interactive visual analytics tool designed to analyze RF ensemble models. <i>RFMap</i> focuses on explaining the inner working mechanism of models through different views describing individual data instance predictions, providing an overview of the entire forest of trees, and highlighting instance input feature values. The interactive nature of <i>RFMap</i> allows users to visually interpret model errors and decisions, establishing the necessary confidence and user trust in RF models and improving performance.Dipankar MazumdarMário Popolin NetoFernando V. PaulovichMDPI AGarticleRandom Forestclassification model visualizationexplainable artificial intelligence (XAI)dimensionality reductionElectronicsTK7800-8360ENElectronics, Vol 10, Iss 2862, p 2862 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Random Forest classification model visualization explainable artificial intelligence (XAI) dimensionality reduction Electronics TK7800-8360
spellingShingle	Random Forest classification model visualization explainable artificial intelligence (XAI) dimensionality reduction Electronics TK7800-8360 Dipankar Mazumdar Mário Popolin Neto Fernando V. Paulovich Random Forest Similarity Maps: A Scalable Visual Representation for Global and Local Interpretation
description	Machine Learning prediction algorithms have made significant contributions in today’s world, leading to increased usage in various domains. However, as ML algorithms surge, the need for transparent and interpretable models becomes essential. Visual representations have shown to be instrumental in addressing such an issue, allowing users to grasp models’ inner workings. Despite their popularity, visualization techniques still present visual scalability limitations, mainly when applied to analyze popular and complex models, such as Random Forests (RF). In this work, we propose <i>Random Forest Similarity Map (RFMap)</i>, a scalable interactive visual analytics tool designed to analyze RF ensemble models. <i>RFMap</i> focuses on explaining the inner working mechanism of models through different views describing individual data instance predictions, providing an overview of the entire forest of trees, and highlighting instance input feature values. The interactive nature of <i>RFMap</i> allows users to visually interpret model errors and decisions, establishing the necessary confidence and user trust in RF models and improving performance.
format	article
author	Dipankar Mazumdar Mário Popolin Neto Fernando V. Paulovich
author_facet	Dipankar Mazumdar Mário Popolin Neto Fernando V. Paulovich
author_sort	Dipankar Mazumdar
title	Random Forest Similarity Maps: A Scalable Visual Representation for Global and Local Interpretation
title_short	Random Forest Similarity Maps: A Scalable Visual Representation for Global and Local Interpretation
title_full	Random Forest Similarity Maps: A Scalable Visual Representation for Global and Local Interpretation
title_fullStr	Random Forest Similarity Maps: A Scalable Visual Representation for Global and Local Interpretation
title_full_unstemmed	Random Forest Similarity Maps: A Scalable Visual Representation for Global and Local Interpretation
title_sort	random forest similarity maps: a scalable visual representation for global and local interpretation
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/7a08344bc0154928a55d6d7975855372
work_keys_str_mv	AT dipankarmazumdar randomforestsimilaritymapsascalablevisualrepresentationforglobalandlocalinterpretation AT mariopopolinneto randomforestsimilaritymapsascalablevisualrepresentationforglobalandlocalinterpretation AT fernandovpaulovich randomforestsimilaritymapsascalablevisualrepresentationforglobalandlocalinterpretation
_version_	1718412357690982400

Random Forest Similarity Maps: A Scalable Visual Representation for Global and Local Interpretation

Ejemplares similares