Automatic Detection and Correction of Multi-class Classification Errors Using System Whole-part Relationships
Real-world dynamic systems such as physical and atmosphereocean systems often exhibit a hierarchical system-subsystem structure. However, the paradigm of making this hierarchical/modular structure and the rich properties they encode a “first-class citizen” of machine learning algorithms is largely a...
Main Authors: | , , , , , , , |
---|---|
Other Authors: | |
Format: | Text |
Language: | English |
Subjects: | |
Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.414.165 http://cucis.ece.northwestern.edu/publications/pdf/CheJen13.pdf |
id |
ftciteseerx:oai:CiteSeerX.psu:10.1.1.414.165 |
---|---|
record_format |
openpolar |
spelling |
ftciteseerx:oai:CiteSeerX.psu:10.1.1.414.165 2023-05-15T17:32:53+02:00 Automatic Detection and Correction of Multi-class Classification Errors Using System Whole-part Relationships Zhengzhang Chen John Jenkins Alok Choudhary Jinfeng Rao Fredrick Semazzi Anatoli V. Melechko Vipin Kumar Nagiza F. Samatova The Pennsylvania State University CiteSeerX Archives application/pdf http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.414.165 http://cucis.ece.northwestern.edu/publications/pdf/CheJen13.pdf en eng http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.414.165 http://cucis.ece.northwestern.edu/publications/pdf/CheJen13.pdf Metadata may be used without restrictions as long as the oai identifier remains attached to it. http://cucis.ece.northwestern.edu/publications/pdf/CheJen13.pdf text ftciteseerx 2016-01-08T03:31:21Z Real-world dynamic systems such as physical and atmosphereocean systems often exhibit a hierarchical system-subsystem structure. However, the paradigm of making this hierarchical/modular structure and the rich properties they encode a “first-class citizen” of machine learning algorithms is largely absent from the literature. Furthermore, traditional data mining approaches focus on designing new classifiers or ensembles of classifiers, while there is a lack of study on detecting and correcting prediction errors of existing forecasting (or classification) algorithms. In this paper, we propose DETECTOR, a hierarchical method for detecting and correcting forecast errors by employing the whole-part relationships between the target system and non-target systems. Experimental results show that DETECTOR can successfully detect and correct forecasting errors made by state-of-art classifier ensemble techniques and traditional single classifier methods at an average rate of 22%, corresponding to a 11 % average forecasting accuracy increase, in seasonal forecasting of hurricanes and landfalling hurricanes in North Atlantic and North African rainfall. 1 Text North Atlantic Unknown |
institution |
Open Polar |
collection |
Unknown |
op_collection_id |
ftciteseerx |
language |
English |
description |
Real-world dynamic systems such as physical and atmosphereocean systems often exhibit a hierarchical system-subsystem structure. However, the paradigm of making this hierarchical/modular structure and the rich properties they encode a “first-class citizen” of machine learning algorithms is largely absent from the literature. Furthermore, traditional data mining approaches focus on designing new classifiers or ensembles of classifiers, while there is a lack of study on detecting and correcting prediction errors of existing forecasting (or classification) algorithms. In this paper, we propose DETECTOR, a hierarchical method for detecting and correcting forecast errors by employing the whole-part relationships between the target system and non-target systems. Experimental results show that DETECTOR can successfully detect and correct forecasting errors made by state-of-art classifier ensemble techniques and traditional single classifier methods at an average rate of 22%, corresponding to a 11 % average forecasting accuracy increase, in seasonal forecasting of hurricanes and landfalling hurricanes in North Atlantic and North African rainfall. 1 |
author2 |
The Pennsylvania State University CiteSeerX Archives |
format |
Text |
author |
Zhengzhang Chen John Jenkins Alok Choudhary Jinfeng Rao Fredrick Semazzi Anatoli V. Melechko Vipin Kumar Nagiza F. Samatova |
spellingShingle |
Zhengzhang Chen John Jenkins Alok Choudhary Jinfeng Rao Fredrick Semazzi Anatoli V. Melechko Vipin Kumar Nagiza F. Samatova Automatic Detection and Correction of Multi-class Classification Errors Using System Whole-part Relationships |
author_facet |
Zhengzhang Chen John Jenkins Alok Choudhary Jinfeng Rao Fredrick Semazzi Anatoli V. Melechko Vipin Kumar Nagiza F. Samatova |
author_sort |
Zhengzhang Chen |
title |
Automatic Detection and Correction of Multi-class Classification Errors Using System Whole-part Relationships |
title_short |
Automatic Detection and Correction of Multi-class Classification Errors Using System Whole-part Relationships |
title_full |
Automatic Detection and Correction of Multi-class Classification Errors Using System Whole-part Relationships |
title_fullStr |
Automatic Detection and Correction of Multi-class Classification Errors Using System Whole-part Relationships |
title_full_unstemmed |
Automatic Detection and Correction of Multi-class Classification Errors Using System Whole-part Relationships |
title_sort |
automatic detection and correction of multi-class classification errors using system whole-part relationships |
url |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.414.165 http://cucis.ece.northwestern.edu/publications/pdf/CheJen13.pdf |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_source |
http://cucis.ece.northwestern.edu/publications/pdf/CheJen13.pdf |
op_relation |
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.414.165 http://cucis.ece.northwestern.edu/publications/pdf/CheJen13.pdf |
op_rights |
Metadata may be used without restrictions as long as the oai identifier remains attached to it. |
_version_ |
1766131186173739008 |