Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed Learners

<p dir="ltr">Internet of Things applications can greatly benefit from accurate prediction models. The performance of prediction models is highly dependent on the quantity and quality of their training data. In this paper, we investigate the creation of a dynamic ensemble from distrib...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Mehdi Mohammadi (5024105) (author)
مؤلفون آخرون: Ala Al-Fuqaha (4434340) (author)
منشور في: 2018
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513526006022144
author Mehdi Mohammadi (5024105)
author2 Ala Al-Fuqaha (4434340)
author2_role author
author_facet Mehdi Mohammadi (5024105)
Ala Al-Fuqaha (4434340)
author_role author
dc.creator.none.fl_str_mv Mehdi Mohammadi (5024105)
Ala Al-Fuqaha (4434340)
dc.date.none.fl_str_mv 2018-10-21T09:00:00Z
dc.identifier.none.fl_str_mv 10.1109/access.2018.2877153
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Exploiting_the_Spatio-Temporal_Patterns_in_IoT_Data_to_Establish_a_Dynamic_Ensemble_of_Distributed_Learners/25304245
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Engineering
Materials engineering
Data models
Predictive models
Training data
Load modeling
Computational modeling
Sensors
Distributed deep neural networks
spatio-temporal analysis
ensemble deep learning
bloom filter
Internet of Things
smart city
dc.title.none.fl_str_mv Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed Learners
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Internet of Things applications can greatly benefit from accurate prediction models. The performance of prediction models is highly dependent on the quantity and quality of their training data. In this paper, we investigate the creation of a dynamic ensemble from distributed deep learning models by considering the spatiotemporal patterns embedded in the training data. Our dynamic ensemble does not depend on offline configurations. Instead, it exploits the spatiotemporal patterns embedded in the training data to generate dynamic weights for the underlying weak distributed deep learners to create a stronger learner. Our evaluation experiments using three real-world datasets in the context of the smart city show that our proposed dynamic ensemble strategy leads to an improved error rate of up to 33% compared to the baseline strategy even when using31of the training data. Moreover, using only 20% of the training data, the error rate of the model slightly increased by up to 2 in terms of mean square error. This increase is 82% less than the 11.3 increase seen in the baseline model. Therefore, our approach contributes to the reduced network traffic while not hindering the accuracy significantly.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2018.2877153" target="_blank">https://dx.doi.org/10.1109/access.2018.2877153</a></p>
eu_rights_str_mv openAccess
id Manara2_0e5ac34dd7b24bd02a2d77990a686f2c
identifier_str_mv 10.1109/access.2018.2877153
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/25304245
publishDate 2018
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed LearnersMehdi Mohammadi (5024105)Ala Al-Fuqaha (4434340)EngineeringMaterials engineeringData modelsPredictive modelsTraining dataLoad modelingComputational modelingSensorsDistributed deep neural networksspatio-temporal analysisensemble deep learningbloom filterInternet of Thingssmart city<p dir="ltr">Internet of Things applications can greatly benefit from accurate prediction models. The performance of prediction models is highly dependent on the quantity and quality of their training data. In this paper, we investigate the creation of a dynamic ensemble from distributed deep learning models by considering the spatiotemporal patterns embedded in the training data. Our dynamic ensemble does not depend on offline configurations. Instead, it exploits the spatiotemporal patterns embedded in the training data to generate dynamic weights for the underlying weak distributed deep learners to create a stronger learner. Our evaluation experiments using three real-world datasets in the context of the smart city show that our proposed dynamic ensemble strategy leads to an improved error rate of up to 33% compared to the baseline strategy even when using31of the training data. Moreover, using only 20% of the training data, the error rate of the model slightly increased by up to 2 in terms of mean square error. This increase is 82% less than the 11.3 increase seen in the baseline model. Therefore, our approach contributes to the reduced network traffic while not hindering the accuracy significantly.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/access.2018.2877153" target="_blank">https://dx.doi.org/10.1109/access.2018.2877153</a></p>2018-10-21T09:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/access.2018.2877153https://figshare.com/articles/journal_contribution/Exploiting_the_Spatio-Temporal_Patterns_in_IoT_Data_to_Establish_a_Dynamic_Ensemble_of_Distributed_Learners/25304245CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/253042452018-10-21T09:00:00Z
spellingShingle Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed Learners
Mehdi Mohammadi (5024105)
Engineering
Materials engineering
Data models
Predictive models
Training data
Load modeling
Computational modeling
Sensors
Distributed deep neural networks
spatio-temporal analysis
ensemble deep learning
bloom filter
Internet of Things
smart city
status_str publishedVersion
title Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed Learners
title_full Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed Learners
title_fullStr Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed Learners
title_full_unstemmed Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed Learners
title_short Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed Learners
title_sort Exploiting the Spatio-Temporal Patterns in IoT Data to Establish a Dynamic Ensemble of Distributed Learners
topic Engineering
Materials engineering
Data models
Predictive models
Training data
Load modeling
Computational modeling
Sensors
Distributed deep neural networks
spatio-temporal analysis
ensemble deep learning
bloom filter
Internet of Things
smart city