The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Vues en texte intégral
146
La surveillance spatiale prédictive, qui prédit des informations spatiales telles que le trafic routier, a beaucoup retenu l'attention dans le contexte des villes intelligentes. L'apprentissage automatique permet une surveillance spatiale prédictive en utilisant une grande quantité de données de capteurs agrégées. La capacité des réseaux mobiles étant strictement limitée, de sérieux retards de transmission se produisent lorsque le trafic de communication est important. Si certaines des données utilisées pour la surveillance spatiale prédictive n'arrivent pas à temps, la précision de la prédiction se dégrade car la prédiction doit être effectuée en utilisant uniquement les données reçues, ce qui implique que les données de prédiction sont « sensibles aux délais ». Une technique d'allocation basée sur l'utilité a suggéré la modélisation des caractéristiques temporelles de ces données sensibles au retard pour une transmission prioritaire. Cependant, aucune étude n’a abordé le modèle temporel pour une transmission prioritaire dans la surveillance spatiale prédictive. Par conséquent, cet article propose un schéma qui permet la création d’un modèle temporel pour la surveillance spatiale prédictive. Le schéma se compose grossièrement de deux étapes : la première consiste à créer des données d'entraînement à partir de données de séries chronologiques originales et d'un modèle d'apprentissage automatique pouvant utiliser les données, tandis que la deuxième étape consiste à modéliser un modèle temporel à l'aide de la sélection de fonctionnalités dans le modèle d'apprentissage. La sélection des fonctionnalités permet d'estimer l'importance des données en termes de contribution des données à la précision des prédictions à partir du modèle d'apprentissage automatique. Cet article considère la prévision du trafic routier comme un scénario et montre que les modèles temporels créés avec le schéma proposé peuvent gérer des ensembles de données spatiales réelles. Une étude numérique a démontré comment notre modèle temporel fonctionne efficacement dans la transmission prioritaire pour la surveillance spatiale prédictive en termes de précision de prédiction.
Keiichiro SATO
Kyoto University
Ryoichi SHINKUMA
Kyoto University
Takehiro SATO
Kyoto University
Eiji OKI
Kyoto University
Takanori IWAI
NEC Corporation
Takeo ONISHI
NEC Corporation
Takahiro NOBUKIYO
NEC Corporation
Dai KANETOMO
NEC Corporation
Kozo SATODA
NEC Corporation
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Keiichiro SATO, Ryoichi SHINKUMA, Takehiro SATO, Eiji OKI, Takanori IWAI, Takeo ONISHI, Takahiro NOBUKIYO, Dai KANETOMO, Kozo SATODA, "Creation of Temporal Model for Prioritized Transmission in Predictive Spatial-Monitoring Using Machine Learning" in IEICE TRANSACTIONS on Communications,
vol. E104-B, no. 8, pp. 951-960, August 2021, doi: 10.1587/transcom.2020EBP3175.
Abstract: Predictive spatial-monitoring, which predicts spatial information such as road traffic, has attracted much attention in the context of smart cities. Machine learning enables predictive spatial-monitoring by using a large amount of aggregated sensor data. Since the capacity of mobile networks is strictly limited, serious transmission delays occur when loads of communication traffic are heavy. If some of the data used for predictive spatial-monitoring do not arrive on time, prediction accuracy degrades because the prediction has to be done using only the received data, which implies that data for prediction are ‘delay-sensitive’. A utility-based allocation technique has suggested modeling of temporal characteristics of such delay-sensitive data for prioritized transmission. However, no study has addressed temporal model for prioritized transmission in predictive spatial-monitoring. Therefore, this paper proposes a scheme that enables the creation of a temporal model for predictive spatial-monitoring. The scheme is roughly composed of two steps: the first involves creating training data from original time-series data and a machine learning model that can use the data, while the second step involves modeling a temporal model using feature selection in the learning model. Feature selection enables the estimation of the importance of data in terms of how much the data contribute to prediction accuracy from the machine learning model. This paper considers road-traffic prediction as a scenario and shows that the temporal models created with the proposed scheme can handle real spatial datasets. A numerical study demonstrated how our temporal model works effectively in prioritized transmission for predictive spatial-monitoring in terms of prediction accuracy.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2020EBP3175/_p
Copier
@ARTICLE{e104-b_8_951,
author={Keiichiro SATO, Ryoichi SHINKUMA, Takehiro SATO, Eiji OKI, Takanori IWAI, Takeo ONISHI, Takahiro NOBUKIYO, Dai KANETOMO, Kozo SATODA, },
journal={IEICE TRANSACTIONS on Communications},
title={Creation of Temporal Model for Prioritized Transmission in Predictive Spatial-Monitoring Using Machine Learning},
year={2021},
volume={E104-B},
number={8},
pages={951-960},
abstract={Predictive spatial-monitoring, which predicts spatial information such as road traffic, has attracted much attention in the context of smart cities. Machine learning enables predictive spatial-monitoring by using a large amount of aggregated sensor data. Since the capacity of mobile networks is strictly limited, serious transmission delays occur when loads of communication traffic are heavy. If some of the data used for predictive spatial-monitoring do not arrive on time, prediction accuracy degrades because the prediction has to be done using only the received data, which implies that data for prediction are ‘delay-sensitive’. A utility-based allocation technique has suggested modeling of temporal characteristics of such delay-sensitive data for prioritized transmission. However, no study has addressed temporal model for prioritized transmission in predictive spatial-monitoring. Therefore, this paper proposes a scheme that enables the creation of a temporal model for predictive spatial-monitoring. The scheme is roughly composed of two steps: the first involves creating training data from original time-series data and a machine learning model that can use the data, while the second step involves modeling a temporal model using feature selection in the learning model. Feature selection enables the estimation of the importance of data in terms of how much the data contribute to prediction accuracy from the machine learning model. This paper considers road-traffic prediction as a scenario and shows that the temporal models created with the proposed scheme can handle real spatial datasets. A numerical study demonstrated how our temporal model works effectively in prioritized transmission for predictive spatial-monitoring in terms of prediction accuracy.},
keywords={},
doi={10.1587/transcom.2020EBP3175},
ISSN={1745-1345},
month={August},}
Copier
TY - JOUR
TI - Creation of Temporal Model for Prioritized Transmission in Predictive Spatial-Monitoring Using Machine Learning
T2 - IEICE TRANSACTIONS on Communications
SP - 951
EP - 960
AU - Keiichiro SATO
AU - Ryoichi SHINKUMA
AU - Takehiro SATO
AU - Eiji OKI
AU - Takanori IWAI
AU - Takeo ONISHI
AU - Takahiro NOBUKIYO
AU - Dai KANETOMO
AU - Kozo SATODA
PY - 2021
DO - 10.1587/transcom.2020EBP3175
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E104-B
IS - 8
JA - IEICE TRANSACTIONS on Communications
Y1 - August 2021
AB - Predictive spatial-monitoring, which predicts spatial information such as road traffic, has attracted much attention in the context of smart cities. Machine learning enables predictive spatial-monitoring by using a large amount of aggregated sensor data. Since the capacity of mobile networks is strictly limited, serious transmission delays occur when loads of communication traffic are heavy. If some of the data used for predictive spatial-monitoring do not arrive on time, prediction accuracy degrades because the prediction has to be done using only the received data, which implies that data for prediction are ‘delay-sensitive’. A utility-based allocation technique has suggested modeling of temporal characteristics of such delay-sensitive data for prioritized transmission. However, no study has addressed temporal model for prioritized transmission in predictive spatial-monitoring. Therefore, this paper proposes a scheme that enables the creation of a temporal model for predictive spatial-monitoring. The scheme is roughly composed of two steps: the first involves creating training data from original time-series data and a machine learning model that can use the data, while the second step involves modeling a temporal model using feature selection in the learning model. Feature selection enables the estimation of the importance of data in terms of how much the data contribute to prediction accuracy from the machine learning model. This paper considers road-traffic prediction as a scenario and shows that the temporal models created with the proposed scheme can handle real spatial datasets. A numerical study demonstrated how our temporal model works effectively in prioritized transmission for predictive spatial-monitoring in terms of prediction accuracy.
ER -