Proposal of Feature Value Selection Method for Time-Critical Learning

Kanami Yuyama, Hiroaki Nishi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The development of IoT has led to the creation of a data-enriched environment that enables data gathering by using distributed sensors and terminals. However, in this environment, the cost of data analysis has increased. Machine learning has gained attention for reducing the cost because enabling automatic data analysis, as well as multidimensional data, is expected. However, for enormous data, such as Big Data, we still have to pay costs. Therefore, selecting feature values when using machine learning technology is essential, especially as inputs of a classifier. Selecting the feature values increases its estimation accuracy. Moreover, the time cost, as well as calculation cost, needs consideration for the actual time-critical use of machine learning, especially in its learning process. Therefore, in this study, we proposed an algorithm that selected suitable feature values in required time. The proposed method consists of two stages: stepwise input selection stage using ANOVA and feature deletion stage according to the contribution rate of the features to estimate accuracy. These selection and deletion processes continue until the required processing time. We confirmed the efficiency of the proposed method by using an environment of a crystallization process in a factory and a household's occupancy estimation. A comparison with the original stepwise input method proved that the proposed method improved the estimation accuracy by 2% and 5% in the estimation of the substance amount of the crystallization process and household's occupancy, respectively.

Original languageEnglish
Title of host publicationProceedings - 2018 IEEE 23rd International Conference on Emerging Technologies and Factory Automation, ETFA 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1365-1371
Number of pages7
Volume2018-September
ISBN (Electronic)9781538671085
DOIs
Publication statusPublished - 2018 Oct 22
Event23rd IEEE International Conference on Emerging Technologies and Factory Automation, ETFA 2018 - Torino, Italy
Duration: 2018 Sep 42018 Sep 7

Other

Other23rd IEEE International Conference on Emerging Technologies and Factory Automation, ETFA 2018
CountryItaly
CityTorino
Period18/9/418/9/7

Fingerprint

Learning systems
Costs
Crystallization
Analysis of variance (ANOVA)
Industrial plants
Classifiers
Sensors
Processing
Internet of things
Big data

Keywords

  • ANOVA
  • Feature value selection
  • k-NN
  • real-time feature selection
  • time-constraint machine learning

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Industrial and Manufacturing Engineering
  • Computer Science Applications

Cite this

Yuyama, K., & Nishi, H. (2018). Proposal of Feature Value Selection Method for Time-Critical Learning. In Proceedings - 2018 IEEE 23rd International Conference on Emerging Technologies and Factory Automation, ETFA 2018 (Vol. 2018-September, pp. 1365-1371). [8502622] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ETFA.2018.8502622

Proposal of Feature Value Selection Method for Time-Critical Learning. / Yuyama, Kanami; Nishi, Hiroaki.

Proceedings - 2018 IEEE 23rd International Conference on Emerging Technologies and Factory Automation, ETFA 2018. Vol. 2018-September Institute of Electrical and Electronics Engineers Inc., 2018. p. 1365-1371 8502622.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yuyama, K & Nishi, H 2018, Proposal of Feature Value Selection Method for Time-Critical Learning. in Proceedings - 2018 IEEE 23rd International Conference on Emerging Technologies and Factory Automation, ETFA 2018. vol. 2018-September, 8502622, Institute of Electrical and Electronics Engineers Inc., pp. 1365-1371, 23rd IEEE International Conference on Emerging Technologies and Factory Automation, ETFA 2018, Torino, Italy, 18/9/4. https://doi.org/10.1109/ETFA.2018.8502622
Yuyama K, Nishi H. Proposal of Feature Value Selection Method for Time-Critical Learning. In Proceedings - 2018 IEEE 23rd International Conference on Emerging Technologies and Factory Automation, ETFA 2018. Vol. 2018-September. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1365-1371. 8502622 https://doi.org/10.1109/ETFA.2018.8502622
Yuyama, Kanami ; Nishi, Hiroaki. / Proposal of Feature Value Selection Method for Time-Critical Learning. Proceedings - 2018 IEEE 23rd International Conference on Emerging Technologies and Factory Automation, ETFA 2018. Vol. 2018-September Institute of Electrical and Electronics Engineers Inc., 2018. pp. 1365-1371
@inproceedings{2b044adfc74f4e238871050203944cb3,
title = "Proposal of Feature Value Selection Method for Time-Critical Learning",
abstract = "The development of IoT has led to the creation of a data-enriched environment that enables data gathering by using distributed sensors and terminals. However, in this environment, the cost of data analysis has increased. Machine learning has gained attention for reducing the cost because enabling automatic data analysis, as well as multidimensional data, is expected. However, for enormous data, such as Big Data, we still have to pay costs. Therefore, selecting feature values when using machine learning technology is essential, especially as inputs of a classifier. Selecting the feature values increases its estimation accuracy. Moreover, the time cost, as well as calculation cost, needs consideration for the actual time-critical use of machine learning, especially in its learning process. Therefore, in this study, we proposed an algorithm that selected suitable feature values in required time. The proposed method consists of two stages: stepwise input selection stage using ANOVA and feature deletion stage according to the contribution rate of the features to estimate accuracy. These selection and deletion processes continue until the required processing time. We confirmed the efficiency of the proposed method by using an environment of a crystallization process in a factory and a household's occupancy estimation. A comparison with the original stepwise input method proved that the proposed method improved the estimation accuracy by 2{\%} and 5{\%} in the estimation of the substance amount of the crystallization process and household's occupancy, respectively.",
keywords = "ANOVA, Feature value selection, k-NN, real-time feature selection, time-constraint machine learning",
author = "Kanami Yuyama and Hiroaki Nishi",
year = "2018",
month = "10",
day = "22",
doi = "10.1109/ETFA.2018.8502622",
language = "English",
volume = "2018-September",
pages = "1365--1371",
booktitle = "Proceedings - 2018 IEEE 23rd International Conference on Emerging Technologies and Factory Automation, ETFA 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Proposal of Feature Value Selection Method for Time-Critical Learning

AU - Yuyama, Kanami

AU - Nishi, Hiroaki

PY - 2018/10/22

Y1 - 2018/10/22

N2 - The development of IoT has led to the creation of a data-enriched environment that enables data gathering by using distributed sensors and terminals. However, in this environment, the cost of data analysis has increased. Machine learning has gained attention for reducing the cost because enabling automatic data analysis, as well as multidimensional data, is expected. However, for enormous data, such as Big Data, we still have to pay costs. Therefore, selecting feature values when using machine learning technology is essential, especially as inputs of a classifier. Selecting the feature values increases its estimation accuracy. Moreover, the time cost, as well as calculation cost, needs consideration for the actual time-critical use of machine learning, especially in its learning process. Therefore, in this study, we proposed an algorithm that selected suitable feature values in required time. The proposed method consists of two stages: stepwise input selection stage using ANOVA and feature deletion stage according to the contribution rate of the features to estimate accuracy. These selection and deletion processes continue until the required processing time. We confirmed the efficiency of the proposed method by using an environment of a crystallization process in a factory and a household's occupancy estimation. A comparison with the original stepwise input method proved that the proposed method improved the estimation accuracy by 2% and 5% in the estimation of the substance amount of the crystallization process and household's occupancy, respectively.

AB - The development of IoT has led to the creation of a data-enriched environment that enables data gathering by using distributed sensors and terminals. However, in this environment, the cost of data analysis has increased. Machine learning has gained attention for reducing the cost because enabling automatic data analysis, as well as multidimensional data, is expected. However, for enormous data, such as Big Data, we still have to pay costs. Therefore, selecting feature values when using machine learning technology is essential, especially as inputs of a classifier. Selecting the feature values increases its estimation accuracy. Moreover, the time cost, as well as calculation cost, needs consideration for the actual time-critical use of machine learning, especially in its learning process. Therefore, in this study, we proposed an algorithm that selected suitable feature values in required time. The proposed method consists of two stages: stepwise input selection stage using ANOVA and feature deletion stage according to the contribution rate of the features to estimate accuracy. These selection and deletion processes continue until the required processing time. We confirmed the efficiency of the proposed method by using an environment of a crystallization process in a factory and a household's occupancy estimation. A comparison with the original stepwise input method proved that the proposed method improved the estimation accuracy by 2% and 5% in the estimation of the substance amount of the crystallization process and household's occupancy, respectively.

KW - ANOVA

KW - Feature value selection

KW - k-NN

KW - real-time feature selection

KW - time-constraint machine learning

UR - http://www.scopus.com/inward/record.url?scp=85057288141&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057288141&partnerID=8YFLogxK

U2 - 10.1109/ETFA.2018.8502622

DO - 10.1109/ETFA.2018.8502622

M3 - Conference contribution

AN - SCOPUS:85057288141

VL - 2018-September

SP - 1365

EP - 1371

BT - Proceedings - 2018 IEEE 23rd International Conference on Emerging Technologies and Factory Automation, ETFA 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -