SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection

Md Rashed-Al-Mahfuz; Md Robiul Hoque; Bimal Kumar Pramanik; Md Ekramul Hamid; Mohammad Ali Moni

doi:10.1109/IC4ME247184.2019.9036529

SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection

Md Rashed-Al-Mahfuz
, Md Robiul Hoque
, Bimal Kumar Pramanik
, Md Ekramul Hamid
, Mohammad Ali Moni

Artificial Intelligence and Cyber Futures Institute

University of Rajshahi
Islamic University, Kushtia
The University of Sydney

Research output: Book chapter/Published conference paper › Conference paper › peer-review

3 Citations (Scopus)

Abstract

Falls are a dangerous problem for people of all ages. Thus, accurate falls detection with minimized false alarms is very important. This study aims to detect falls and activities of daily living (ADLs) using acceleration data and to introduce an effective feature selection criterion to reduce the false positive rate of the falls detection systems. The falls detection system in this study consists of three stages. At the first stage, we have harnessed some feature extraction techniques to have discriminative features from the acceleration data. Then we have used feature selection criterions to select effective features in the detection task. At the last stage, we used Support Vector Machine (SVM) to classify the selected features in falls and ADLs. We have used raw acceleration data and extracted all the features. Then we selected features based on the Minimum Redundancy Maximum Relevance (MRMR) criterion and Double Input Symmetrical Relevance (DISR) in the fall detection experiment. We have found that the DISR feature selection criterion is more effective in acceleration based fall detection system. The results show 100% classification accuracy and zero false positive rates in fall detection for the DISR based selected features.

Original language	English
Title of host publication	5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering
Subtitle of host publication	IC4ME2 2019
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Number of pages	5
ISBN (Electronic)	9781728130606
ISBN (Print)	9781728130613 (Print on demand)
DOIs	https://doi.org/10.1109/IC4ME247184.2019.9036529
Publication status	Published - Jul 2019
Event	5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering: IC4ME2 2019 - University of Rajshahi, Rajshahi, Bangladesh Duration: 11 Jul 2019 → 12 Jul 2019 https://web.archive.org/web/20191028191905/http://dept.ru.ac.bd/ic4me2/2019/ (Conference website) https://web.archive.org/web/20191030061438/http://dept.ru.ac.bd/ic4me2/2019/?page_id=115 (Call for papers)

Publication series

Name	5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2 2019

Conference

Conference	5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering
Country/Territory	Bangladesh
City	Rajshahi
Period	11/07/19 → 12/07/19
Other	The International Conference on Computer, Communication, Chemical, Material and Electronic Engineering (IC4ME2-2019) will be held from July 11~12, 2019 at University of Rajshahi in Bangladesh. This conference is a sequel of our first conference ICMEIE-2015. The conference will gather world-class researchers, engineers and educators engaged in the fields of Materials, Electronics, Chemical and Information Engineering to meet and present their latest activities. The main theme of this conference is Networking and Collaboration.
Internet address	https://web.archive.org/web/20191028191905/http://dept.ru.ac.bd/ic4me2/2019/ (Conference website) https://web.archive.org/web/20191030061438/http://dept.ru.ac.bd/ic4me2/2019/?page_id=115 (Call for papers)

Access to Document

10.1109/IC4ME247184.2019.9036529

Cite this

Rashed-Al-Mahfuz, M., Hoque, M. R., Pramanik, B. K., Hamid, M. E., & Moni, M. A. (2019). SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection. In 5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering: IC4ME2 2019 Article 9036529 (5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2 2019). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/IC4ME247184.2019.9036529

Rashed-Al-Mahfuz, Md ; Hoque, Md Robiul ; Pramanik, Bimal Kumar et al. / SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection. 5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering: IC4ME2 2019. IEEE, Institute of Electrical and Electronics Engineers, 2019. (5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2 2019).

@inproceedings{05839b9f5cb0407a97540d068257f19b,

title = "SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection",

abstract = "Falls are a dangerous problem for people of all ages. Thus, accurate falls detection with minimized false alarms is very important. This study aims to detect falls and activities of daily living (ADLs) using acceleration data and to introduce an effective feature selection criterion to reduce the false positive rate of the falls detection systems. The falls detection system in this study consists of three stages. At the first stage, we have harnessed some feature extraction techniques to have discriminative features from the acceleration data. Then we have used feature selection criterions to select effective features in the detection task. At the last stage, we used Support Vector Machine (SVM) to classify the selected features in falls and ADLs. We have used raw acceleration data and extracted all the features. Then we selected features based on the Minimum Redundancy Maximum Relevance (MRMR) criterion and Double Input Symmetrical Relevance (DISR) in the fall detection experiment. We have found that the DISR feature selection criterion is more effective in acceleration based fall detection system. The results show 100\% classification accuracy and zero false positive rates in fall detection for the DISR based selected features.",

keywords = "acceleration data, DISR criterion, fall detection, feature extraction, MRMR criterion, SVM",

author = "Md Rashed-Al-Mahfuz and Hoque, \{Md Robiul\} and Pramanik, \{Bimal Kumar\} and Hamid, \{Md Ekramul\} and Moni, \{Mohammad Ali\}",

year = "2019",

month = jul,

doi = "10.1109/IC4ME247184.2019.9036529",

language = "English",

isbn = "9781728130613 (Print on demand)",

series = "5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2 2019",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

booktitle = "5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering",

address = "United States",

note = "5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering : IC4ME2 2019 ; Conference date: 11-07-2019 Through 12-07-2019",

url = "https://web.archive.org/web/20191028191905/http://dept.ru.ac.bd/ic4me2/2019/, https://web.archive.org/web/20191030061438/http://dept.ru.ac.bd/ic4me2/2019/?page\_id=115",

}

Rashed-Al-Mahfuz, M, Hoque, MR, Pramanik, BK, Hamid, ME & Moni, MA 2019, SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection. in 5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering: IC4ME2 2019., 9036529, 5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2 2019, IEEE, Institute of Electrical and Electronics Engineers, 5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, Rajshahi, Bangladesh, 11/07/19. https://doi.org/10.1109/IC4ME247184.2019.9036529

SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection. / Rashed-Al-Mahfuz, Md; Hoque, Md Robiul; Pramanik, Bimal Kumar et al.
5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering: IC4ME2 2019. IEEE, Institute of Electrical and Electronics Engineers, 2019. 9036529 (5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2 2019).

Research output: Book chapter/Published conference paper › Conference paper › peer-review

TY - GEN

T1 - SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection

AU - Rashed-Al-Mahfuz, Md

AU - Hoque, Md Robiul

AU - Pramanik, Bimal Kumar

AU - Hamid, Md Ekramul

AU - Moni, Mohammad Ali

PY - 2019/7

Y1 - 2019/7

N2 - Falls are a dangerous problem for people of all ages. Thus, accurate falls detection with minimized false alarms is very important. This study aims to detect falls and activities of daily living (ADLs) using acceleration data and to introduce an effective feature selection criterion to reduce the false positive rate of the falls detection systems. The falls detection system in this study consists of three stages. At the first stage, we have harnessed some feature extraction techniques to have discriminative features from the acceleration data. Then we have used feature selection criterions to select effective features in the detection task. At the last stage, we used Support Vector Machine (SVM) to classify the selected features in falls and ADLs. We have used raw acceleration data and extracted all the features. Then we selected features based on the Minimum Redundancy Maximum Relevance (MRMR) criterion and Double Input Symmetrical Relevance (DISR) in the fall detection experiment. We have found that the DISR feature selection criterion is more effective in acceleration based fall detection system. The results show 100% classification accuracy and zero false positive rates in fall detection for the DISR based selected features.

AB - Falls are a dangerous problem for people of all ages. Thus, accurate falls detection with minimized false alarms is very important. This study aims to detect falls and activities of daily living (ADLs) using acceleration data and to introduce an effective feature selection criterion to reduce the false positive rate of the falls detection systems. The falls detection system in this study consists of three stages. At the first stage, we have harnessed some feature extraction techniques to have discriminative features from the acceleration data. Then we have used feature selection criterions to select effective features in the detection task. At the last stage, we used Support Vector Machine (SVM) to classify the selected features in falls and ADLs. We have used raw acceleration data and extracted all the features. Then we selected features based on the Minimum Redundancy Maximum Relevance (MRMR) criterion and Double Input Symmetrical Relevance (DISR) in the fall detection experiment. We have found that the DISR feature selection criterion is more effective in acceleration based fall detection system. The results show 100% classification accuracy and zero false positive rates in fall detection for the DISR based selected features.

KW - acceleration data

KW - DISR criterion

KW - fall detection

KW - feature extraction

KW - MRMR criterion

KW - SVM

UR - https://www.scopus.com/pages/publications/85082991023

UR - https://www.scopus.com/pages/publications/85082991023#tab=citedBy

U2 - 10.1109/IC4ME247184.2019.9036529

DO - 10.1109/IC4ME247184.2019.9036529

M3 - Conference paper

AN - SCOPUS:85082991023

SN - 9781728130613 (Print on demand)

T3 - 5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2 2019

BT - 5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - 5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering

Y2 - 11 July 2019 through 12 July 2019

ER -

Rashed-Al-Mahfuz M, Hoque MR, Pramanik BK, Hamid ME, Moni MA. SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection. In 5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering: IC4ME2 2019. IEEE, Institute of Electrical and Electronics Engineers. 2019. 9036529. (5th International Conference on Computer, Communication, Chemical, Materials and Electronic Engineering, IC4ME2 2019). doi: 10.1109/IC4ME247184.2019.9036529

SVM model for feature selection to increase accuracy and reduce false positive rate in falls detection

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this