Data mining and knowledge discovery from physiological sensors

Research output: Book chapter/Published conference paperConference paper

Abstract

We present in this paper our method for discovering logic rules
from physiological data by applying a fusion of preprocessing and
data mining using decision trees. We have focused on four sensors
representative of wearable technology capabilities namely;
plethysmography, galvanic skin response, respiration, and body
temperature, sourced from a highly cited dataset in Affective
Computing. The method involved generating a number of datasets
from the physiological data and subsequently performing
classification using the C4.5 decision tree algorithm with a focus
on knowledge discovery. The findings of this research demonstrate
that preprocessing data into three classes with extreme boundaries
and a neutral class, as well as classifying the two classes without
the neutral class can produce high-quality rules. The discovered
knowledge in the form of the top 4 rules was mapped on the valence
and arousal emotional model. Finally, these rules are interpreted
with the aid of box and whisker plots in the context of the
underlying physiological processes.
Original languageEnglish
Title of host publicationIn Proc. of the 12th ACM International Conference on Pervasive Technologies Related to Assistive Environments (PETRA 2019), June5-7, 2019
Place of PublicationUnited States
PublisherACM
Pages468-474
Number of pages7
ISBN (Print)9781450362320
DOIs
Publication statusPublished - 05 Jun 2019
Event12th ACM International Conference on PErvasive Technologies Related to Assistive Environments : PETRA 2019 - Island of Rhodes, Greece
Duration: 05 Jun 201907 Jun 2019
http://www.petrae.org/docs/Petra19_program.pdf
http://petrae.org/past.html

Conference

Conference12th ACM International Conference on PErvasive Technologies Related to Assistive Environments
CountryGreece
CityIsland of Rhodes
Period05/06/1907/06/19
Internet address

Fingerprint

Decision trees
Data mining
Plethysmography
Sensors
Skin
Fusion reactions
Wearable technology

Cite this

Costadopoulos, N., Islam, Z., & Tien, D. (2019). Data mining and knowledge discovery from physiological sensors. In In Proc. of the 12th ACM International Conference on Pervasive Technologies Related to Assistive Environments (PETRA 2019), June5-7, 2019 (pp. 468-474). United States: ACM. https://doi.org/10.1145/3316782.3322771
Costadopoulos, Nectarios ; Islam, Zahid ; Tien, David. / Data mining and knowledge discovery from physiological sensors. In Proc. of the 12th ACM International Conference on Pervasive Technologies Related to Assistive Environments (PETRA 2019), June5-7, 2019. United States : ACM, 2019. pp. 468-474
@inproceedings{d80105521cc3497982116a3836d5be38,
title = "Data mining and knowledge discovery from physiological sensors",
abstract = "We present in this paper our method for discovering logic rulesfrom physiological data by applying a fusion of preprocessing anddata mining using decision trees. We have focused on four sensorsrepresentative of wearable technology capabilities namely;plethysmography, galvanic skin response, respiration, and bodytemperature, sourced from a highly cited dataset in AffectiveComputing. The method involved generating a number of datasetsfrom the physiological data and subsequently performingclassification using the C4.5 decision tree algorithm with a focuson knowledge discovery. The findings of this research demonstratethat preprocessing data into three classes with extreme boundariesand a neutral class, as well as classifying the two classes withoutthe neutral class can produce high-quality rules. The discoveredknowledge in the form of the top 4 rules was mapped on the valenceand arousal emotional model. Finally, these rules are interpretedwith the aid of box and whisker plots in the context of theunderlying physiological processes.",
author = "Nectarios Costadopoulos and Zahid Islam and David Tien",
year = "2019",
month = "6",
day = "5",
doi = "10.1145/3316782.3322771",
language = "English",
isbn = "9781450362320",
pages = "468--474",
booktitle = "In Proc. of the 12th ACM International Conference on Pervasive Technologies Related to Assistive Environments (PETRA 2019), June5-7, 2019",
publisher = "ACM",

}

Costadopoulos, N, Islam, Z & Tien, D 2019, Data mining and knowledge discovery from physiological sensors. in In Proc. of the 12th ACM International Conference on Pervasive Technologies Related to Assistive Environments (PETRA 2019), June5-7, 2019. ACM, United States, pp. 468-474, 12th ACM International Conference on PErvasive Technologies Related to Assistive Environments , Island of Rhodes, Greece, 05/06/19. https://doi.org/10.1145/3316782.3322771

Data mining and knowledge discovery from physiological sensors. / Costadopoulos, Nectarios; Islam, Zahid; Tien, David.

In Proc. of the 12th ACM International Conference on Pervasive Technologies Related to Assistive Environments (PETRA 2019), June5-7, 2019. United States : ACM, 2019. p. 468-474.

Research output: Book chapter/Published conference paperConference paper

TY - GEN

T1 - Data mining and knowledge discovery from physiological sensors

AU - Costadopoulos, Nectarios

AU - Islam, Zahid

AU - Tien, David

PY - 2019/6/5

Y1 - 2019/6/5

N2 - We present in this paper our method for discovering logic rulesfrom physiological data by applying a fusion of preprocessing anddata mining using decision trees. We have focused on four sensorsrepresentative of wearable technology capabilities namely;plethysmography, galvanic skin response, respiration, and bodytemperature, sourced from a highly cited dataset in AffectiveComputing. The method involved generating a number of datasetsfrom the physiological data and subsequently performingclassification using the C4.5 decision tree algorithm with a focuson knowledge discovery. The findings of this research demonstratethat preprocessing data into three classes with extreme boundariesand a neutral class, as well as classifying the two classes withoutthe neutral class can produce high-quality rules. The discoveredknowledge in the form of the top 4 rules was mapped on the valenceand arousal emotional model. Finally, these rules are interpretedwith the aid of box and whisker plots in the context of theunderlying physiological processes.

AB - We present in this paper our method for discovering logic rulesfrom physiological data by applying a fusion of preprocessing anddata mining using decision trees. We have focused on four sensorsrepresentative of wearable technology capabilities namely;plethysmography, galvanic skin response, respiration, and bodytemperature, sourced from a highly cited dataset in AffectiveComputing. The method involved generating a number of datasetsfrom the physiological data and subsequently performingclassification using the C4.5 decision tree algorithm with a focuson knowledge discovery. The findings of this research demonstratethat preprocessing data into three classes with extreme boundariesand a neutral class, as well as classifying the two classes withoutthe neutral class can produce high-quality rules. The discoveredknowledge in the form of the top 4 rules was mapped on the valenceand arousal emotional model. Finally, these rules are interpretedwith the aid of box and whisker plots in the context of theunderlying physiological processes.

U2 - 10.1145/3316782.3322771

DO - 10.1145/3316782.3322771

M3 - Conference paper

SN - 9781450362320

SP - 468

EP - 474

BT - In Proc. of the 12th ACM International Conference on Pervasive Technologies Related to Assistive Environments (PETRA 2019), June5-7, 2019

PB - ACM

CY - United States

ER -

Costadopoulos N, Islam Z, Tien D. Data mining and knowledge discovery from physiological sensors. In In Proc. of the 12th ACM International Conference on Pervasive Technologies Related to Assistive Environments (PETRA 2019), June5-7, 2019. United States: ACM. 2019. p. 468-474 https://doi.org/10.1145/3316782.3322771