No Cover Image

Journal article 111 views 3 downloads

Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data / Rory P. Wilson; Mark D. Holton; Agustina di Virgilio; Hannah Williams; Emily L. C. Shepard; Sergio Lambertucci; Flavio Quintana; Juan E. Sala; Bharathan Balaji; Eun Sun Lee; Mani Srivastava; D. Michael Scantlebury; Carlos M. Duarte

Methods in Ecology and Evolution, Volume: 9, Issue: 11, Pages: 2206 - 2215

Swansea University Author: Holton, Mark

Abstract

1. The development of multisensor animal‐attached tags, recording data at high frequencies, has enormous potential in allowing us to define animal behaviour.2. The high volumes of data, are pushing us towards machine‐learning as a powerful option for distilling out behaviours. However, with increasi...

Full description

Published in: Methods in Ecology and Evolution
ISSN: 2041210X
Published: 2018
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa43784
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2018-09-12T12:59:43Z
last_indexed 2019-06-05T10:53:44Z
id cronfa43784
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2019-05-30T12:13:00Z</datestamp><bib-version>v2</bib-version><id>43784</id><entry>2018-09-12</entry><title>Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data</title><alternativeTitle></alternativeTitle><author>Mark Holton</author><firstname>Mark</firstname><surname>Holton</surname><active>true</active><ORCID>0000-0001-8834-3283</ORCID><ethesisStudent>false</ethesisStudent><sid>0e1d89d0cc934a740dcd0a873aed178e</sid><email>a7f54302246ae22ec0e2aeff1ecca1c0</email><emailaddr>aN+2TR1WwHRUjd7rdvLOvsjwe531u+mO/3IG3xe5jMg=</emailaddr><date>2018-09-12</date><deptcode>SCS</deptcode><abstract>1. The development of multisensor animal&#x2010;attached tags, recording data at high frequencies, has enormous potential in allowing us to define animal behaviour.2. The high volumes of data, are pushing us towards machine&#x2010;learning as a powerful option for distilling out behaviours. However, with increasing parallel lines of data, systems become more likely to become processor limited and thereby take appreciable amounts of time to resolve behaviours.3. We suggest a Boolean approach whereby critical changes in recorded parameters are used as sequential templates with defined flexibility (in both time and degree) to determine individual behavioural elements within a behavioural sequence that, together, makes up a single, defined behaviour.4. We tested this approach, and compared it to a suite of other behavioural identification methods, on a number of behaviours from tag&#x2010;equipped animals; sheep grazing, penguins walking, cheetah stalking prey and condors thermalling.5. Overall behaviour recognition using our new approach was better than most other methods due to; (1) its ability to deal with behavioural variation and (2) the speed with which the task was completed because extraneous data are avoided in the process.6. We suggest that this approach is a promising way forward in an increasingly data&#x2010;rich environment and that workers sharing algorithms can provide a powerful library for the benefit of all involved in such work.</abstract><type>Journal article</type><journal>Methods in Ecology and Evolution</journal><volume>9</volume><journalNumber>11</journalNumber><paginationStart>2206</paginationStart><paginationEnd>2215</paginationEnd><publisher></publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>2041210X</issnPrint><issnElectronic/><keywords>accelerometer, behaviour, behaviour identification, bioinformatics, software</keywords><publishedDay>0</publishedDay><publishedMonth>0</publishedMonth><publishedYear>2018</publishedYear><publishedDate>2018-01-01</publishedDate><doi>10.1111/2041-210X.13069</doi><url></url><notes>Data deposited in the Dryad Digital Repository https://doi.org/10.5061/dryad.56mh682 (Wilson et al., 2018)</notes><college>College of Science</college><department>Computer Science</department><CollegeCode>CSCI</CollegeCode><DepartmentCode>SCS</DepartmentCode><institution/><researchGroup>Swansea Lab for Animal Movement</researchGroup><supervisor/><sponsorsfunders/><grantnumber/><degreelevel/><degreename>None</degreename><lastEdited>2019-05-30T12:13:00Z</lastEdited><Created>2018-09-12T09:32:19Z</Created><path><level id="1">College of Science</level><level id="2">Computer Science</level></path><authors><author><firstname>Rory P.</firstname><surname>Wilson</surname><orcid/><order>1</order></author><author><firstname>Mark D.</firstname><surname>Holton</surname><orcid/><order>2</order></author><author><firstname>Agustina</firstname><surname>di Virgilio</surname><orcid/><order>3</order></author><author><firstname>Hannah</firstname><surname>Williams</surname><orcid/><order>4</order></author><author><firstname>Emily L. C.</firstname><surname>Shepard</surname><orcid/><order>5</order></author><author><firstname>Sergio</firstname><surname>Lambertucci</surname><orcid/><order>6</order></author><author><firstname>Flavio</firstname><surname>Quintana</surname><orcid/><order>7</order></author><author><firstname>Juan E.</firstname><surname>Sala</surname><orcid/><order>8</order></author><author><firstname>Bharathan</firstname><surname>Balaji</surname><orcid/><order>9</order></author><author><firstname>Eun Sun</firstname><surname>Lee</surname><orcid/><order>10</order></author><author><firstname>Mani</firstname><surname>Srivastava</surname><orcid/><order>11</order></author><author><firstname>D. Michael</firstname><surname>Scantlebury</surname><orcid/><order>12</order></author><author><firstname>Carlos M.</firstname><surname>Duarte</surname><orcid/><order>13</order></author></authors><documents><document><filename>0043784-12092018094001.pdf</filename><originalFilename>wilson_supp2018.pdf</originalFilename><uploaded>2018-09-12T09:40:01Z</uploaded><type>Output</type><contentLength>611662</contentLength><contentType>application/pdf</contentType><version>NA</version><cronfaStatus>true</cronfaStatus><action>Updated Copyright</action><actionDate>28/09/2018</actionDate><embargoDate>2019-07-24T00:00:00</embargoDate><documentNotes>Supplementary Material</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language></document><document><filename>0043784-28092018134210.pdf</filename><originalFilename>43784.pdf</originalFilename><uploaded>2018-09-28T13:42:10Z</uploaded><type>Output</type><contentLength>427910</contentLength><contentType>application/pdf</contentType><version>AM</version><cronfaStatus>true</cronfaStatus><action>Published to Cronfa</action><actionDate>28/09/2018</actionDate><embargoDate>2019-07-24T00:00:00</embargoDate><documentNotes/><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents></rfc1807>
spelling 2019-05-30T12:13:00Z v2 43784 2018-09-12 Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data Mark Holton Mark Holton true 0000-0001-8834-3283 false 0e1d89d0cc934a740dcd0a873aed178e a7f54302246ae22ec0e2aeff1ecca1c0 aN+2TR1WwHRUjd7rdvLOvsjwe531u+mO/3IG3xe5jMg= 2018-09-12 SCS 1. The development of multisensor animal‐attached tags, recording data at high frequencies, has enormous potential in allowing us to define animal behaviour.2. The high volumes of data, are pushing us towards machine‐learning as a powerful option for distilling out behaviours. However, with increasing parallel lines of data, systems become more likely to become processor limited and thereby take appreciable amounts of time to resolve behaviours.3. We suggest a Boolean approach whereby critical changes in recorded parameters are used as sequential templates with defined flexibility (in both time and degree) to determine individual behavioural elements within a behavioural sequence that, together, makes up a single, defined behaviour.4. We tested this approach, and compared it to a suite of other behavioural identification methods, on a number of behaviours from tag‐equipped animals; sheep grazing, penguins walking, cheetah stalking prey and condors thermalling.5. Overall behaviour recognition using our new approach was better than most other methods due to; (1) its ability to deal with behavioural variation and (2) the speed with which the task was completed because extraneous data are avoided in the process.6. We suggest that this approach is a promising way forward in an increasingly data‐rich environment and that workers sharing algorithms can provide a powerful library for the benefit of all involved in such work. Journal article Methods in Ecology and Evolution 9 11 2206 2215 2041210X accelerometer, behaviour, behaviour identification, bioinformatics, software 0 0 2018 2018-01-01 10.1111/2041-210X.13069 Data deposited in the Dryad Digital Repository https://doi.org/10.5061/dryad.56mh682 (Wilson et al., 2018) College of Science Computer Science CSCI SCS Swansea Lab for Animal Movement None 2019-05-30T12:13:00Z 2018-09-12T09:32:19Z College of Science Computer Science Rory P. Wilson 1 Mark D. Holton 2 Agustina di Virgilio 3 Hannah Williams 4 Emily L. C. Shepard 5 Sergio Lambertucci 6 Flavio Quintana 7 Juan E. Sala 8 Bharathan Balaji 9 Eun Sun Lee 10 Mani Srivastava 11 D. Michael Scantlebury 12 Carlos M. Duarte 13 0043784-12092018094001.pdf wilson_supp2018.pdf 2018-09-12T09:40:01Z Output 611662 application/pdf NA true Updated Copyright 28/09/2018 2019-07-24T00:00:00 Supplementary Material true eng 0043784-28092018134210.pdf 43784.pdf 2018-09-28T13:42:10Z Output 427910 application/pdf AM true Published to Cronfa 28/09/2018 2019-07-24T00:00:00 true eng
title Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data
spellingShingle Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data
Holton, Mark
title_short Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data
title_full Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data
title_fullStr Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data
title_full_unstemmed Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data
title_sort Give the machine a hand: A Boolean time-based decision-tree template for rapidly finding animal behaviours in multisensor data
author_id_str_mv 0e1d89d0cc934a740dcd0a873aed178e
author_id_fullname_str_mv 0e1d89d0cc934a740dcd0a873aed178e_***_Holton, Mark
author Holton, Mark
author2 Rory P. Wilson
Mark D. Holton
Agustina di Virgilio
Hannah Williams
Emily L. C. Shepard
Sergio Lambertucci
Flavio Quintana
Juan E. Sala
Bharathan Balaji
Eun Sun Lee
Mani Srivastava
D. Michael Scantlebury
Carlos M. Duarte
format Journal article
container_title Methods in Ecology and Evolution
container_volume 9
container_issue 11
container_start_page 2206
publishDate 2018
institution Swansea University
issn 2041210X
doi_str_mv 10.1111/2041-210X.13069
college_str College of Science
hierarchytype
hierarchy_top_id collegeofscience
hierarchy_top_title College of Science
hierarchy_parent_id collegeofscience
hierarchy_parent_title College of Science
department_str Computer Science{{{_:::_}}}College of Science{{{_:::_}}}Computer Science
document_store_str 1
active_str 1
researchgroup_str Swansea Lab for Animal Movement
description 1. The development of multisensor animal‐attached tags, recording data at high frequencies, has enormous potential in allowing us to define animal behaviour.2. The high volumes of data, are pushing us towards machine‐learning as a powerful option for distilling out behaviours. However, with increasing parallel lines of data, systems become more likely to become processor limited and thereby take appreciable amounts of time to resolve behaviours.3. We suggest a Boolean approach whereby critical changes in recorded parameters are used as sequential templates with defined flexibility (in both time and degree) to determine individual behavioural elements within a behavioural sequence that, together, makes up a single, defined behaviour.4. We tested this approach, and compared it to a suite of other behavioural identification methods, on a number of behaviours from tag‐equipped animals; sheep grazing, penguins walking, cheetah stalking prey and condors thermalling.5. Overall behaviour recognition using our new approach was better than most other methods due to; (1) its ability to deal with behavioural variation and (2) the speed with which the task was completed because extraneous data are avoided in the process.6. We suggest that this approach is a promising way forward in an increasingly data‐rich environment and that workers sharing algorithms can provide a powerful library for the benefit of all involved in such work.
published_date 2018-01-01T05:12:23Z
_version_ 1647521712872357888
score 10.859247