A benchmark for computational analysis of animal behavior, using animal-borne tags
Abstract Background Animal-borne sensors (‘bio-loggers’) can record a suite of kinematic and environmental data, which are used to elucidate animal ecophysiology and improve conservation efforts. Machine learning techniques are used for interpreting the large amounts of data recorded by bio-loggers,...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2024-12-01
|
| Series: | Movement Ecology |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40462-024-00511-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850100516776837120 |
|---|---|
| author | Benjamin Hoffman Maddie Cusimano Vittorio Baglione Daniela Canestrari Damien Chevallier Dominic L. DeSantis Lorène Jeantet Monique A. Ladds Takuya Maekawa Vicente Mata-Silva Víctor Moreno-González Anthony M. Pagano Eva Trapote Outi Vainio Antti Vehkaoja Ken Yoda Katherine Zacarian Ari Friedlaender |
| author_facet | Benjamin Hoffman Maddie Cusimano Vittorio Baglione Daniela Canestrari Damien Chevallier Dominic L. DeSantis Lorène Jeantet Monique A. Ladds Takuya Maekawa Vicente Mata-Silva Víctor Moreno-González Anthony M. Pagano Eva Trapote Outi Vainio Antti Vehkaoja Ken Yoda Katherine Zacarian Ari Friedlaender |
| author_sort | Benjamin Hoffman |
| collection | DOAJ |
| description | Abstract Background Animal-borne sensors (‘bio-loggers’) can record a suite of kinematic and environmental data, which are used to elucidate animal ecophysiology and improve conservation efforts. Machine learning techniques are used for interpreting the large amounts of data recorded by bio-loggers, but there exists no common framework for comparing the different machine learning techniques in this domain. This makes it difficult to, for example, identify patterns in what works well for machine learning-based analysis of bio-logger data. It also makes it difficult to evaluate the effectiveness of novel methods developed by the machine learning community. Methods To address this, we present the Bio-logger Ethogram Benchmark (BEBE), a collection of datasets with behavioral annotations, as well as a modeling task and evaluation metrics. BEBE is to date the largest, most taxonomically diverse, publicly available benchmark of this type, and includes 1654 h of data collected from 149 individuals across nine taxa. Using BEBE, we compare the performance of deep and classical machine learning methods for identifying animal behaviors based on bio-logger data. As an example usage of BEBE, we test an approach based on self-supervised learning. To apply this approach to animal behavior classification, we adapt a deep neural network pre-trained with 700,000 h of data collected from human wrist-worn accelerometers. Results We find that deep neural networks out-perform the classical machine learning methods we tested across all nine datasets in BEBE. We additionally find that the approach based on self-supervised learning out-performs the alternatives we tested, especially in settings when there is a low amount of training data available. Conclusions In light of these results, we are able to make concrete suggestions for designing studies that rely on machine learning to infer behavior from bio-logger data. Therefore, we expect that BEBE will be useful for making similar suggestions in the future, as additional hypotheses about machine learning techniques are tested. Datasets, models, and evaluation code are made publicly available at https://github.com/earthspecies/BEBE , to enable community use of BEBE. |
| format | Article |
| id | doaj-art-597868eb1dc44df98d4cbf34e245dda7 |
| institution | DOAJ |
| issn | 2051-3933 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | BMC |
| record_format | Article |
| series | Movement Ecology |
| spelling | doaj-art-597868eb1dc44df98d4cbf34e245dda72025-08-20T02:40:17ZengBMCMovement Ecology2051-39332024-12-0112112510.1186/s40462-024-00511-8A benchmark for computational analysis of animal behavior, using animal-borne tagsBenjamin Hoffman0Maddie Cusimano1Vittorio Baglione2Daniela Canestrari3Damien Chevallier4Dominic L. DeSantis5Lorène Jeantet6Monique A. Ladds7Takuya Maekawa8Vicente Mata-Silva9Víctor Moreno-González10Anthony M. Pagano11Eva Trapote12Outi Vainio13Antti Vehkaoja14Ken Yoda15Katherine Zacarian16Ari Friedlaender17Earth Species ProjectEarth Species ProjectUniversity de LeónUniversity de LeónCNRS BoreaGeorgia College & State UniversityAfrican Institute for Mathematical Sciences, University of StellenboschDepartment of ConservationOsaka UniversityUniversity Texas El PasoUniversity de LeónUS Geological SurveyUniversity de LeónUniversity of HelsinkiTampere UniversityNagoya UniversityEarth Species ProjectUniversity of California Santa CruzAbstract Background Animal-borne sensors (‘bio-loggers’) can record a suite of kinematic and environmental data, which are used to elucidate animal ecophysiology and improve conservation efforts. Machine learning techniques are used for interpreting the large amounts of data recorded by bio-loggers, but there exists no common framework for comparing the different machine learning techniques in this domain. This makes it difficult to, for example, identify patterns in what works well for machine learning-based analysis of bio-logger data. It also makes it difficult to evaluate the effectiveness of novel methods developed by the machine learning community. Methods To address this, we present the Bio-logger Ethogram Benchmark (BEBE), a collection of datasets with behavioral annotations, as well as a modeling task and evaluation metrics. BEBE is to date the largest, most taxonomically diverse, publicly available benchmark of this type, and includes 1654 h of data collected from 149 individuals across nine taxa. Using BEBE, we compare the performance of deep and classical machine learning methods for identifying animal behaviors based on bio-logger data. As an example usage of BEBE, we test an approach based on self-supervised learning. To apply this approach to animal behavior classification, we adapt a deep neural network pre-trained with 700,000 h of data collected from human wrist-worn accelerometers. Results We find that deep neural networks out-perform the classical machine learning methods we tested across all nine datasets in BEBE. We additionally find that the approach based on self-supervised learning out-performs the alternatives we tested, especially in settings when there is a low amount of training data available. Conclusions In light of these results, we are able to make concrete suggestions for designing studies that rely on machine learning to infer behavior from bio-logger data. Therefore, we expect that BEBE will be useful for making similar suggestions in the future, as additional hypotheses about machine learning techniques are tested. Datasets, models, and evaluation code are made publicly available at https://github.com/earthspecies/BEBE , to enable community use of BEBE.https://doi.org/10.1186/s40462-024-00511-8Machine learningBio-loggersAnimal behaviorAccelerometersTime seriesSelf-supervised Learning |
| spellingShingle | Benjamin Hoffman Maddie Cusimano Vittorio Baglione Daniela Canestrari Damien Chevallier Dominic L. DeSantis Lorène Jeantet Monique A. Ladds Takuya Maekawa Vicente Mata-Silva Víctor Moreno-González Anthony M. Pagano Eva Trapote Outi Vainio Antti Vehkaoja Ken Yoda Katherine Zacarian Ari Friedlaender A benchmark for computational analysis of animal behavior, using animal-borne tags Movement Ecology Machine learning Bio-loggers Animal behavior Accelerometers Time series Self-supervised Learning |
| title | A benchmark for computational analysis of animal behavior, using animal-borne tags |
| title_full | A benchmark for computational analysis of animal behavior, using animal-borne tags |
| title_fullStr | A benchmark for computational analysis of animal behavior, using animal-borne tags |
| title_full_unstemmed | A benchmark for computational analysis of animal behavior, using animal-borne tags |
| title_short | A benchmark for computational analysis of animal behavior, using animal-borne tags |
| title_sort | benchmark for computational analysis of animal behavior using animal borne tags |
| topic | Machine learning Bio-loggers Animal behavior Accelerometers Time series Self-supervised Learning |
| url | https://doi.org/10.1186/s40462-024-00511-8 |
| work_keys_str_mv | AT benjaminhoffman abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT maddiecusimano abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT vittoriobaglione abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT danielacanestrari abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT damienchevallier abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT dominicldesantis abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT lorenejeantet abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT moniquealadds abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT takuyamaekawa abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT vicentematasilva abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT victormorenogonzalez abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT anthonympagano abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT evatrapote abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT outivainio abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT anttivehkaoja abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT kenyoda abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT katherinezacarian abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT arifriedlaender abenchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT benjaminhoffman benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT maddiecusimano benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT vittoriobaglione benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT danielacanestrari benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT damienchevallier benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT dominicldesantis benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT lorenejeantet benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT moniquealadds benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT takuyamaekawa benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT vicentematasilva benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT victormorenogonzalez benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT anthonympagano benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT evatrapote benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT outivainio benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT anttivehkaoja benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT kenyoda benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT katherinezacarian benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags AT arifriedlaender benchmarkforcomputationalanalysisofanimalbehaviorusinganimalbornetags |