{"doi":"10.1002/ece3.72722","title":"<scp>AccelerometerBehavior</scp> : R Package for Classifying Ungulate Behaviors Into Three States","abstract":"ABSTRACT Advances in technology provide new opportunities to study animal behavior at increasingly fine scales. GPS collars for wildlife are commonly equipped with accelerometers, which record fine‐scale movements with relatively little energy demand, yet the data remain underutilized. We paired activity data with behavioral states from direct observations and developed random forest models to classify behaviors into ‘stationary’, ‘foraging’, and ‘traveling’ states for 3 ungulate species (bighorn sheep, Ovis canadensis moose, Alces alces , and mule deer, Odocoileus hemionus ). Our algorithm achieved an overall classification accuracy of ≥ 87% and an area under the receiver operating curve of ≥ 0.93 for all species. The mean class error rate was 15.65% (range 4.4%–26.8%). We also developed a general ‘ungulate’ model (classification accuracy of 90% and area under the receiver operating curve of 0.95) to be applied to species lacking observation data. We developed an R package, AccelerometerBehavior, that allows users to classify ungulate behavior using our models that were validated with observation data. Unlike other packages, AccelerometerBehavior allows users to apply existing models to new activity datasets without needing their own direct observations, which can be difficult to obtain. AccelerometerBehavior facilitates the use of activity data and expands its potential for understanding ungulate behavior. Additionally, for each species, we compared activity budgets developed using AccelerometerBehavior (5‐min resolution) with those developed from Hidden‐Markov models using GPS data (1‐h resolution). Activity budgets developed using AccelerometerBehavior estimated that animals spent substantially more time stationary and less time foraging than those developed from Hidden‐Markov models and GPS data, emphasizing the need to consider the method and resolution of data when remotely assessing animal behavior. AccelerometerBehavior, which is hosted on GitHub (https://github.com/STRankins/AccelerometerBehavior), is simple and allows for the expanded use of activity data that are continuously collected on GPS collars yet are underutilized to study ungulate behavior.","journal":"Ecology and Evolution","year":2025,"id":529604,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":2,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9329,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1408899,"name":"Seth T. Rankins","orcid":null,"position":1,"is_corresponding":false},{"id":1408900,"name":"Lindsay Millward","orcid":null,"position":2,"is_corresponding":false},{"id":1408901,"name":"Jack N. Gavin","orcid":null,"position":3,"is_corresponding":false},{"id":1408902,"name":"Daniel P. Thompson","orcid":null,"position":4,"is_corresponding":false},{"id":1408903,"name":"John A. Crouse","orcid":null,"position":5,"is_corresponding":false},{"id":1408904,"name":"Peach Van Wick","orcid":null,"position":6,"is_corresponding":false},{"id":1408905,"name":"Katie Anderson","orcid":null,"position":7,"is_corresponding":false},{"id":672422,"name":"Clinton W. Epps","orcid":"0000-0001-6577-1840","position":8,"is_corresponding":false},{"id":561771,"name":"Anna E. Jolles","orcid":"0000-0003-3074-9912","position":9,"is_corresponding":false},{"id":635982,"name":"Brianna R. Beechler","orcid":"0000-0002-9711-4340","position":10,"is_corresponding":false},{"id":1408906,"name":"Rebecca L. Levine","orcid":null,"position":11,"is_corresponding":false},{"id":1408907,"name":"Tayler N. LaSharr","orcid":null,"position":12,"is_corresponding":false},{"id":1408908,"name":"Brittany L. Wagler","orcid":null,"position":13,"is_corresponding":false},{"id":1408909,"name":"Rebekah T. Rafferty","orcid":null,"position":14,"is_corresponding":false},{"id":1408910,"name":"Alyson B. Courtemanch","orcid":null,"position":15,"is_corresponding":false},{"id":1408911,"name":"Tony W. Mong","orcid":null,"position":16,"is_corresponding":false},{"id":1245096,"name":"Kevin L. Monteith","orcid":"0000-0003-4834-5465","position":17,"is_corresponding":false},{"id":1408441,"name":"Rachel A. Smiley","orcid":"0000-0001-7230-508X","position":0,"is_corresponding":true}],"reference_count":21,"raw_metadata":null,"created_at":"2026-07-19T02:50:56.971987Z","pmid":"41426641","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}