...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Marine biology >Development and application of a machine learning algorithm for classification of elasmobranch behaviour from accelerometry data
【24h】

Development and application of a machine learning algorithm for classification of elasmobranch behaviour from accelerometry data

机译：机器学习算法用于加速度计数据分类弹bra行为的开发与应用

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Discerning behaviours of free-ranging animals allows for quantification of their activity budget, providing important insight into ecology. Over recent years, accelerometers have been used to unveil the cryptic lives of animals. The increased ability of accelerometers to store large quantities of high resolution data has prompted a need for automated behavioural classification. We assessed the performance of several machine learning (ML) classifiers to discern five behaviours performed by accelerometer-equipped juvenile lemon sharks (Negaprion brevirostris) at Bimini, Bahamas (25 degrees 44'N, 79 degrees 16'W). The sharks were observed to exhibit chafing, burst swimming, headshaking, resting and swimming in a semi-captive environment and these observations were used to ground-truth data for ML training and testing. ML methods included logistic regression, an artificial neural network, two random forest models, a gradient boosting model and a voting ensemble (VE) model, which combined the predictions of all other (base) models to improve classifier performance. The macro-averaged F-measure, an indicator of classifier performance, showed that the VE model improved overall classification (F-measure 0.88) above the strongest base learner model, gradient boosting (0.86). To test whether the VE model provided biologically meaningful results when applied to accelerometer data obtained from wild sharks, we investigated headshaking behaviour, as a proxy for prey capture, in relation to the variables: time of day, tidal phase and season. All variables were significant in predicting prey capture, with predations most likely to occur during early evening and less frequently during the dry season and high tides. These findings support previous hypotheses from sporadic visual observations.
机译：识别自由放养动物的行为可以量化其活动预算，从而提供对生态学的重要见解。近年来，加速度计已用于揭示动物的神秘生活。加速度计存储大量高分辨率数据的能力增强，促使人们需要进行自动的行为分类。我们评估了几种机器学习（ML）分类器的性能，以识别由配备了加速度计的少年柠檬鲨（Negaprion brevirostris）在巴哈马群岛比米尼（25度44'N，79度16'W）所执行的五种行为。观察到鲨鱼在半圈养环境中表现出擦伤，突然游泳，摇头，休息和游泳，这些观察结果被用于对真相数据进行ML训练和测试。机器学习方法包括逻辑回归，人工神经网络，两个随机森林模型，梯度提升模型和投票合奏（VE）模型，它们结合了所有其他（基础）模型的预测以提高分类器性能。宏观平均F度量是分类器性能的指标，它表明VE模型比最强的基础学习器模型梯度提升（0.86）改善了整体分类（F度量0.88）。为了测试当将VE模型应用于从野生鲨鱼获得的加速度计数据时是否提供了生物学上有意义的结果，我们调查了有关变量（一天中的时间，潮汐阶段和季节）的摇头行为（作为捕获猎物的代理）。所有变量都在预测猎物捕获方面具有重要意义，捕食最有可能在傍晚发生，而在旱季和涨潮时更不常见。这些发现支持了零星视觉观察的先前假设。

著录项

  • 来源
    《Marine biology》 |2018年第4期|62.1-62.19|共19页
  • 作者

    Brewster L. R.; Dale J. J.; Guttridge T. L.; Gruber S. H.; Hansell A. C.; Elliott M.; Cowx I. G.; Whitney N. M.; Gleiss A. C.;

  • 作者单位

    Bimini Biol Field Stn Fdn, South Bimini, Bahamas;

    Stanford Univ, Dept Biol, Hopkins Marine Stn, Pacific Grove, CA 93950 USA;

    Bimini Biol Field Stn Fdn, South Bimini, Bahamas;

    Bimini Biol Field Stn Fdn, South Bimini, Bahamas;

    Univ Massachusetts Dartmouth, Sch Marine Sci & Technol, Dept Fisheries Oceanog, 836 South Rodney French Blvd, New Bedford, MA 02719 USA;

    Univ Hull, Inst Estuarine & Coastal Studies, Kingston Upon Hull HU6 7RX, N Humberside, England;

    Univ Hull, Hull Int Fisheries Inst, Kingston Upon Hull HU6 7RX, N Humberside, England;

    Cent Wharf, Anderson Cabot Ctr Ocean Life, New England Aquarium, Boston, MA 02110 USA;

    Murdoch Univ, Sch Vet & Life Sci, Ctr Fish & Fisheries Res, 90 South St, Perth, WA 6150, Australia;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号