How to treat mixed behavior segments in supervised machine learning of behavioural modes from inertial measurement data - treat mixed segments in training

Aug 5, 2026

Upload by Oliver

Supervised machine learning is a standard tool in behavioural ecology for inferring behavioural modes from inertial measurements recorded by bio-loggers. Model accuracy depends on design decisions, including whether mixed behaviour segments—segments containing more than one behaviour—are included or excluded from training data. A simulation study across four accelerometer datasets with behavioural observations tested this choice and evaluated robustness to segment length, sample size, and mixture level. Including mixed segments improves classification accuracy when mixed segments constitute over ~10% of test data, with species-specific exceptions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Abstract

Introduction

Bio-logging and supervised machine learning for behavioural modes

Mixed segments in training data: inclusion vs exclusion

Methods and datasets

Simulation design

Study species and accelerometer data

Results

Accuracy improvements with mixed segments in training

Robustness checks (segment length, sample size, mixture degree)

Exceptions and species-specific effects

Recommendations

When to include mixed segments during training

When to avoid excess inclusion (e.g., captivity-to-wild)

Resheff etal. Movement Ecology (2024) 12:44 [https://doi.org/10.1186/s40462-024-00485-7](https://doi.org/10.1186/s40462-024-00485-7) Movement Ecology METHODOLOGY Open Access How to treat mixed behavior segments in supervised machine learning of behavioural modes from inertial measurement data Yehezkel S. Resheff1*, Hanna M. Bensch2,3, Markus Zöttl2,3, Roi Harel4,5,6,7, Akiko Matsumoto‑Oda8, Margaret C. Crofoot4,5,6,7, Sara Gomez9, Luca Börger9 and Shay Rotics10,11 Abstract The application of supervised machine learning methods to identify behavioural modes from inertial measurements of bio‑loggers has become a standard tool in behavioural ecology. Several design choices can affect the accuracy of identifying the behavioural modes. One such choice is the inclusion or exclusion of segments consisting of more than a single behaviour (mixed segments) in the machine learning model training data. Currently, the common prac‑ tice is to ignore such segments during model training. In this paper we tested the hypothesis that including mixed segments in model training will improve accuracy, as the model would perform better in identifying them in the test data. We test this hypothesis using a series of data simulations on four datasets of accelerometer data coupled with behaviour observations, obtained from four study species (Damaraland mole‑rats, meerkats, olive baboons, polar bears) . Results show that when a substantial proportion of the test data are mixed behaviour segments (above~ 10%), including mixed segments in machine learning model training improves the accuracy of classification. These results were consistent across the four study species, and robust to changes in segment length, sample size, and degree of mixture within the mixed segments. However, we also find that in some cases (particularly in baboons) models trained with mixed segments show reduced accuracy in classifying test data containing only single behav‑ iour (pure) segments, compared to models trained without mixed segments. Based on these results, we recommend that when the classification model is expected to deal with a substantial proportion of mixed behaviour segments (> 10%), it is beneficial to include them in model training, otherwise, it is unnecessary but also not harmful. The excep‑ tion is when there is a basis to assume that the training data contains a higher rate of mixed segments than the actual (unobserved) data to be classified—such a situation may occur particularly when training data are collected in cap‑ tivity and used to classify data from the wild. In this case, excess inclusion of mixed segments in training data should probably be avoided. Keywords Body‑acceleration, Bio‑logging, Machine learning, Animal behaviour *Correspondence: Yehezkel S. Resheff [hezi. resheff@gmail.com](hezi. resheff@gmail.com) Full list of author information is available at the end of the article © The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit [http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver ()[. The Creative Commons Public Domain Dedication waiver (](http://creativecommons.org/li

FAQs

Why do mixed behaviour segments matter in supervised machine learning for behavioural modes?

Mixed segments contain more than one behaviour within the same time window, so the inertial signal reflects multiple behaviours. Whether these segments are included in training can affect how accurately the model classifies behavioural modes.

What was tested in the study regarding including mixed segments during training?

The study tested the hypothesis that including mixed segments in model training improves accuracy, because the trained model would better identify them in test data. The evaluation used simulation experiments on accelerometer datasets paired with behavioural observations.

When does including mixed segments in training improve classification accuracy?

When a substantial portion of the test data are mixed behaviour segments—above about 10%—including mixed segments in training improves classification accuracy. The benefit was consistent across four study species under multiple robustness checks.

Are there cases where mixed-segment training can reduce accuracy?

Yes. In some situations, particularly for baboons, models trained with mixed segments showed reduced accuracy when classifying test data that contained only single-behaviour (pure) segments compared with models trained without mixed segments.

Aug 5, 2026

Upload by Oliver