Understanding animal behavior using their trajectories: A case study of gender specific trajectory trends

Ilya Ardakani, Koichi Hashimoto, Ken Yoda

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Generally, behavior and movement could be closely attributed with each other. In other words, in most cases, behavior expressed as a set of specific movement patterns in time. These movement traces or trajectories representing behavior could provide a window into the underlying state of the subjects. In this study, analogies drawn between text and trajectories which allowed us to employ sentiment analysis and topic model methods to analyze trajectories. It is assumed that trajectories consist of key points which are commonly and frequently traversed. It is proposed that analogously in trajectory analysis, key points frequency would encapsulate information about the subject or the key points in trajectory generated by latent distribution which attributed to certain behavior or specific group of subjects with similar behavioral features. To test this hypothesis, an experiment was conducted which examines the influence of gender in composition of key points in birds’ trajectories logged from a seabird species called Streaked Shearwater Calonectris leucomelas. It was shown that genders have specific distribution over the key points. Therefore, key points membership in trajectory could be attributed to a specific gender and even a simple classifier would provide information about the gender of the subject simply by observing the trajectory’s key points. It was concluded that like text, trajectories composed of smaller elements which could be associated to a specific latent state. Learning or exploiting these associations revealed essential information about identity and behavior of the subject of observation.

Original languageEnglish
Title of host publicationDistributed, Ambient and Pervasive Interactions
Subtitle of host publicationTechnologies and Contexts - 6th International Conference, DAPI 2018, Held as Part of HCI International 2018, Proceedings
EditorsShin’ichi Konomi, Norbert Streitz
PublisherSpringer Verlag
Pages3-22
Number of pages20
ISBN (Print)9783319911304
DOIs
Publication statusPublished - 2018
Event6th International Conference on Distributed, Ambient and Pervasive Interactions, DAPI 2018 Held as Part of HCI International 2018 - Las Vegas, United States
Duration: 2018 Jul 152018 Jul 20

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10922 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other6th International Conference on Distributed, Ambient and Pervasive Interactions, DAPI 2018 Held as Part of HCI International 2018
CountryUnited States
CityLas Vegas
Period18/7/1518/7/20

Keywords

  • Animal behavior
  • Animal movement
  • Trajectory mining

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Understanding animal behavior using their trajectories: A case study of gender specific trajectory trends'. Together they form a unique fingerprint.

Cite this