東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Trajectory Segmentation for Behavior...

Salami, Mary Idera,

Linked to FindBook

Google Book

Amazon

博客來

Trajectory Segmentation for Behavioral Mode Detection in Animal Movement /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Trajectory Segmentation for Behavioral Mode Detection in Animal Movement // Mary Idera Salami.
Author:	Salami, Mary Idera,
Description:	1 electronic resource (69 pages)
Notes:	Source: Masters Abstracts International, Volume: 87-02.
Contained By:	Masters Abstracts International87-02.
Subject:	Geography. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31995282
ISBN:	9798291539200

Trajectory Segmentation for Behavioral Mode Detection in Animal Movement /
Salami, Mary Idera,

Trajectory Segmentation for Behavioral Mode Detection in Animal Movement /Mary Idera Salami. - 1 electronic resource (69 pages)

Source: Masters Abstracts International, Volume: 87-02.

Accurately segmenting animal movement trajectories into behavioral states is crucial for understanding ecological dynamics and advancing conservation strategies. Traditional segmentation methods often struggle when processing complex, noisy, high-resolution movement data. This research proposes a hybrid modeling framework that combines the probabilistic state inference capabilities of Hidden Markov Models (HMMs) with the long-range temporal dependency learning of Long Short-Term Memory (LSTM) networks. Optimal segmentation performance was achieved by integrating the complementary strengths of HMMs and LSTMs within a unified model architecture. The performance of the hybrid model was evaluated by analyzing the territorial behavior of tigers and the migratory patterns of turkey vultures. Results demonstrate that the hybrid HMM-LSTM system outperforms individual HMM and LSTM models in accurately identifying behavioral states. However, it shows limitations in distinguishing between hunting and exploring behaviors in tigers, as well as migratory and non-migratory states in vultures. The findings of this thesis establish the hybrid HMM-LSTM model as a powerful approach to improve the segmentation and interpretation of complex animal movement patterns.

English

ISBN: 9798291539200Subjects--Topical Terms:

524010
Geography.
Subjects--Index Terms:

Animal behavior

Trajectory Segmentation for Behavioral Mode Detection in Animal Movement /
LDR:02725nmm a22004453i 4500 001 2427599
005 20260911085209.5
006 m o d
007 cr|nu|||||||||
008 260929s2025 miu||||||m |||||||eng d
020 $a 9798291539200
035 $a (MiAaPQD)AAI31995282
035 $a AAI31995282
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Salami, Mary Idera, $e author. $3 3814547
245 1 0 $a Trajectory Segmentation for Behavioral Mode Detection in Animal Movement / $c Mary Idera Salami.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2025
300 $a 1 electronic resource (69 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 87-02.
500 $a Advisors: Dodge, Somayeh Committee members: Nelson, Trisalyn; Frazier, Amy.
502 $b M.A. $c University of California, Santa Barbara $d 2025.
520 $a Accurately segmenting animal movement trajectories into behavioral states is crucial for understanding ecological dynamics and advancing conservation strategies. Traditional segmentation methods often struggle when processing complex, noisy, high-resolution movement data. This research proposes a hybrid modeling framework that combines the probabilistic state inference capabilities of Hidden Markov Models (HMMs) with the long-range temporal dependency learning of Long Short-Term Memory (LSTM) networks. Optimal segmentation performance was achieved by integrating the complementary strengths of HMMs and LSTMs within a unified model architecture. The performance of the hybrid model was evaluated by analyzing the territorial behavior of tigers and the migratory patterns of turkey vultures. Results demonstrate that the hybrid HMM-LSTM system outperforms individual HMM and LSTM models in accurately identifying behavioral states. However, it shows limitations in distinguishing between hunting and exploring behaviors in tigers, as well as migratory and non-migratory states in vultures. The findings of this thesis establish the hybrid HMM-LSTM model as a powerful approach to improve the segmentation and interpretation of complex animal movement patterns.
546 $a English
590 $a School code: 0035
650 4 $a Geography. $3 524010
650 4 $a Ecology. $3 516476
650 4 $a Behavioral sciences. $3 529833
653 $a Animal behavior
653 $a Hidden Markov Models
653 $a Hybrid models
653 $a Long Short-Term Memory
653 $a Movement ecology
653 $a Trajectory segmentation
690 $a 0366
690 $a 0329
690 $a 0602
710 2 $a University of California, Santa Barbara. $b Geography. $e degree granting institution. $3 3814548
720 1 $a Dodge, Somayeh $e degree supervisor.
773 0 $t Masters Abstracts International $g 87-02.
790 $a 0035
791 $a M.A.
792 $a 2025
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31995282