20.500.12556/RUL-189963 A Hybrid Analytical Pipeline for Circadian Rhythm Modeling from Accelerometer Data: A Maned Wolf Case Study Hibridni pristop za modeliranje cirkadianih ritmov na podlagi podatkov pospeškometrov: študija primera grivastega volka This thesis presents a reproducible analytical pipeline for characterising circadian activity patterns from biologger time series, demonstrated on maned wolves (Chrysocyon brachyurus) in Brazil. The pipeline was developed during the process of the analysis of 408,171 observations from 15 subjects, recorded by implanted biologgers. Activity was quantified using the firmware-derived count of active minutes per 15-minute epoch; the accompanying per-axis mean accelerations were shown to describe body orientation rather than movement, and were reserved as an independent posture channel. Per-subject Gaussian Hidden Markov Models (HMMs) supported an interpretable four-state behavioural representation (deep rest, light rest, moderate activity, high activity), whose ordering was corroborated by the posture channel: mean posture magnitude declined monotonically with increasing activity. Circadian structure was quantified using multi-component cosinor regression. The five-harmonic model clearly outperformed the single-component baseline at population level (R2 = 0.993 vs. R2 = 0.878), confirming that multi-harmonic rhythmometry is required to capture the observed diel waveform. The fitted harmonics are ecologically coherent, with the fundamental peaking at 23:04 and the 12-hour component at 06:02 and 18:02. The framework was extended to Hidden Semi-Markov Models (HSMMs) with explicit Poisson duration distributions; comparison with HMM outputs showed substantial duration-related differences. For inferential modelling, sex was retained as the only group-level covariate in linear mixed models. Fixed sex effects were modest, while individual-level heterogeneity was substantial relative to the group-level sex effect (conditional R2 = 0.974-0.984; ICC = 0.954-0.971). Overall, the thesis contributes a coherent and extensible workflow that links behavioural-state inference, circadian rhythm modeling, and individual-variation analysis without requiring labelled behavioural ground truth. Magistrska naloga predstavlja ponovljiv analitični cevovod za karakterizacijo cirkadianih vzorcev aktivnosti iz časovnih vrst vsadnih zapisovalnikov (biologgerjev), prikazan na primeru grivastega volka (Chrysocyon brachyurus) v Braziliji. Analiza, v okviru katere smo razvili predstavljen cevovod, vključuje 408.171 opazovanj pri 15 osebkih. Aktivnost kvantificiramo s številom aktivnih minut na 15-minutni interval, ki ga izračuna programska oprema naprave. Pokazali smo, da povprečne vrednosti posameznih osi ne merijo intenzivnosti gibanja, temveč orientacijo telesa, zato jih uporabimo kot neodvisen kanal za preverjanje stanj. Za identifikacijo vedenjskih stanj uporabimo Gaussove skrite Markove modele (HMM), prilagojene po posameznem osebku, in ohranimo štiristopenjsko predstavitev stanj (globok počitek, lahek počitek, zmerna aktivnost, visoka aktivnost). Razvrstitev stanj potrdi neodvisni kanal orientacije: povprečna magnituda drže monotono pada z naraščajočo aktivnostjo. Cirkadiano strukturo kvantificiramo z večkomponentno kosinorno regresijo. Model s petimi harmoniki na ravni populacije bistveno preseže enokomponentni model (R2 = 0,993 proti R2 = 0,878), kar potrjuje potrebo po večharmoničnem opisu dnevnega vzorca aktivnosti. Osnovni harmonik ima vrh ob 23:04, 12-urna pa ob 06:02 in 18:02, kar približno ustreza jutranjemu in večernemu somraku. Ogrodje razširimo tudi na skrite pol-Markove modele (HSMM) z eksplicitnimi Poissonovimi porazdelitvami trajanja; primerjava s HMM pokaže opazne razlike v dinamiki trajanja vedenjskih epizod. Za inferenčno modeliranje v končni različici naloge ohranimo spol kot edino skupinsko kovariato v linearnih mešanih modelih. Fiksni učinki spola so zmerni, medtem ko je medosebkovna heterogenost velika v primerjavi s skupinskim učinkom spola (pogojni R2 = 0,974-0,984; ICC = 0,954-0,971). Ključni prispevek naloge je enoten in razširljiv pristop, ki brez označenih vedenjskih podatkov povezuje inferenco vedenjskih stanj, modeliranje cirkadianih ritmov in analizo individualnih razlik. analytical pipeline Python package open-source software hidden Markov models hidden semi-Markov models circadian rhythms accelerometry cosinor regression linear mixed models maned wolf analitični cevovod Python knjižnica odprtokodna programska oprema skriti Markovski modeli skriti pol-Markovski modeli cirkadiani ritmi akcelerometrija regresija kosinor linearni mešani modeli grivasti volk true false false Angleški jezik Slovenski jezik Magistrsko delo/naloga 2026-10-19 11:05:23 2026-10-19 11:05:36 2026-10-30 11:34:12 0000-00-00 00:00:00 2026 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 38576 303235331 Jukan_Nermin_-_Hibridni_pristop_za_modeliranje_cirkadianih_ritmov_na_podlagi_podatkov_pospeskome.pdf Jukan_Nermin_-_Hibridni_pristop_za_modeliranje_cirkadianih_ritmov_na_podlagi_podatkov_pospeskome.pdf 1 39E4CFA92D6B135BA5B77F402157D656 41d1e67770125b67151a9c1f533644b1e1240cf4753b84478a03feeff2936ec3 1f6682dd-cb9c-11f1-8bc5-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=250883 Fakulteta za računalništvo in informatiko 0 0 0