Leutrim Uka

Leutrim Uka

Auto-generated profile, based on publicly available information

Leutrim Uka


Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 June 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
LULeutrim Uka
Leutrim Uka
Severino PintoSeverino Pinto
Severino Pinto
Gundula HoffmannGundula Hoffmann
Gundula Hoffmann
MHMarina M.-C. Höhne
Marina M.-C. Höhne
Automated cattle posture-classification systems frequently report near-perfect accuracy, yet their robustness under realistic deployment conditions remains largely unknown. In particular, it is unclear whether multimodal sensor fusion improves generalisation or leads models to rely on context-specific signals that fail under distribution shift. Here, we evaluate the robustness of automated posture classification (lying versus standing) using collar accelerometers, rumen-bolus sensors, and environmental measurements collected from a pasture-based beef cattle herd across two consecutive years (2024-2025). XGBoost served as the primary model, with Logistic Regression, Random Forest, and Long Short-Term Memory (LSTM) networks evaluated as comparative baselines. Model robustness was assessed under progressively more stringent evaluation protocols, ranging from conventional random train–test splits to leave-one-animal-out validation and cross-year evaluation on an independent cohort of previously unseen animals recorded one year later. While multimodal models achieved strong within-year performance (macro-F1 = 0.94), the performance declined substantially under cross-year evaluation (macro-F1 = 0.49). Unexpectedly, the collar-only model outperformed all multimodal configurations under temporal shift (macro-F1 = 0.54), indicating that additional physiological and environmental information did not improve robustness. Explainability analysis revealed persistent reliance on rumen-bolus activity and environmental variables even when predictive performance deteriorated, a pattern consistent with shortcut learning in which models exploit context-dependent signals that do not remain reliable under covariate shift. Distribution-shift diagnostics further confirmed substantial differences in the feature distributions between recording years. Our findings demonstrate that commonly used evaluation protocols can substantially overestimate real-world performance and that multimodal sensor fusion may reduce, rather than improve, robustness under temporal distribution shift. More broadly, the results highlight that benchmark accuracy alone is insufficient to assess deployment readiness and underscore the need for robustness-centred evaluation in livestock-monitoring research.Together, our results highlight the importance of evaluating livestock-monitoring systems under realistic deployment conditions and demonstrate how explainability analysis can reveal hidden model dependencies and provide insight into model behaviour.