# Evaluation of Stress Detection as Time Series Events -- A Novel Window-Based F1-Metric

arXiv: https://arxiv.org/abs/2509.03240 · HTML: https://mlsanity.com/paper/2509.03240/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 33 references that are in the map.
- First posted on arXiv 3 Sep 2025.

## Builds on (most cited first, 10 of 10)

- [Large Language Models](https://mlsanity.com/paper/2307.05782/) (2023)
- [A decoder-only foundation model for time-series forecasting](https://mlsanity.com/paper/2310.10688/) (2023)
- [Large Brain Model for Learning Generic Representations with Tremendous EEG Data in BCI](https://mlsanity.com/paper/2405.18765/) (2024)
- [Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications](https://mlsanity.com/paper/1802.03903/) (2018)
- [Towards a Rigorous Evaluation of Time-series Anomaly Detection](https://mlsanity.com/paper/2109.05257/) (2021)
- [An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Series](https://mlsanity.com/paper/2109.11428/) (2021)
- [Empowering Time Series Analysis with Foundation Models: A Comprehensive Survey](https://mlsanity.com/paper/2405.02358/) (2024)
- [A multimodal sensor dataset for continuous stress detection of nurses in a hospital](https://mlsanity.com/paper/2108.07689/) (2021)
- [ADARP: A Multi Modal Dataset for Stress and Alcohol Relapse Quantification in Real Life Setting](https://mlsanity.com/paper/2206.14568/) (2022)
- [Predicting Driver Self-Reported Stress by Analyzing the Road Scene](https://mlsanity.com/paper/2109.13225/) (2021)

## Abstract (arXiv)

Accurate evaluation of event detection in time series is essential for applications such as stress monitoring with wearable devices, where ground truth is typically annotated as single-point events, even though the underlying phenomena are gradual and temporally diffused. Standard metrics like F1 and point-adjusted F1 (F1pa) often misrepresent model performance in such real-world, imbalanced datasets. We introduce a window-based F1 metric (F1w) that incorporates temporal tolerance, enabling a more robust assessment of event detection when exact alignment is unrealistic. Empirical analysis in three physiological datasets, two in-the-wild (ADARP, Wrist Angel) and one experimental (ROAD), indicates that F1w reveals meaningful model performance patterns invisible to conventional metrics, while its window size can be adapted to domain knowledge to avoid overestimation. We show that the choice of evaluation metric strongly influences the interpretation of model performance: using predictions from TimesFM, only our temporally tolerant metrics reveal statistically significant improvements over random and null baselines in the two in-the-wild use cases. This work addresses key gaps in time series evaluation and provides practical guidance for healthcare applications where requirements for temporal precision vary by context.