MJMatthias Jakobs

Matthias Jakobs

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Matthias Jakobs has a total of 30 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Size Matters: A Large-Scale, Real-World Dataset of Smartphone-Based Obstructive Sleep Apnea Screening in the General Population

    by Christoph Janott, Marcel Braun, Torsten Eggert, Matthias Jakobs, Simone Salzger, Jörg Ulrich Sommer, Clemens Heiser, Christoph Schöbel

    Background: Obstructive sleep apnea (OSA) is a highly prevalent sleep-related breathing disorder that remains substantially underdiagnosed and is causally linked to cardiovascular and metabolic morbidity. Clinical diagnosis relies on the apnea–hypopnea index (AHI) derived from polysomnography (PSG) or home sleep apnea testing. Low-threshold, clinically validated screening tools that can be used repeatedly outside a clinical setting may help to reduce underdiagnosis and to capture longitudinal information. Methods: We describe a real-world dataset generated by a medical device-grade, smartphone-only OSA screening application (Snorefox M) that calculates the AHI from nocturnal breathing sounds. All users who provided informed consent for the research use of their anonymized data between March 2023 and December 2025 were included. The diagnostic performance of the application had previously been validated against parallel PSG. Additional post hoc analysis of classification performance at additional AHI thresholds, together with the linear correlation between the app-derived and PSG-derived AHI are described. Results: The dataset comprises 239,132 measurements from 31,647 consenting adults. A total of 14,351 users contributed at least two valid nights, yielding 221,820 nights of longitudinal screening data. Across all attempted measurements, 96.5% yielded a valid analysis result. The mean per-measurement AHI was 14.3 (median 11.8, SD 11.0). Voluntary, self-reported demographic information was available for a subgroup of 8092 users (mean age 54.4 years, mean ESS score 7.2). Against PSG, the application detected moderate-to-severe OSA (AHI ≥ 15) with a sensitivity of 0.91 and a specificity of 0.83, and severe OSA (AHI ≥ 30) with a sensitivity of 0.93 and a specificity of 0.97. The app-derived AHI correlated strongly with the PSG-derived AHI (Pearson r = 0.80, p < 0.001). The app-derived AHI discriminated against the PSG reference with an area under the curve (AUC) of 0.951 for severe OSA (AHI ≥ 30), 0.925 for moderate-to-severe OSA (AHI ≥ 15) and 0.877 for the AHI ≥ 10 threshold. Conclusions: This is, to our knowledge, the largest longitudinal dataset of unsupervised, real-world OSA screening from a general adult population using a clinically validated, low-threshold smartphone tool. Smartphone-based screening can complement established diagnostic pathways, support pre-clinical triage and generate longitudinal data not accessible through single-night testing, while self-selection and the unsupervised nature of data capture constrain representativeness. By introducing this resource, this work provides the basis for the in-depth analyses of demographics, symptom–result concordance and night-to-night variability.

  • Decentralized Time Series Classification with ROCKET Features

    by Bruno Casella, Matthias Jakobs, Marco Aldinucci, Sebastian Buschjäger

  • Open Access

    Decentralized Time Series Classification with ROCKET Features

    by Bruno Casella, Matthias Jakobs, Marco Aldinucci, Sebastian Buschjäger

    Time series classification (TSC) is a critical task with applications in various domains, including healthcare, finance, and industrial monitoring. Due to privacy concerns and data regulations, Federated Learning has emerged as a promising approach for learning from distributed time series data without centralizing raw information. However, most FL solutions rely on a client-server architecture, which introduces robustness and confidentiality risks related to the distinguished role of the server, which is a single point of failure and can observe knowledge extracted from clients. To address these challenges, we propose DROCKS, a fully decentralized FL framework for TSC that leverages ROCKET (RandOm Convolutional KErnel Transform) features. In DROCKS, the global model is trained by sequentially traversing a structured path across federation nodes, where each node refines the model and selects the most effective local kernels before passing them to the successor. Extensive experiments on the UCR archive demonstrate that DROCKS outperforms state-of-the-art client-server FL approaches while being more resilient to node failures and malicious attacks. Our code is available at https://anonymous.4open.science/r/DROCKS-7FF3/README.md.

  • Open Access

    Interpretable Rules for Online Failure Prediction: A Case Study on the Metro do Porto dataset

    by Matthias Jakobs, Bruno Veloso, Joao Gama

    Due to their high predictive performance, predictive maintenance applications have increasingly been approached with Deep Learning techniques in recent years. However, as in other real-world application scenarios, the need for explainability is often stated but not sufficiently addressed. This study will focus on predicting failures on Metro trains in Porto, Portugal. While recent works have found high-performing deep neural network architectures that feature a parallel explainability pipeline, the generated explanations are fairly complicated and need help explaining why the failures are happening. This work proposes a simple online rule-based explainability approach with interpretable features that leads to straightforward, interpretable rules. We showcase our approach on MetroPT2 and find that three specific sensors on the Metro do Porto trains suffice to predict the failures present in the dataset with simple rules.

  • Open Access

    Explainable Adaptive Tree-based Model Selection for Time Series Forecasting

    by Matthias Jakobs, Amal Saadallah

    Tree-based models have been successfully applied to a wide variety of tasks, including time series forecasting. They are increasingly in demand and widely accepted because of their comparatively high level of interpretability. However, many of them suffer from the overfitting problem, which limits their application in real-world decision-making. This problem becomes even more severe in online-forecasting settings where time series observations are incrementally acquired, and the distributions from which they are drawn may keep changing over time. In this context, we propose a novel method for the online selection of tree-based models using the TreeSHAP explainability method in the task of time series forecasting. We start with an arbitrary set of different tree-based models. Then, we outline a performance-based ranking with a coherent design to make TreeSHAP able to specialize the tree-based forecasters across different regions in the input time series. In this framework, adequate model selection is performed online, adaptively following drift detection in the time series. In addition, explainability is supported on three levels, namely online input importance, model selection, and model output explanation. An extensive empirical study on various real-world datasets demonstrates that our method achieves excellent or on-par results in comparison to the state-of-the-art approaches as well as several baselines.

  • An Empirical Evaluation of the Rashomon Effect in Explainable Machine Learning

    by Sebastian Müller, Vanessa Toborek, Katharina Beckh, Matthias Jakobs, Christian Bauckhage, Pascal Welke


  • Shapley Values with Uncertain Value Functions

    by Raoul Heese, Sascha Mücke, Matthias Jakobs, Thore Gerlach, Nico Piatkowski


  • A Unified Framework for Assessing Energy Efficiency of Machine Learning

    by Raphael Fischer, Matthias Jakobs, Sascha Mücke, Katharina Morik


  • Open Access

    Yes we care!-Certification for machine learning methods through the care label framework

    by Katharina J. Morik, Helena Kotthaus, Raphael Fischer, Sascha Mücke, Matthias Jakobs, Nico Piatkowski, Andreas Pauly, Lukas Heppe, Danny Heinrich

    Machine learning applications have become ubiquitous. Their applications range from embedded control in production machines over process optimization in diverse areas (e.g., traffic, finance, sciences) to direct user interactions like advertising and recommendations. This has led to an increased effort of making machine learning trustworthy. Explainable and fair AI have already matured. They address the knowledgeable user and the application engineer. However, there are users that want to deploy a learned model in a similar way as their washing machine. These stakeholders do not want to spend time in understanding the model, but want to rely on guaranteed properties. What are the relevant properties? How can they be expressed to the stake- holder without presupposing machine learning knowledge? How can they be guaranteed for a certain implementation of a machine learning model? These questions move far beyond the current state of the art and we want to address them here. We propose a unified framework that certifies learning methodsviacare labels. They are easy to understand and draw inspiration from well-known certificates like textile labels or property cards of electronic devices. Our framework considers both, the machine learning theory and a given implementation. We test the implementation's compliance with theoretical properties and bounds.