Gutenberg Open Science

The Open Science Repository of Johannes Gutenberg University Mainz.

Visible. Citable. Open.

Recent Submissions

ZeitschriftenaufsatzAccess status: Open Access ,
Feature-weighted maximum representative subsampling
(2026)
Hauptmann, Tony
;
Kramer, Stefan
In the social sciences, it is often necessary to debias studies and surveys before valid conclusions can be drawn. Debiasing algorithms enable the computational removal of bias using sample weights. However, an issue arises when only a subset of features is highly biased, while the rest are already representative. Algorithms need to substantially alter the sample distribution to handle a few highly biased features, which can, in turn, introduce bias into otherwise representative variables. To address this issue, we developed a method that uses feature weights to minimize the impact of highly biased features on the computation of sample weights. Our algorithm is based on Maximum Representative Subsampling (MRS), which debiases datasets by iteratively removing elements from a non-representative sample to align it with a representative one. The new algorithm, named feature-weighted MRS (FW-MRS), decreases the emphasis on highly biased features, allowing it to retain more instances for downstream tasks. The feature weights are derived from the feature importance of a domain classifier trained to differentiate between the representative and non-representative datasets. We validated FW-MRS using eight tabular datasets, each of which we artificially biased. Biased features can be important for downstream tasks, and focusing less on them could reduce generalization. For this reason, we assessed the generalization performance of FW-MRS on downstream tasks and found no statistically significant differences. Additionally, FW-MRS was applied to a real-world dataset from the social sciences. The source code is available at https://github.com/kramerlab/FeatureWeightDebiasing.
ZeitschriftenaufsatzAccess status: Open Access ,
Rhodoliths can act as daily resolution paleotemperature archives in the Red Sea
(2026)
Li, Lena Y.
;
Bernal-Tamayo, Juan P.
;
Hetzinger, Steffen
;
Halfar, Jochen
;
Rich, Walter A.
Long-term, high-frequency temperature records are critical for evaluating the vulnerability of marine ecosystems to ongoing ocean warming. Commonly used paleoceanographic archives, such as corals and bivalves, face regional and ecological constraints. Rhodoliths, free-living coralline red algal nodules, are established and globally accessible alternatives, yet their complex three-dimensional growth structures hinder the construction of continuous age models. Here we present a workflow combining staining-based growth calibration, semi-automated increment detection using micro-computed tomography, and dynamic time warping to assemble a daily chronology spanning 133 days (March–July) from seven fruticose thalli of a single rhodolith from the central Red Sea. The resulting multi-thallus chronology, combined with a multi-element temperature proxy (Magnesium-to-Strontium), yields closer agreement with in-situ logger temperatures (R² = 0.91, RMSE = 0.63 °C) than individual thalli or single-element proxies, establishing a proof-of-concept framework for daily-resolved paleotemperature reconstructions from tropical rhodoliths.
ZeitschriftenaufsatzAccess status: Open Access ,
Music-integrated strength–proprioceptive training improves lower-limb performance and postural balance in adolescents with visual impairment : a randomized controlled trial
(2026)
Baccouch, Rym
;
Maatoug, Hana
;
Borji, Rihab
;
Ammar, Achraf
;
Fendri, Thouraya
Adolescents with visual impairment (VI) often exhibit reduced physical fitness (PF) and functional autonomy due to limited sensory input and restricted opportunities for physical activity (PA). While integrating music into exercise programs has been shown to enhance motor engagement and sensory compensation in other populations, its potential role within physical training for adolescents with VI remains underexplored. Therefore, this study investigated whether integrating music listening into a combined strength–proprioceptive physical training program could further enhance PF in adolescents with congenital VI. Fifty-nine volunteer participants were randomly assigned to a physical training group (PTG), music-physical training group (MPTG) and control group (CG). The sit-to-stand test (STST), the Killy test, the functional reach test (FRT), and the single-leg stance test (SLST) were conducted before and after 8 weeks of training to assess PF. Our results showed significant pre–post improvements in all assessed PF outcomes in both PTG and MPTG (all p < 0.001) following the intervention. While both PTG and MPTG demonstrated significant pre–post improvements, post-intervention scores in STST (36.80 ± 6.17 vs. 29.47 ± 3.23 reps), Killy test (77.42 ± 20.10 vs. 63.57 ± 21.16 s), and SLST (25.09 ± 4.33 vs. 20.31 ± 5.99 s) were significantly (p < 0.001; p < 0.01; p < 0.001 respectively) higher in the MPTG compared to the PTG. However, no significant between-group difference was observed for the FRT (42.04 ± 4.81 vs. 40.84 ± 5.82 cm) following the intervention. No meaningful changes were observed in the control group. These findings suggest that integrating music into strength–proprioceptive training may provide an additional benefit for selected functional capacities in adolescents with VI, which may in turn support greater autonomy in daily activities.