Online-controlled experiment evaluation - Data Set Digital learning platforms are more and more used in blended classroom scenarios in Germany. However, as learning processes are different among students, adaptive learning platforms can offer personalized learning, e.g. by individual feedback and corrections, task sequencing, or recommendations. As digital learning platforms are already used in classroom settings, we propose the transformation of these plat-forms into adaptive learning environments. To measure the effectiveness and improvements achieved through the adaptions an online-controlled experiment design is created. In our experiment, we therefore investigate the effectiveness of different inter-ventions on a large user group in a four-month online-controlled experiment. For this purpose, the highly frequented German learning platform Orthografietrainer.net was transformed into an adaptive learning platform and users were randomly assigned to different interventions. The experimental design is published here: N. Rzepka, K. Simbeck, H.-G. Müller, and N. Pinkwart An Online Controlled Experiment Design to Support the Transformation of Digital Learning towards Adaptive Learning Platforms Proceedings of the 14th International Conference on Computer Supported Education - Volume 2: CSEDU,, SciTePress, 2022, ISBN 978-989-758-562-3 The architectural concept is published here: Rzepka, N., Simbeck, K., Müller, H.-G. & Pinkwart, N., (2022). Adaptive Learning as a Service – A concept to extend digital learning platforms?. In: Henning, P. A., Striewe, M.-0. 0. & Wölfel, M.-0. 0. (Hrsg.), 20. Fachtagung Bildungstechnologien (DELFI). Bonn: Gesellschaft für Informatik e.V.. (S. 237-238). DOI: 10.18420/delfi2022-049 The findings of this experiment are published here: tba The code to this evaluation can be found on Zenodo: 10.5281/zenodo.7755546

Online-controlled experiment evaluation - Data Set Digital learning platforms are more and more used in blended classroom scenarios in Germany. However, as learning processes are different among students, adaptive learning platforms can offer personalized learning, e.g. by individual feedback and corrections, task sequencing, or recommendations. As digital learning platforms are already used in classroom settings, we propose the transformation of these plat-forms into adaptive learning environments. To measure the effectiveness and improvements achieved through the adaptions an online-controlled experiment design is created. In our experiment, we therefore investigate the effectiveness of different inter-ventions on a large user group in a four-month online-controlled experiment. For this purpose, the highly frequented German learning platform Orthografietrainer.net was transformed into an adaptive learning platform and users were randomly assigned to different interventions. The experimental design is published here: N. Rzepka, K. Simbeck, H.-G. Müller, and N. Pinkwart An Online Controlled Experiment Design to Support the Transformation of Digital Learning towards Adaptive Learning Platforms Proceedings of the 14th International Conference on Computer Supported Education - Volume 2: CSEDU,, SciTePress, 2022, ISBN 978-989-758-562-3 The architectural concept is published here: Rzepka, N., Simbeck, K., Müller, H.-G. & Pinkwart, N., (2022). Adaptive Learning as a Service – A concept to extend digital learning platforms?. In: Henning, P. A., Striewe, M.-0. 0. & Wölfel, M.-0. 0. (Hrsg.), 20. Fachtagung Bildungstechnologien (DELFI). Bonn: Gesellschaft für Informatik e.V.. (S. 237-238). DOI: 10.18420/delfi2022-049 The findings of this experiment are published here: tba The code to this evaluation can be found on Zenodo: 10.5281/zenodo.7755546

Solution Probability Model and Fairness Evaluation This in-session prediction model seeks to predict the users’ performance on the Orthografietrainer.net platform. The target variable is binary and predicts if the user will do the following sentence correctly or not. For fairness evaluations the best models (MLP and DTE), and the worst model (SVM) are considered. A random state is not set, thus, results might differ marginally. A detailed description of the solution probability model and the fairness evaluation can be found here: tba

Solution Probability Model and Fairness Evaluation This in-session prediction model seeks to predict the users’ performance on the Orthografietrainer.net platform. The target variable is binary and predicts if the user will do the following sentence correctly or not. For fairness evaluations the best models (MLP and DTE), and the worst model (SVM) are considered. A random state is not set, thus, results might differ marginally. A detailed description of the solution probability model and the fairness evaluation can be found here: tba

In-session dropout prediction model This project describes an in-session prediction model that predicts student early dropout from online learning exercises.
Dropout prediction models for Massive Open Online Courses (MOOCs) have shown high accuracy rates in
the past and make personalized interventions possible. While MOOCs have traditionally high dropout rates,
school homework and assignments are supposed to be completed by all learners. In the pandemic, online
learning platforms were used to support school teaching. In this setting, dropout predictions have to be designed differently as a simple dropout from the (mandatory) class is not possible. The aim of our work is to
transfer traditional temporal dropout prediction models to in-session dropout prediction for school-supporting
learning platforms. For this purpose, we used data from more than 164,000 sessions by 52,000 users of the
online language learning platform orthografietrainer.net. We calculated time-progressive machine learning
models that predict dropout after each step (completed sentence) in the assignment using learning process
data. The multilayer perceptron is outperforming the baseline algorithms with up to 87% accuracy. By extending the binary prediction with dropout probabilities, we were able to design a personalized intervention
strategy that distinguishes between motivational and subject-specific interventions.
A random state is not set, thus, results might differ marginally. Whole project described in:
N. Rzepka, K. Simbeck, H.-G. Müller, and N. Pinkwart
Keep It Up: In-session Dropout Prediction to Support Blended Classroom Scenarios
Proceedings of the 14th International Conference on Computer Supported Education - Volume 2: CSEDU,
SciTePress, 2022, ISBN 978-989-758-562-3

In-session dropout prediction model This project describes an in-session prediction model that predicts student early dropout from online learning exercises.
Dropout prediction models for Massive Open Online Courses (MOOCs) have shown high accuracy rates in
the past and make personalized interventions possible. While MOOCs have traditionally high dropout rates,
school homework and assignments are supposed to be completed by all learners. In the pandemic, online
learning platforms were used to support school teaching. In this setting, dropout predictions have to be designed differently as a simple dropout from the (mandatory) class is not possible. The aim of our work is to
transfer traditional temporal dropout prediction models to in-session dropout prediction for school-supporting
learning platforms. For this purpose, we used data from more than 164,000 sessions by 52,000 users of the
online language learning platform orthografietrainer.net. We calculated time-progressive machine learning
models that predict dropout after each step (completed sentence) in the assignment using learning process
data. The multilayer perceptron is outperforming the baseline algorithms with up to 87% accuracy. By extending the binary prediction with dropout probabilities, we were able to design a personalized intervention
strategy that distinguishes between motivational and subject-specific interventions.
A random state is not set, thus, results might differ marginally. Whole project described in:
N. Rzepka, K. Simbeck, H.-G. Müller, and N. Pinkwart
Keep It Up: In-session Dropout Prediction to Support Blended Classroom Scenarios
Proceedings of the 14th International Conference on Computer Supported Education - Volume 2: CSEDU,
SciTePress, 2022, ISBN 978-989-758-562-3

We propose a new approach to the problem of high-dimensional multivariate ANOVA via bootstrapping max statistics that involve the differences of sample mean vectors. The proposed method proceeds via the construction of simultaneous confidence regions for the differences of population mean vectors. It is suited to simultaneously test the equality of several pairs of mean vectors of potentially more than two populations. By exploiting the variance decay property that is a natural feature in relevant applications, we are able to provide dimension-free and nearly-parametric convergence rates for Gaussian approximation, bootstrap approximation, and the size of the test. We demonstrate the proposed approach with ANOVA problems for functional data and sparse count data. The proposed methodology is shown to work well in simulations and several real data applications.

We propose a new approach to the problem of high-dimensional multivariate ANOVA via bootstrapping max statistics that involve the differences of sample mean vectors. The proposed method proceeds via the construction of simultaneous confidence regions for the differences of population mean vectors. It is suited to simultaneously test the equality of several pairs of mean vectors of potentially more than two populations. By exploiting the variance decay property that is a natural feature in relevant applications, we are able to provide dimension-free and nearly-parametric convergence rates for Gaussian approximation, bootstrap approximation, and the size of the test. We demonstrate the proposed approach with ANOVA problems for functional data and sparse count data. The proposed methodology is shown to work well in simulations and several real data applications.

This dataset contains the digitized treatments in Plazi based on the original journal article Bravo, Maria Fernanda Montoya, Müller, Hans-Georg, Arango, Claudia P., Tigreros, Paulo, Melzer, Roland R. (2009): Morphology of shallow-water sea spiders from the Colombian Caribbean. add _ journal _ name _ here 32 (1): 9-34, DOI: 10.5281/zenodo.16850989