Increasingly available open medical and health datasets encourage data-driven research with a promise of improving patient care through knowledge discovery and algorithm development. Among efficient approaches to such high-dimensional problems are a number of machine learning methods, which are applied in this paper to pressure ulcer prediction in modular critical care data. An inherent property of many health-related datasets is a high number of irregularly sampled time-variant and scarcely populated features, often exceeding the number of observations. Although machine learning methods are known to work well under such circumstances, many choices regarding model and data processing exist. In particular, this paper address both theoretical and practical aspects related to the application of six classification models to pressure ulcers, while utilizing one of the largest available Medical Information Mart for Intensive Care (MIMIC-IV) databases. Random forest, with an accuracy of 96%, is the best-performing approach among the considered machine learning algorithms.

As energy distribution systems evolve from a traditional hierarchical load structure towards distributed smart grids, flexibility is increasingly investigated as both a key measure and core challenge of grid balancing. This paper contributes to the theoretical framework for quantifying network flexibility potential by introducing a machine learning based node characterization. In particular, artificial neural networks are considered for classification of historic demand data from several network substations. Performance of the resulting classifiers is evaluated with respect to clustering analysis and parameter space of the models considered, while the bootstrapping based statistical evaluation is reported in terms of mean confusion matrices. The resulting meta-models of individual nodes can be further utilized on a network level to mitigate the difficulties associated with identifying, implementing and actuating many small sources of energy flexibility, compared to the few large ones traditionally acknowledged.

This paper aims to analyse possibilities of train type identification in railway switches and crossings (S&C) based on accelerometer data by using contemporary machine learning methods such as neural networks. That is a unique approach since trains have been only identified in a straight track. Accelerometer sensors placed around the S&C structure were the source of input data for subsequent models. Data from four S&C at different locations were considered and various neural network architectures evaluated. The research indicated the feasibility to identify trains in S&C using neural networks from accelerometer data. Models trained at one location are generally transferable to another location despite differences in geometrical parameters, substructure, and direction of passing trains. Other challenges include small dataset and speed variation of the trains that must be considered for accurate identification. Results are obtained using statistical bootstrapping and are presented in a form of confusion matrices.

We present CWRCzech, Click Web Ranking dataset for Czech, a 100M query-document Czech click dataset for relevance ranking with user behavior data collected from search engine logs of Seznam.cz. To the best of our knowledge, CWRCzech is the largest click dataset with raw text published so far. It provides document positions in the search results as well as information about user behavior: 27.6M clicked documents and 10.8M dwell times. In addition, we also publish a manually annotated Czech test for the relevance task, containing nearly 50k query-document pairs, each annotated by at least 2 annotators. Finally, we analyze how the user behavior data improve relevance ranking and show that models trained on data automatically harnessed at sufficient scale can surpass the performance of models trained on human annotated data. CWRCzech is published under an academic non-commercial license and is available to the research community at https://github.com/seznam/CWRCzech.

Cracking is a critical issue in civil engineering, yet its reliable and straightforward assessment remains challenging. This study applies convolutional neural networks (CNNs) for automated crack recognition and quantitative analysis. A dataset of 1,002 images with corresponding masks, featuring diverse crack patterns and surface textures, was created through systematic variation of slag-based and cement paste compositions and moist-curing durations. Based on this dataset, the open-source CrackenPy toolkit was developed, comprising a Python library, a pre-trained model, and the dataset itself. Unlike existing tools limited to binary segmentation (e.g., crack vs. background), CrackenPy identifies four classes: cracks, pores, matrix, and background. It automatically detects specimens and computes pixel-intensity and CNN-derived metrics, including crack ratio and mean crack thickness. Continuous monitoring using a custom-built apparatus enabled correlations between image-derived metrics and independently measured shrinkage. The resulting reproducible image-based measurement procedure supports quantitative binder research and has potential for structural health monitoring.