SoftwareX最新文献

This manuscript introduces Subgroups, an openly accessible Python library designed to ease the use of Subgroup Discovery (SD) algorithms for machine learning and data science. The Subgroups Library offers several advantages: (1) Efficiency Enhancement: Developed in native Python, unlike other software available, the library prioritizes efficiency to ensure seamless execution of SD algorithms; (2) User-Friendly Interface: Modeled after the popular scikit-learn framework, the library boasts an intuitive interface, streamlining the utilization process for practitioners and non-expert programmers; (3) Trustworthy Algorithm Implementations: Drawing from scientific publications authored by leading experts, the Subgroups Library incorporates rigorously tested algorithmic implementations, ensuring reliability and accuracy in results; (4) Customization and Expansion: The modular architecture of the library facilitates effortless integration of additional quality measures, data structures, and SD algorithms, empowering users to tailor their analyses to specific needs and explore new avenues of research. Furthermore, the Subgroups Library has been successfully employed in diverse scientific papers and projects, underscoring its efficacy and versatility as a valuable tool for SD exploration and application.

Particle tracking is a common post processing method in groundwater hydrology. In this paper we describe Ichnos, a particle tracking code able to work with flow simulations obtained from either finite difference, finite element, adaptive mesh, or mesh free methods. Ichnos can trace virtual particles (streamlines) in flow fields of any fluid dynamics context, but its application is here focused on groundwater-based flow fields. The code is written in C++ and the structure of the code allows for it to be easily extended. In this study we describe the main features of the code and present several illustrations.

This paper presents the pyrepo-mcda Python package upgrade with the implementation of the Preference Vector Method (PVM) multi-criteria method. This upgrade extends the scope of multi-criteria decision analysis offered by this package. Several advantages of the PVM method, such as the reduction of the participation of decision-makers, the possibility of giving individual preference vectors, and the possibility of modification and further development, are in the interest of decision-makers in various multi-criteria decision analysis problems, particularly in the sustainability assessment.

Several Python libraries have been released for training time series forecasting models in the last few years. Most include classical statistical approaches, machine learning models, and recent deep learning architectures. Despite the great work for releasing such open-source resources, a tool that allows testing Deep Learning architectures in a framework that guarantees transparent input output management, reproducibility of the results, and expandability of the supported models is still lacking. With DSIPTS, we fill this gap, providing the community with a tool for training and comparing Deep Learning models in the time series forecasting field.

The growing demand for effective healthcare has driven advances in digital health. This digitization supposes a challenge from the point of view of privacy and the treatment of sensitive personal data while providing non-intrusive and easy-to-use digital mechanisms. This paper presents Contigo: a health monitoring system that integrates a mobile application and a web platform for detecting anomalies using Federated Learning techniques. The mobile application collects health and personal data to train a personal predictive model. It is then anonymized and aggregated into a global model to improve efficiency, reducing adoption time for new users. At the same time, the web platform allows healthcare professionals to access the data for its analysis and validation. Contigo addresses the need for user-friendly digital mechanisms in healthcare, addressing privacy concerns while improving data-driven decision-making for professionals and personalized patient care. This approach ensures privacy and facilitates continuous model improvement, providing personalized, proactive, and non-intrusive patient health analytics.

The conformal mapping function from the interior of the complex plane's unit circle to the exterior of any simple closed curve on the real plane finds widespread applications, including the use of complex variable methods in elasticity studies. Our MATLAB toolbox employs numerical methods to solve such conformal mapping functions, applicable to physical domains featuring simple closed curves of arbitrary shapes, and even extending to slit-like structures. Featuring a user-friendly GUI program, the toolbox efficiently computes conformal mapping functions, streamlining the solving process.

This article introduces the latest features and enhancements in TopoHub, an open repository for network topologies used in networking research. The major update includes a new collection of global-scale backbone topologies, generated based on population density and incorporating submarine communication cables. The web interface has been upgraded to support interactive exploration of networks, including panning and zooming. Additionally, the new version addresses usability improvements and bug fixes informed by user feedback, enhancing the overall functionality and user experience of the platform.

Load profiling for residential aggregates encounters challenges due to data scarcity and the inadequacy of standard profiles obtained from statistical analyses. In the absence of hourly data, many methods rely on standard profiles, which could lead to significant errors in consumption estimation, especially for evaluating specific aggregates. This article presents PyARC, a Python-based algorithm trainable with customizable consumption data, which addresses the problem related to evaluating the energy consumption of specific aggregates by using typological profiles extracted from similar users, thereby improving accuracy. The algorithm's innovative approach uses Association Rule Mining and Random Forest Classification to reconstruct the load profiles of aggregates, providing a more robust solution for estimating the electrical load with limited data.

The Multiclass Classification Performance (MCP) curve is an innovative method to visualize the performance of a classifier for multiclass datasets. On the other hand, the Imbalanced Multiclass Classification Performance (IMCP) curve is a novel approach to visualizing classifier performance on multiclass datasets that exhibit class imbalance, i.e. the proportions of (two or more) class labels are unequal. We have developed an open-source Python package that encompasses the functionality required to calculate and visualize these two novel classification performance measures, along with providing the calculation of the area under the curves. The MCP and IMCP curves offer advantages over the traditional ROC (Receiver Operating Characteristic) curve when dealing with multiclass and imbalanced datasets, respectively. They provide more informative insights into classifier behavior, especially in scenarios involving multiple classes or uneven class distribution.

The growing interest in the application of structured light has led to an increase in the use of spatial light modulators across users at all levels of experience. Hence, there is a need for software that controls the device, designs holograms and gathers experimental feedback. To meet these demands we present the Binary Complex Amplitude App - a standalone Matlab application that provides a graphic user interface with a full control of the Digital Micromirror Device, enabling hologram design and camera preview. We show that with all-in-one application, the user at any level of experience can operate the device and do not lose any of its capabilities.