Pub Date : 2026-02-01Epub Date: 2025-12-05DOI: 10.1016/j.softx.2025.102463
Damian Frąszczak, Edyta Frąszczak
Website phishing represents a significant cyber threat, where attackers create fraudulent websites that imitate legitimate sites to deceive users. Continuous monitoring and detection of malicious websites are crucial for mitigating this threat. This paper introduces PhishingWebCollector, an open-source Python library designed to simplify the collection and integration of phishing feeds. It is an appropriate tool for real-time blacklist updates, creating historical datasets for research, and serving as a foundation for developing AI-based phishing detection systems. Identifying phishing and spoofed websites helps generate high-quality datasets necessary for training models in automated website classification and threat identification. Leveraging Python’s asyncio, it processes multiple feeds concurrently to achieve optimal performance. Available on PyPI with extensive documentation and examples, PhishingWebCollector offers a resource-efficient solution for cybersecurity professionals and researchers.
{"title":"PhishingWebCollector: Async python library for automated phishing feed collection","authors":"Damian Frąszczak, Edyta Frąszczak","doi":"10.1016/j.softx.2025.102463","DOIUrl":"10.1016/j.softx.2025.102463","url":null,"abstract":"<div><div>Website phishing represents a significant cyber threat, where attackers create fraudulent websites that imitate legitimate sites to deceive users. Continuous monitoring and detection of malicious websites are crucial for mitigating this threat. This paper introduces PhishingWebCollector, an open-source Python library designed to simplify the collection and integration of phishing feeds. It is an appropriate tool for real-time blacklist updates, creating historical datasets for research, and serving as a foundation for developing AI-based phishing detection systems. Identifying phishing and spoofed websites helps generate high-quality datasets necessary for training models in automated website classification and threat identification. Leveraging Python’s asyncio, it processes multiple feeds concurrently to achieve optimal performance. Available on PyPI with extensive documentation and examples, PhishingWebCollector offers a resource-efficient solution for cybersecurity professionals and researchers.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102463"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145693271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-12-11DOI: 10.1016/j.softx.2025.102465
Achbarou Omar , Lachgar Mohamed , El Badouri Youssef , Rhouddani Monsef , Tahalli Anas , El Dhimni Roa
SafeOps+ is an open-source platform dedicated to automating security analysis and compliance in software development. Through a modular architecture combining a Python backend and React frontend, it integrates reference tools like Checkov and Semgrep to detect vulnerabilities and misconfigurations. Its intuitive web interface allows users to launch analyses, consult detailed reports, and track audit history. SafeOps+ facilitates the adoption of DevSecOps practices, improves traceability and reproducibility, and is designed for development teams as well as researchers and trainers in software security.
{"title":"SafeOps+: An integrated platform for automated security analysis and compliance in DevSecOps pipelines","authors":"Achbarou Omar , Lachgar Mohamed , El Badouri Youssef , Rhouddani Monsef , Tahalli Anas , El Dhimni Roa","doi":"10.1016/j.softx.2025.102465","DOIUrl":"10.1016/j.softx.2025.102465","url":null,"abstract":"<div><div>SafeOps+ is an open-source platform dedicated to automating security analysis and compliance in software development. Through a modular architecture combining a Python backend and React frontend, it integrates reference tools like Checkov and Semgrep to detect vulnerabilities and misconfigurations. Its intuitive web interface allows users to launch analyses, consult detailed reports, and track audit history. SafeOps+ facilitates the adoption of DevSecOps practices, improves traceability and reproducibility, and is designed for development teams as well as researchers and trainers in software security.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102465"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145748537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-12-11DOI: 10.1016/j.softx.2025.102466
Elżbieta Jowik , Agnieszka Jastrzębska , Michał Gamrot
Nowcasting means forecasting in fine detail, by any method, over very short horizons, from the present into the immediate future. Originally used in meteorology, the term was later adopted in economics to describe an early assessment of the economy’s current (“now”) state. It is like a weather forecast for the economy – but instead of projecting rainfall or temperature, economists use nowcasts to make a judgment about whether the economy is growing or shrinking, and whether the balance of risk is toward heating up (increasing inflation) or cooling down (lost output and rising unemployment).
Our research looks at the practical side of short-term macroeconomic forecasting and the science behind it. We propose a Python package that combines machine learning and econometrics – canonical time-series models and modern algorithms – to read the economy as it moves, reacts, and reshapes itself. Every nowcast is estimated from the ground up, not just with new data, but with updated variables, model structures, and parameters, allowing it to respond to evolving macroeconomic dynamics, structural breaks, and policy interventions in real time. Explainable AI (XAI) principles, applied along the way, ensure that the results are fully auditable. Users know which variables matter most and how each new piece of information changes the outlook.
In that sense, the package is more than a forecasting solution. It is a tool for understanding how information flows through the economy. Grounded in strong theoretical foundations and designed for evidence-based empirical analysis, it provides a way to work with real-time data without locking users into a specific way of modeling or thinking about it.
{"title":"Forecast what matters, when it matters: Introducing Maynard, a tool for modern nowcasting","authors":"Elżbieta Jowik , Agnieszka Jastrzębska , Michał Gamrot","doi":"10.1016/j.softx.2025.102466","DOIUrl":"10.1016/j.softx.2025.102466","url":null,"abstract":"<div><div>Nowcasting means forecasting in fine detail, by any method, over very short horizons, from the present into the immediate future. Originally used in meteorology, the term was later adopted in economics to describe an early assessment of the economy’s current (“now”) state. It is like a weather forecast for the economy – but instead of projecting rainfall or temperature, economists use nowcasts to make a judgment about whether the economy is growing or shrinking, and whether the balance of risk is toward heating up (increasing inflation) or cooling down (lost output and rising unemployment).</div><div>Our research looks at the practical side of short-term macroeconomic forecasting and the science behind it. We propose a Python package that combines machine learning and econometrics – canonical time-series models and modern algorithms – to read the economy as it moves, reacts, and reshapes itself. Every nowcast is estimated from the ground up, not just with new data, but with updated variables, model structures, and parameters, allowing it to respond to evolving macroeconomic dynamics, structural breaks, and policy interventions in real time. Explainable AI (XAI) principles, applied along the way, ensure that the results are fully auditable. Users know which variables matter most and how each new piece of information changes the outlook.</div><div>In that sense, the package is more than a forecasting solution. It is a tool for understanding how information flows through the economy. Grounded in strong theoretical foundations and designed for evidence-based empirical analysis, it provides a way to work with real-time data without locking users into a specific way of modeling or thinking about it.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102466"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145749396","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2026-01-23DOI: 10.1016/j.softx.2026.102528
Vadym Chibrikov, Justyna Cybulska, Artur Zdunek
The issue of food quality and its control has become a daily routine for humanity, driven by both health and economic reasons. Among other components, plant cell wall components—such as cellulose, hemicellulose, and pectin—have several pro-health roles in the human organism that are barely discussed in public. To address this, there is a clear need for a portable digital framework built on accurate, accessible, and scientifically-proven data. Here, our commitment was the development of FibreApp, an Android/iOS mobile application that helps users obtain data on the chemical composition of common fruit and vegetable species available on the European market. FibreApp's architecture was designed as a hybrid local/offline system that integrates on-device machine learning for visual identification with a pre-loaded, unified database of fruit and vegetable compositional parameters. A machine learning-powered livestream tool for image classification of fruits and vegetables was included in the app by rigorously following a systematic image acquisition protocol, coupled with a transfer learning approach using pre-trained feature extractors to train the machine learning models. The latter performed well despite significant changes in lighting and diverse polar orientations, as well as during polyclass image classification. FibreApp was released and field-tested, positioning it to capture a niche in improving public awareness of fruits and vegetables as a source of functional polysaccharides.
{"title":"FibreApp: Mobile machine learning tool for fruit and vegetable fiber content","authors":"Vadym Chibrikov, Justyna Cybulska, Artur Zdunek","doi":"10.1016/j.softx.2026.102528","DOIUrl":"10.1016/j.softx.2026.102528","url":null,"abstract":"<div><div>The issue of food quality and its control has become a daily routine for humanity, driven by both health and economic reasons. Among other components, plant cell wall components—such as cellulose, hemicellulose, and pectin—have several pro-health roles in the human organism that are barely discussed in public. To address this, there is a clear need for a portable digital framework built on accurate, accessible, and scientifically-proven data. Here, our commitment was the development of <em>FibreApp</em>, an Android/iOS mobile application that helps users obtain data on the chemical composition of common fruit and vegetable species available on the European market. <em>FibreApp</em>'s architecture was designed as a hybrid local/offline system that integrates on-device machine learning for visual identification with a pre-loaded, unified database of fruit and vegetable compositional parameters. A machine learning-powered livestream tool for image classification of fruits and vegetables was included in the app by rigorously following a systematic image acquisition protocol, coupled with a transfer learning approach using pre-trained feature extractors to train the machine learning models. The latter performed well despite significant changes in lighting and diverse polar orientations, as well as during polyclass image classification. <em>FibreApp</em> was released and field-tested, positioning it to capture a niche in improving public awareness of fruits and vegetables as a source of functional polysaccharides.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102528"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2026-01-29DOI: 10.1016/j.softx.2026.102537
Robby Ulung Pambudi, Ary Mazharuddin Shiddiqi, Royyana Muslim Ijtihadie, Muhammad Nabil Akhtar Raya Amoriza, Hardy Tee, Fadhl Akmal Madany, Rizky Januar Akbar, Dini Adni Navastara
The increasing demand for scalable and responsive Large Language Model (LLM) applications has accelerated the need for distributed inference systems capable of handling high concurrency and heterogeneous GPU resources. This paper introduces DiLLeMa, an extensible framework for distributed LLM deployment on multi-GPU clusters, designed to improve inference efficiency through workload parallelization and adaptive resource management. Built upon the Ray distributed computing framework, DiLLeMa orchestrates LLM inference across multiple nodes while maintaining balanced GPU utilization and low-latency response. The system integrates a FastAPI-based backend for coordination and API management, a React-based frontend for interactive access, and a vLLM inference engine optimized for high-throughput execution. Complementary modules for data preprocessing, semantic embedding, and vector-based retrieval further enhance contextual relevance during response generation. Illustrative examples demonstrate that DiLLeMa effectively reduces inference latency and scales efficiently.
{"title":"DiLLeMa: An extensible and scalable framework for distributed large language models (LLMs) inference on multi-GPU clusters","authors":"Robby Ulung Pambudi, Ary Mazharuddin Shiddiqi, Royyana Muslim Ijtihadie, Muhammad Nabil Akhtar Raya Amoriza, Hardy Tee, Fadhl Akmal Madany, Rizky Januar Akbar, Dini Adni Navastara","doi":"10.1016/j.softx.2026.102537","DOIUrl":"10.1016/j.softx.2026.102537","url":null,"abstract":"<div><div>The increasing demand for scalable and responsive Large Language Model (LLM) applications has accelerated the need for distributed inference systems capable of handling high concurrency and heterogeneous GPU resources. This paper introduces DiLLeMa, an extensible framework for distributed LLM deployment on multi-GPU clusters, designed to improve inference efficiency through workload parallelization and adaptive resource management. Built upon the Ray distributed computing framework, DiLLeMa orchestrates LLM inference across multiple nodes while maintaining balanced GPU utilization and low-latency response. The system integrates a <em>FastAPI</em>-based backend for coordination and API management, a <em>React</em>-based frontend for interactive access, and a vLLM inference engine optimized for high-throughput execution. Complementary modules for data preprocessing, semantic embedding, and vector-based retrieval further enhance contextual relevance during response generation. Illustrative examples demonstrate that DiLLeMa effectively reduces inference latency and scales efficiently.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102537"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146077489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2026-01-07DOI: 10.1016/j.softx.2025.102505
M. Yochum , F. Karimi , F. Wendling, M. Al Harrach
In this new version of the NeoCoMM (Neocortical Computational Microscale Model) software, we present an updated neuroinspired computational model of the cortical column that includes neuroplasticity and a transcranial Electric Stimulation (tES) modeling platform. The neuroplasticity update consists of three types of long- term plasticity models based on the calcium dynamics that are incorporated into the principal cells (PCs) of the network. For tES, a new panel in the GUI was added to simulate the electric field parameters allowing the user to simulate the impact of both Direct (tDCS) and Alternating (tACS) Current Stimulation on the network dynamics.
{"title":"Version [2.0]-[NeoCoMM: Neocortical neuro-inspired computational model for realistic microscale simulations]","authors":"M. Yochum , F. Karimi , F. Wendling, M. Al Harrach","doi":"10.1016/j.softx.2025.102505","DOIUrl":"10.1016/j.softx.2025.102505","url":null,"abstract":"<div><div>In this new version of the NeoCoMM (Neocortical Computational Microscale Model) software, we present an updated neuroinspired computational model of the cortical column that includes neuroplasticity and a transcranial Electric Stimulation (tES) modeling platform. The neuroplasticity update consists of three types of long- term plasticity models based on the calcium dynamics that are incorporated into the principal cells (PCs) of the network. For tES, a new panel in the GUI was added to simulate the electric field parameters allowing the user to simulate the impact of both Direct (tDCS) and Alternating (tACS) Current Stimulation on the network dynamics.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102505"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925768","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-12-17DOI: 10.1016/j.softx.2025.102484
Saad Noufel, Nadir Maaroufi, Mehdi Najib, Mohamed Bakhouya
Image inpainting, the process of reconstructing missing or damaged regions in images, remains a critical challenge in computer vision with applications spanning medical imaging, remote sensing, and digital heritage preservation. While data-driven approaches dominate current research, model-driven methods retain significant value in scenarios with limited training data or specialized domain requirements. This paper presents HySim-IRIS, a hybrid similarity interactive restoration and inpainting suite, as a comprehensive GUI-based image inpainting application. The software features a novel hybrid similarity measure combining Chebyshev and Minkowski distances for patch-based inpainting, alongside a modern Qt-based interface with built-in mask editing tools, exhaustive parameter search capabilities, and comprehensive performance analytics. The application provides both CPU and GPU-accelerated implementations, with the latter achieving up to 20 speedup for high-resolution images.
{"title":"HySim-IRIS: Hybrid similarity interactive restoration and inpainting suite","authors":"Saad Noufel, Nadir Maaroufi, Mehdi Najib, Mohamed Bakhouya","doi":"10.1016/j.softx.2025.102484","DOIUrl":"10.1016/j.softx.2025.102484","url":null,"abstract":"<div><div>Image inpainting, the process of reconstructing missing or damaged regions in images, remains a critical challenge in computer vision with applications spanning medical imaging, remote sensing, and digital heritage preservation. While data-driven approaches dominate current research, model-driven methods retain significant value in scenarios with limited training data or specialized domain requirements. This paper presents HySim-IRIS, a hybrid similarity interactive restoration and inpainting suite, as a comprehensive GUI-based image inpainting application. The software features a novel hybrid similarity measure combining Chebyshev and Minkowski distances for patch-based inpainting, alongside a modern Qt-based interface with built-in mask editing tools, exhaustive parameter search capabilities, and comprehensive performance analytics. The application provides both CPU and GPU-accelerated implementations, with the latter achieving up to 20<span><math><mo>×</mo></math></span> speedup for high-resolution images.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102484"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145797944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2025-12-09DOI: 10.1016/j.softx.2025.102473
José Hugo Barrón-Zambrano , Marco Aurelio Nuño-Maganda , Melchor Hernández-Díaz , José de Jesús Rangel-Magdaleno , Yahir Hernández-Mier
Education in Science, Technology, Engineering, Arts, and Mathematics (STEAM) is crucial for developing essential skills in today’s society. A key issue for researchers in the Education and Behavioral Sciences (EBS) fields is to assess the evolution of Computational Thinking (CT) in learners through the use of educational robotics, which is a powerful tool that enhances learning by allowing students to apply theoretical knowledge to real-world scenarios. In this article, we propose a 2D-3D virtual and physical robotic platform for STEM/STEAM education, enabling users to interact with a low-cost line-following educational robotic platform, equipped with an onboard computer, sensors, and actuators. The platform is user-programmable and integrates the ROS operating system to define the robot’s movement and path, as well as to visualize the robot’s movement in real-time. The platform is also accessible to educators and the general public for exploratory and pedagogical use. We report results related to the application of the competent Computational Thinking Test (cCTt) instrument to a small group of students, which may be of particular relevance to the Education and Behavioral Sciences (EBS) community for validating the acquisition of CT skills through the proposed platform for larger learner groups.
{"title":"Mobile2D-3D-RoboticSim: A robotic platform for computational thinking assessment in STEM and STEAM education","authors":"José Hugo Barrón-Zambrano , Marco Aurelio Nuño-Maganda , Melchor Hernández-Díaz , José de Jesús Rangel-Magdaleno , Yahir Hernández-Mier","doi":"10.1016/j.softx.2025.102473","DOIUrl":"10.1016/j.softx.2025.102473","url":null,"abstract":"<div><div>Education in Science, Technology, Engineering, Arts, and Mathematics (STEAM) is crucial for developing essential skills in today’s society. A key issue for researchers in the Education and Behavioral Sciences (EBS) fields is to assess the evolution of Computational Thinking (CT) in learners through the use of educational robotics, which is a powerful tool that enhances learning by allowing students to apply theoretical knowledge to real-world scenarios. In this article, we propose a 2D-3D virtual and physical robotic platform for STEM/STEAM education, enabling users to interact with a low-cost line-following educational robotic platform, equipped with an onboard computer, sensors, and actuators. The platform is user-programmable and integrates the ROS operating system to define the robot’s movement and path, as well as to visualize the robot’s movement in real-time. The platform is also accessible to educators and the general public for exploratory and pedagogical use. We report results related to the application of the competent Computational Thinking Test (cCTt) instrument to a small group of students, which may be of particular relevance to the Education and Behavioral Sciences (EBS) community for validating the acquisition of CT skills through the proposed platform for larger learner groups.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102473"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145748538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2026-02-05DOI: 10.1016/j.softx.2026.102551
Marta Muñoz-Muñoz, Christian Luna, Juan A. Lara, C Romero
The prediction and prevention of students at risk of dropout are two of the most important challenges in the educational domain. Although some commercial predictive tools support at-risk estimation and provide explanations of the associated factors, none of them offer recommendations to address or reverse potential dropout cases. This paper proposes Dropout Insight as a prescriptive web-based interactive tool that automates the entire data-mining process to suggest specific decisions. It supports the loading and processing of student data, the selection of the best predictive model, and the visualization of results through interpretation techniques based on explainers. The tool provides a clear and visually intuitive interface that enables users to explore risk factors and simulate alternative scenarios, including instructors and other stakeholders, without prior knowledge of data mining. It offers not only traditional individual counterfactual explanations, but also novel group counterfactuals, which generate hypothetical clusters or groups of students with similar behavioral profiles. These groups help recover the largest possible number of at-risk students with less effort and cost by offering a single, shared recommendation for intervention. By integrating automated prediction tools with visual, explainable artificial intelligence methods and counterfactual reasoning, the tool becomes a highly valuable and innovative resource to support pedagogical decision-making and guide proactive educational policies aimed at preventing dropout.
{"title":"Dropout insight: Educational risk dashboard with counterfactual explanations","authors":"Marta Muñoz-Muñoz, Christian Luna, Juan A. Lara, C Romero","doi":"10.1016/j.softx.2026.102551","DOIUrl":"10.1016/j.softx.2026.102551","url":null,"abstract":"<div><div>The prediction and prevention of students at risk of dropout are two of the most important challenges in the educational domain. Although some commercial predictive tools support at-risk estimation and provide explanations of the associated factors, none of them offer recommendations to address or reverse potential dropout cases. This paper proposes Dropout Insight as a prescriptive web-based interactive tool that automates the entire data-mining process to suggest specific decisions. It supports the loading and processing of student data, the selection of the best predictive model, and the visualization of results through interpretation techniques based on explainers. The tool provides a clear and visually intuitive interface that enables users to explore risk factors and simulate alternative scenarios, including instructors and other stakeholders, without prior knowledge of data mining. It offers not only traditional individual counterfactual explanations, but also novel group counterfactuals, which generate hypothetical clusters or groups of students with similar behavioral profiles. These groups help recover the largest possible number of at-risk students with less effort and cost by offering a single, shared recommendation for intervention. By integrating automated prediction tools with visual, explainable artificial intelligence methods and counterfactual reasoning, the tool becomes a highly valuable and innovative resource to support pedagogical decision-making and guide proactive educational policies aimed at preventing dropout.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102551"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146187366","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-01Epub Date: 2026-01-06DOI: 10.1016/j.softx.2025.102495
Majid Mohammadhosseinzadeh, Hossein Ghorbani-Menghari, Ji Hoon Kim
This study introduces an integrated workflow for simulating powder compaction through a hybrid discrete element method (DEM) and multi-particle finite element method (MPFEM) approach. PyCompact, by integrating FreeCAD, LIGGGHTS, ParaView, LS-PrePost, and OpenRadioss with two in-house Python scripts for automated data translation and mesh generation, the framework covers the full simulation cycle: geometric modelling, particle generation, finite element meshing, compaction analysis, and visualization. The workflow was validated using experimental data from two Fe-Si-Al-P powders with distinct particle size distributions. Results demonstrated a maximum relative density deviation of only 2.5 % compared to experiments, matching ABAQUS predictions. This work introduces the first validated DEM-MPFEM framework that eliminates licensing barriers for the core simulation steps, enhances reproducibility, and adapts to various powder compaction applications in academic and industrial settings.
{"title":"PyCompact: An integrated workflow for discrete element method–multi-particle finite element method for powder compaction simulation","authors":"Majid Mohammadhosseinzadeh, Hossein Ghorbani-Menghari, Ji Hoon Kim","doi":"10.1016/j.softx.2025.102495","DOIUrl":"10.1016/j.softx.2025.102495","url":null,"abstract":"<div><div>This study introduces an integrated workflow for simulating powder compaction through a hybrid discrete element method (DEM) and multi-particle finite element method (MPFEM) approach. PyCompact, by integrating FreeCAD, LIGGGHTS, ParaView, LS-PrePost, and OpenRadioss with two in-house Python scripts for automated data translation and mesh generation, the framework covers the full simulation cycle: geometric modelling, particle generation, finite element meshing, compaction analysis, and visualization. The workflow was validated using experimental data from two Fe-Si-Al-P powders with distinct particle size distributions. Results demonstrated a maximum relative density deviation of only 2.5 % compared to experiments, matching ABAQUS predictions. This work introduces the first validated DEM-MPFEM framework that eliminates licensing barriers for the core simulation steps, enhances reproducibility, and adapts to various powder compaction applications in academic and industrial settings.</div></div>","PeriodicalId":21905,"journal":{"name":"SoftwareX","volume":"33 ","pages":"Article 102495"},"PeriodicalIF":2.4,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号