Was ist die Webseite?

MAKE MDPI

Startseite
Schweiz
Basel
MAKE MDPI

Machine Learning and Knowledge Extraction (ISSN 2504-4990) is a peer-reviewed, scholarly open access

31/07/2025

🧪 A Simple Yet Powerful Hybrid Machine Learning Approach to Aid Decision-Making in Laboratory Experiments 🤖📊

We demonstrate that combining Ordinary Least Squares, Gaussian Processes, Expected Improvement, and K-Means clustering can significantly reduce experimental load—replicating expert-level performance with just 25 virtual trials. ⚗️🔍 This hybrid machine learning pipeline offers a compelling proof of concept for integrating AI into active laboratory experimentation. 💡🧬

Authors: Bernardo Campos Diocaretz, Ágota Tűzesi, Ph.D and Andrei Herdean

📖 Read more at: https://www.mdpi.com/2504-4990/7/3/60 #

31/07/2025

📉 A Novel Approach to Company Bankruptcy Prediction Using Convolutional Neural Networks and Generative Adversarial Networks

Discover how D'Ercole & Me leverage the power of Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs) to enhance bankruptcy prediction. By transforming financial data into images and enriching it with synthetic data, their model boosts predictive accuracy, a game-changer for financial risk assessment.

👥 Authors: Alessia D’Ercole and Gianluigi Me

📖 Read more: https://www.mdpi.com/2504-4990/7/3/63 #

30/07/2025

🧠 Explaining Deep Q-Learning Experience Replay with SHapley Additive exPlanations

Unravelling the opacity of Deep Reinforcement Learning (DRL), our study delves into optimizing resource use. Contrary to the trend of increasing Experience Replay capacity, we intentionally reduce it, discovering a path to resource-efficient DRL.

🎮 Across 20 Atari games and capacities from millions of transitions to thousands, we show that reducing capacity from ten thousand transitions to five thousand doesn't significantly impact rewards. To enhance interpretability, we visualize Experience Replay with the Deep SHAP Explainer, providing transparent explanations for agent decisions.

🔍 Our findings advocate for a cautious reduction in Experience Replay, emphasizing interpretable decision explanations for efficient DRL.

👥 Authors: Rob Sullivan and Luca Longo

🔬 Artificial Intelligence & Cognitive Load Research (AICL) Lab: https://lnkd.in/d7Ejp9qw

🔗 Read more: https://www.mdpi.com/2504-4990/5/4/72

30/07/2025

🚨 Trending in MAKE 🚨

This article is gaining momentum fast, with a notable citation boost just last week! 📈

📄 Using Machine Learning with Eye-Tracking Data to Predict if a Recruiter Will Approve a Resume

Eye-tracking and machine learning predicted how recruiters would treat computer science resumes.

👥 Authors: Angel Pina, Corbin Petersheim, Josh Cherian, Joanna Lahey, Gerianne Alexander and Tracy Hammond

📚 Read the full article: https://www.mdpi.com/2504-4990/5/3/38 #

29/07/2025

📈 Trending Article – Rapid Citation Growth

📄 Augmenting Deep Neural Networks with Symbolic Educational Knowledge: Towards Trustworthy and Interpretable AI for Education ✏️🤖

Hybrid human-AI intelligence for education enhances deep neural networks with symbolic educational knowledge to build more trustworthy, bias-resistant, responsible, and interpretable AI for education.

👥 Authors: Danial Hooshyar, Roger Azevedo and Yeongwook Yang

🔗 Read the full article: https://www.mdpi.com/2504-4990/6/1/28

29/07/2025

📢 Trending Article – Rapid Citation Growth

📄 Painting the Black Box White: Experimental Findings from Applying XAI to an ECG Reading Setting

XAI on ECG: study of 44 cardiologists reveals how explanations affect trust, bias and automation reliance, warning against superficial “white-box” transparency.

📈 One of the most rapidly cited papers from last week.

🔗 Read the full article: https://www.mdpi.com/2504-4990/5/1/17

27/07/2025

📢 Upcoming Webinar | MAKE Journal
🎙️ Topics Webinar: AI in Medical Imaging
📅 4 August 2025
🕘 09:00 (CEST)

🔗 Join us for a deep dive into the role of Artificial Intelligence in transforming medical imaging technologies and clinical practice.

Discover cutting-edge research, real-world applications, and expert insights from scholars driving innovation at the intersection of AI and healthcare.

🔍 Do not miss this opportunity to engage with the scientific community and stay updated on emerging trends in AI-powered diagnostics.

📌 Read more: https://www.mdpi.com/2624-800X/5/3/49

17/07/2025

🧠🏥 Featured Paper from MAKE

AI Advances in ICU with an Emphasis on Sepsis Prediction: An Overview

Sepsis remains a leading cause of mortality in ICUs. This paper provides a comprehensive overview of how AI tools are being applied to improve early detection, patient outcomes, and resource management in critical care settings.

🗂️ A valuable reference for researchers and clinicians working at the intersection of AI and critical care.

📖 Read it in MAKE: https://brnw.ch/21wUdMt

17/07/2025

🧠💬 New Special Issue in MAKE: Language Acquisition and Understanding

Can AI truly understand language, or just predict it?

This Special Issue delves into the gap between statistical pattern matching and genuine comprehension, exploring how AI systems learn and understand language beyond scale.

👥 Guest Editors:
Dr. Michal Ptaszynski | Dr. Rafal Rzepka | Prof. Dr. Masaharu Yoshioka

🔗https://brnw.ch/21wUdM4

11/07/2025

🌍 Benchmarking with a Language Model Initial Selection for Text Classification Tasks

How can we make AI benchmarking more sustainable? This study introduces LMDFit, an innovative pre-selection approach for language models that cuts down computational load and reduces carbon emissions by 37%, without compromising performance.

Explore how smarter benchmarking can help create greener AI:
🔗 https://brnw.ch/21wU4fg

11/07/2025

📰 A Comparative Analysis of European Media Coverage of the Israel–Gaza War Using Hesitant Fuzzy Linguistic Term Sets

This study applies Hesitant Fuzzy Linguistic Term Sets combined with sentiment analysis to explore how media across European countries cover the Israel–Gaza war. By capturing nuanced perceptions and attitudes, the research offers deeper insights into diverse national viewpoints on this sensitive issue.

🔗 Read the full article: https://brnw.ch/21wU49m

10/07/2025

🩺 Analyzing the Impact of Data Augmentation on the Explainability of Deep Learning-Based Medical Image Classification

This study explores how different data augmentation techniques affect the explainability of deep learning models for medical image classification using LIME and SHAP. Key insights reveal that TrivialAugment and Flipping + Cropping improve explainability, with LIME showing better coherence than SHAP.

🔗 Read the full article: https://brnw.ch/21wU2yC

Adresse

Basel

Webseite

https://www.mdpi.com/journal/make

Benachrichtigungen

Lassen Sie sich von uns eine E-Mail senden und seien Sie der erste der Neuigkeiten und Aktionen von MAKE MDPI erfährt. Ihre E-Mail-Adresse wird nicht für andere Zwecke verwendet und Sie können sich jederzeit abmelden.