IEEE Access · 2024 · 127 citations · 134 references
ComputingComputer ScienceArtificial IntelligenceAutomationMachine LearningActive LearningEthics Of Automated Decision-makingHuman-machine InterfacesIndustrial AutomationReinforcement LearningActive Learning With Human FeedbackAutomated Machine LearningDomain-specific ExpertiseHuman-automation CollaborationHuman-in-the-loop LearningPrediction ModelsUser EngagementUser Expertise Modeling
Machine learning is widely used in Industry 4.0 for tasks such as image, language, financial, and biomedical analysis, yet it struggles with decision‑making, prompting interest in human‑in‑the‑loop systems that face methodological, technical, and ethical challenges. This paper aims to explore how incorporating human expertise into ML automation can improve prediction accuracy and to examine the methodologies, challenges, and opportunities of HITL systems. The authors review HITL system methodologies, identify key issues such as data quality, bias, and user engagement, and evaluate performance‑enhancing approaches like active learning, iterative ML, and reinforcement learning. The study demonstrates that HITL systems can improve decision‑making accountability and transparency, offering benefits for researchers, practitioners, and policymakers.
Machine learning (ML) has become a popular technique for various automation tasks in the era of Industry 4.0, such as the analysis and synthesis of visual data such as images and videos, natural language and speech, financial data, and biomedical applications. However, ML-based automation techniques are facing difficulties like decision-making, thus incorporating user expertise into the system might be advantageous. The goal of adding human domain expertise with ML-based automation is to provide more accurate prediction models. Human-in-the-loop (HITL) systems that integrate human expertise with ML algorithms are becoming more and more common in various industries. However, there are a number of methodological, technical, and ethical difficulties with the development and application of HITL systems. This paper aims to explore the methodologies, challenges, and opportunities associated with HITL systems implementations.We also discuss a number of issues that must be resolved for HITL systems to be effective, including data quality, bias, and user engagement. Besides, we also explored several approaches that can be utilized to enhance the performance of HITL systems, such as active learning (AL), iterative ML, and reinforcement learning, as well as the current state of the art in HITL systems.We also selectively highlighted the advantages of HITL systems, such as their potential to increase decision-making process accountability and transparency by utilizing human experience to improve ML decision-making capability. The paper will be very useful for researchers, practitioners, and policymakers.
134
DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · 2023 · 76.3K citations · Full text
Evaluating the Quality of Medical Care
Avedis Donabedian · Milbank Quarterly · 2005 · 6.1K citations
Public Affairs, Health Services Administration, Quality Management +12