explainable ai

GROWING OYSTER MUSHROOMS: A STARTER GUIDE FOR FIRST-TIME GROWERS

A short, free introductory guide to growing oyster mushrooms (Pleurotus spp.) for first-time and small-scale growers.The guide teaches the shape of the craft rather than a recipe: what the oyster mushroom is and why it is a forgiving first crop; the two phases of cultivation — colonisation and fruiting — that organise...

Who Verifies the Verifiers? AI, Local Knowledge Production, and Credibility in Tanzania’s 2025 Election

This study examines how Factchecking AI tools and community-based credibility practices interacted during Tanzania’s 2025 presidential election. Using data from five regions, we analyzed 342 media materials, 87 interviews, field observations, and 22 distinct contested election events. The findings identify three forms...

An Efficient and Privacy-Preserving Framework for Edge AI Using Federated Learning and Homomorphic Encryption

Edge Artificial Intelligence (Edge AI) supports intelligent processing close to where data is generated. However, training machine-learning models across edge devices is difficult when devices have limited memory, processing power, battery capacity, and network bandwidth. Federated learning (FL) addresses one important...

Artificial Intelligence in Education: Impact on Human Skills, Critical Thinking, Independent Learning, and Academic Performance Subtitle: Balancing Technological Innovation with Independent Thinking and Genuine Learning

Alternative Titles The Impact of Generative Artificial Intelligence on Human Cognitive Skills and Educational Outcomes Artificial Intelligence and the Future of Human Learning: Does Technological Assistance Undermine Independent Thinking? Generative AI in Education: Examining the Relationship Between AI Dependence, Cri...

Prostate Commander – PROSTATEx validation

Preregistration of a diagnostic accuracy study (STARD 2015) of Prostate Commander, a rule-based, explainable implementation of PI-RADS 2.1 for prostate MRI (AI used only for gland segmentation; ADC from measured b-values; rule-based lesion detection and categorisation). Index test: frozen software (11 Oct 2026, SHA-256...

Structured PREreview of "HE-Guardrail: A Homomorphic Guardrail Against Jailbreak Attacks for Encrypted Large Language Model Inference"

This Zenodo record is a permanently preserved version of a Structured PREreview. You can view the complete PREreview at https://prereview.org/reviews/23289740. Does the introduction explain the objective of the research presented in the preprint? Yes The introduction makes the problem clear early on: because the server...

Towards Trustworthy Intelligent Tutoring Systems for Nursing Education: Integrating Nurse-Informed Design with a Frugal Neuro-Symbolic Architecture

Nursing education is increasingly expected to provide personalised, scalable, and clinically meaningful learning experiences while responding to educator shortages, growing student numbers, limited placement opportunities, and rapidly evolving healthcare environments. These pressures have led to the adoption of technol...

Explainable deep learning for predicting lignin removal efficiency in deep eutectic solvent-based biomass fractionation

An extended theory of planned behavior model of generative AI use intention for music composition among Chinese music students

Findings suggest that educators could support students in developing calibrated judgments of GenAI reliability, evaluating its suitability for specific composition tasks, and addressing AI-related anxiety while preserving authorship and creative agency.

Beyond Technocentrism in AI-Supported Special Education: The PAISEN Model of Teacher-Mediated Inclusive Learning

Artificial intelligence (AI) offers new possibilities for inclusive teaching and learning for students with special educational needs (SEN), yet AI-in-SEN research remains largely technocentric, often describing system capabilities without explaining how outputs become pedagogically meaningful. This conceptual paper de...

Evaluating and explaining LLM classification for automatic screening in Systematic Reviews

Evidence synthesis methods, such as systematic reviews (SR), require research teams to find, select, evaluate, and synthesize all available evidence to answer a specific research question in a reproducible and transparent, bias-reduced manner. Early in the process, teams must screen all unique titles and abstracts gath...

Provenance-Aware Hybrid Search in AI-Mediated Digital Libraries

A provenance-aware hybrid search framework for digital libraries that connects digital library content and metadata with hybrid retrieval and AI synthesis, provenance-aware interface design, and user interaction is proposed.