With the proliferation of misinformation on the web, automatic misinformation detection methods are becoming an increasingly important subject of study. Large language models have produced the best results among content-based methods, which rely on the text of the article rather than the metadata or network features. However, finetuning such a model requires significant training data, which has led to the automatic creation of large-scale misinformation detection datasets. In these datasets, articles are not labelled directly. Rather, each news site is labelled for reliability by an established fact-checking organisation and every article is subsequently assigned the corresponding label based on the reliability score of the news source in question. A recent paper has explored the biases present in one such dataset, NELA-GT-2018, and shown that the models are at least partly learning the stylistic and other features of different news sources rather than the features of unreliable news. We confirm a part of their findings. Apart from studying the characteristics and potential biases of the datasets, we also find it important to examine in what way the model architecture influences the results. We therefore explore which text features or combinations of features are learned by models based on contextual word embeddings as opposed to basic bag-of-words models. To elucidate this, we perform extensive error analysis aided by the SHAP post-hoc explanation technique on a debiased portion of the dataset. We validate the explanation technique on our inherently interpretable baseline model.
BACKGROUND: The quality standards of the Dutch Society of Intensive Care require monitoring of the satisfaction of patient's relatives with respect to care. Currently, no suitable instrument is available in the Netherlands to measure this. This study describes the development and psychometric evaluation of the questionnaire-based Consumer Quality Index 'Relatives in Intensive Care Unit' (CQI 'R-ICU'). The CQI 'R-ICU' measures the perceived quality of care from the perspective of patients' relatives, and identifies aspects of care that need improvement.METHODS: The CQI 'R-ICU' was developed using a mixed method design. Items were based on quality of care aspects from earlier studies and from focus group interviews with patients' relatives. The time period for the data collection of the psychometric evaluation was from October 2011 until July 2012. Relatives of adult intensive care patients in one university hospital and five general hospitals in the Netherlands were approached to participate. Psychometric evaluation included item analysis, inter-item analysis, and factor analysis.RESULTS: Twelve aspects were noted as being indicators of quality of care, and were subsequently selected for the questionnaire's vocabulary. The response rate of patients' relatives was 81% (n = 455). Quality of care was represented by two clusters, each showing a high reliability: 'Communication' (α = .80) and 'Participation' (α = .84). Relatives ranked the following aspects for quality of care as most important: no conflicting information, information from doctors and nurses is comprehensive, and health professionals take patients' relatives seriously. The least important care aspects were: need for contact with peers, nuisance, and contact with a spiritual counsellor. Aspects that needed the most urgent improvement (highest quality improvement scores) were: information about how relatives can contribute to the care of the patient, information about the use of meal-facilities in the hospital, and involvement in decision-making on the medical treatment of the patient.CONCLUSIONS: The CQI 'R-ICU' evaluates quality of care from the perspective of relatives of intensive care patients and provides practical information for quality assurance and improvement programs. The development and psychometric evaluation of the CQI 'R-ICU' led to a draft questionnaire, sufficient to justify further research into the reliability, validity, and the discriminative power of the questionnaire.
MULTIFILE
The design of health game rewards for preadolescents Videogames are a promising strategy for child health interventions, but their impact can vary depending on the game mechanics used. This study investigated achievement-based ‘rewards’ and their design among preadolescents (8-12 years) to assess their effect and explain how they work. In a 2 (game reward achievement system: social vs. personal) x 2 (game reward context: in-game vs. out-game) between-subjects design, 178 children were randomly assigned to one of four conditions. Findings indicated that a ‘personal’ achievement system (showing one’s own high scores) led to more attention and less frustration than a ‘social’ achievement system (showing also high scores of others) which, in turn, increased children’s motivation to make healthy food choices. Furthermore, ‘out’-game rewards (tangible stickers allocated outside the game environment) were liked more than ‘in’-game rewards (virtual stickers allocated in the game environment), leading to greater satisfaction and, in turn, a higher motivation to make healthy food choices.
LINK
Developing a framework that integrates Advanced Language Models into the qualitative research process.Qualitative research, vital for understanding complex phenomena, is often limited by labour-intensive data collection, transcription, and analysis processes. This hinders scalability, accessibility, and efficiency in both academic and industry contexts. As a result, insights are often delayed or incomplete, impacting decision-making, policy development, and innovation. The lack of tools to enhance accuracy and reduce human error exacerbates these challenges, particularly for projects requiring large datasets or quick iterations. Addressing these inefficiencies through AI-driven solutions like AIDA can empower researchers, enhance outcomes, and make qualitative research more inclusive, impactful, and efficient.The AIDA project enhances qualitative research by integrating AI technologies to streamline transcription, coding, and analysis processes. This innovation enables researchers to analyse larger datasets with greater efficiency and accuracy, providing faster and more comprehensive insights. By reducing manual effort and human error, AIDA empowers organisations to make informed decisions and implement evidence-based policies more effectively. Its scalability supports diverse societal and industry applications, from healthcare to market research, fostering innovation and addressing complex challenges. Ultimately, AIDA contributes to improving research quality, accessibility, and societal relevance, driving advancements across multiple sectors.
The textile and clothing sector belongs to the world’s biggest economic activities. Producing textiles is highly energy-, water- and chemical-intensive and consequently the textile industry has a strong impact on environment and is regarded as the second greatest polluter of clean water. The European textile industry has taken significant steps taken in developing sustainable manufacturing processes and materials for example in water treatment and the development of biobased and recycled fibres. However, the large amount of harmful and toxic chemicals necessary, especially the synthetic colourants, i.e. the pigments and dyes used to colour the textile fibres and fabrics remains a serious concern. The limited range of alternative natural colourants that is available often fail the desired intensity and light stability and also are not provided at the affordable cost . The industrial partners and the branch organisations Modint and Contactgroep Textiel are actively searching for sustainable alternatives and have approached Avans to assist in the development of the colourants which led to the project Beauti-Fully Biobased Fibres project proposal. The objective of the Beauti-Fully Biobased Fibres project is to develop sustainable, renewable colourants with improved light fastness and colour intensity for colouration of (biobased) man-made textile fibres Avans University of Applied Science, Zuyd University of Applied Sciences, Wageningen University & Research, Maastricht University and representatives from the textile industry will actively collaborate in the project. Specific approaches have been identified which build on knowledge developed by the knowledge partners in earlier projects. These will now be used for designing sustainable, renewable colourants with the improved quality aspects of light fastness and intensity as required in the textile industry. The selected approaches include refining natural extracts, encapsulation and novel chemical modification of nano-particle surfaces with chromophores.
Phosphorus is an essential element for life, whether in the agricultural sector or in the chemical industry to make products such as flame retardants and batteries. Almost all the phosphorus we use are mined from phosphate rocks. Since Europe scarcely has any mine, we therefore depend on imported phosphate, which poses a risk of supply. To that effect, Europe has listed phosphate as one of its main critical raw materials. This creates a need for the search for alternative sources of phosphate such as wastewater, since most of the phosphate we use end up in our wastewater. Additionally, the direct discharge of wastewater with high concentration of phosphorus (typically > 50 ppb phosphorus) creates a range of environmental problems such as eutrophication . In this context, the Dutch start-up company, SusPhos, created a process to produce biobased flame retardants using phosphorus recovered from municipal wastewater. Flame retardants are often used in textiles, furniture, electronics, construction materials, to mention a few. They are important for safety reasons since they can help prevent or spread fires. Currently, almost all the phosphate flame retardants in the market are obtained from phosphate rocks, but SusPhos is changing this paradigm by being the first company to produce phosphate flame retardants from waste. The process developed by SusPhos to upcycle phosphate-rich streams to high-quality flame retardant can be considered to be in the TRL 5. The company seeks to move further to a TRL 7 via building and operating a demo-scale plant in 2021/2022. BioFlame proposes a collaboration between a SME (SusPhos), a ZZP (Willem Schipper Consultancy) and HBO institute group (Water Technology, NHL Stenden) to expand the available expertise and generate the necessary infrastructure to tackle this transition challenge.
