Among patients with Coronavirus disease (COVID-19), the ability to identify patients at risk for deterioration during their hospital stay is essential for effective patient allocation and management. To predict patient risk for critical COVID-19 based on status at admission using machine-learning models. Retrospective study based on a database of tertiary medical center with designated departments for patients with COVID-19. Patients with severe COVID-19 at admission, based on low oxygen saturation, low partial arterial oxygen pressure, were excluded. The primary outcome was risk for critical disease, defined as mechanical ventilation, multi-organ failure, admission to the ICU, and/or death. Three different machine-learning models were used to predict patient deterioration and compared to currently suggested predictors and to the APACHEII risk-prediction score. Among 6995 patients evaluated, 162 were hospitalized with non-severe COVID-19, of them, 25 (15.4%) patients deteriorated to critical COVID-19. Machine-learning models outperformed the all other parameters, including the APACHE II score (ROC AUC of 0.92 vs. 0.79, respectively), reaching 88.0% sensitivity, 92.7% specificity and 92.0% accuracy in predicting critical COVID-19. The most contributory variables to the models were APACHE II score, white blood cell count, time from symptoms to admission, oxygen saturation and blood lymphocytes count. Machine-learning models demonstrated high efficacy in predicting critical COVID-19 compared to the most efficacious tools available. Hence, artificial intelligence may be applied for accurate risk prediction of patients with COVID-19, to optimize patients triage and in-hospital allocation, better prioritization of medical resources and improved overall management of the COVID-19 pandemic.

Journal of the Learning SciencesPublication details, including instructions for authors and subscription information:In this article, we elaborate methodologies to study construction of knowledge in argumentative activities. For this purpose, we report on a quasi-empirical study on construction of knowledge through successive argumentative activities on a controversial issue. A group of 120 fifth grade students participated in successive argumentative activities; some activities involved individuals and some involved collectives. According to a first methodology, construction of knowledge was measured through arguments/outcomes produced. We developed tools for evaluating changes in individual and collective arguments. In the study, we showed the generally beneficial effect of argumentative activities on collective and individual arguments/outcomes. The significant discrepancies between collective and individual arguments suggested that individual students only partly internalized the collectively constructed arguments. We developed a qualitative methodology to refine this hypothesis as well as other hypotheses concerning the interpretation of the quantitative study. The integration of the quantitative and qualitative methodologies for studying argumentation helped identify several mechanisms of construction of knowledge in argumentative activities. In particular, it brought new light on the mediating role of representational tools such as Argumentative Maps or Pro-Con tables.The idea that construction of knowledge emerges from social and cultural contexts is of course not new (Vygotsky, 1986). However, adequate methodologies for evidenc-ing construction of knowledge in rich contexts are difficult to elaborate. For example, although argumentation is recognized as potentially leading to construction of knowledge, experimental studies focusing on the changes that individuals and groups undergo during and after argumentative activities are rare. In this article we attempt to contribute to the elaboration of methodologies for studying construction of knowledge in context, in argumentative activities. We show the methodologies we developed through an experimental study on construction of knowledge through successive argumentative activities. The quantitative measure of construction of knowledge relies on a research setting in which individual argument-outcomes alternated with collective argument-outcomes. The participants were 120 fifth grade students who engaged in argumentative activities on a controversial issue (whether to permit or forbid experiments on animals). Students first completed a questionnaire to express their standpoint individually. They then formed triads and engaged in argumentative talk. At this stage, triads had at their disposal short texts presenting information on the issue. At the end of the conversation, individuals completed the questionnaire again. The triads went on with their argumentative talk and displayed their arguments. In a first group (G1, N1 = 60), triads used a computerized tool, ...

No abstract

No abstract

Identifying metaphorical language-use (e.g., sweet child) is one of the challenges facing natural language processing. This paper describes three novel algorithms for automatic metaphor identification. The algorithms are variations of the same core algorithm. We evaluate the algorithms on two corpora of Reuters and the New York Times articles. The paper presents the most comprehensive study of metaphor identification in terms of scope of metaphorical phrases and annotated corpora size. Algorithms’ performance in identifying linguistic phrases as metaphorical or literal has been compared to human judgment. Overall, the algorithms outperform the state-of-the-art algorithm with 71% precision and 27% averaged improvement in prediction over the base-rate of metaphors in the corpus.