This study is aimed to explore parent-child interaction in a game-based learning activity. The participants were 10 parent-child dyads. The children were 10.4 years old in average and the parents were 45.4 years old. ...
详细信息
Traditional evaluation metrics like ROUGE compare lexical overlap between the reference and generated summaries without taking argumentative structure into account, which is important for legal summaries. In this pape...
详细信息
ISBN:
(纸本)9781643684727;9781643684734
Traditional evaluation metrics like ROUGE compare lexical overlap between the reference and generated summaries without taking argumentative structure into account, which is important for legal summaries. In this paper, we propose a novel legal summarization evaluation framework that utilizes GPT-4 to generate a set of question-answer pairs that cover main points and information in the reference summary. GPT-4 is then used to produce answers based on the generated summary for the questions from the reference summary. Finally, GPT-4 grades the answers from the reference summary and the generated summary. We examined the correlation between GPT-4 grading and human grading. The results suggest that this question-answering approach with GPT-4 can be a useful tool for gauging the quality of the summary.
Crime scene investigation serves as an important source of information for law enforcement officers when running a case. Effective leads are uncovered based on the knowledge, technique, and experience of the crime sce...
详细信息
Crime scene investigation serves as an important source of information for law enforcement officers when running a case. Effective leads are uncovered based on the knowledge, technique, and experience of the crime scene investigator. This study utilized eye-tracking technology to understand different decision makers'visual behaviors and decision-making process when studying a crime scene picture. A total of 42 college students majored in crime investigation were recruited as the subjects of the study. Eye movement data were collected and analyzed using t-test and ANOVA on eye-tracking measures of areas of interest (AOI) for visual attention distributions as well as lag sequence analysis (LSA), heat map, and scan path observations for visual transfer patterns. The results showed that participants who made the correct decision were faster in identifying the key target. Regardless of whether they made the correct decision or not, their visual attention was fixated on the answer option of their choosing. Additionally, the subjects' attention must be shifted to the key target in order to make the correct decision. This study successfully applied the eye-tracking method to explore the investigation and decision-making process of inspecting a crime scene picture. Future studies can apply the eye-tracking method to deeply understand the process of information processing in investigating criminal scene pictures and provide more suggestions for professional training curriculum and instruction.
The COVID-19 crisis affected the world globally in 2020 and caused school closure around the world, which promptly embraced online learning and raised ethical awareness for teachers and students. Online learning that ...
详细信息
The COVID-19 crisis affected the world globally in 2020 and caused school closure around the world, which promptly embraced online learning and raised ethical awareness for teachers and students. Online learning that is based on continuously gaining relevance in educational institutions, is challenged by lack of documented ethics that maintain rights and encourage applying online learning. Unfortunately, online learning ethics are still obscure at this moment. This study aims to propose online code of ethics for Palestinian schools. Applying the mixed methods approach: A focus group with teachers and a questionnaire that was built to explore the teachers' degree of agreement on the proposed online learning ethics. The code was finally being evaluated by online learning experts who have extensive expertise in the field. Results revealed principles of online learning ethics that were highly accepted by school teachers. The researchers recommended conducting future research on the impact of this code on learning and teaching.
The Next Generation Science Standards (NGSS) have spurred renewed interest in the epistemologies that students adopt as they engage in science practices. One framework for characterizing students' epistemologies i...
详细信息
The Next Generation Science Standards (NGSS) have spurred renewed interest in the epistemologies that students adopt as they engage in science practices. One framework for characterizing students' epistemologies is the epistemologies in practice framework (Berland et al. (2016), Journal of Research in Science Teaching, 53(7), 1082-1112), which focuses on students' meaningful use of four epistemic considerations: Nature, Generality, Justification, and Audience. To date, research based on the framework has primarily examined students' use of the epistemic considerations in the context of diagrammatic modeling. However, with computational technologies becoming more prevalent in science classrooms, the framework could be applied to investigate students' engagement in computational modeling. Moreover, computational modeling could be particularly beneficial to a fast-growing population of multilingual learners (MLs) in the U.S. K-12 context, who benefit from leveraging multiple meaning-making resources (e.g., code, dynamic visualization). This study examined MLs' meaningful use of four epistemic considerations in the context of computational modeling in an elementary science classroom. Fifth-grade MLs (N = 11) participated in two interviews about computational models they had developed as part of two NGSS-designed instructional units that integrated computational modeling (in addition to other model types). Findings indicated that, while students used all four epistemic considerations across the interviews, some considerations (Nature and Generality) were used more frequently than others (Justification and Audience). Beyond diagrammatic modeling, computational modeling offered unique affordances for MLs to meaningfully use the considerations as well as to communicate this use, though not without some emergent challenges. Overall, this study highlights the promise of computational modeling for providing a rich sense-making and meaning-making context for MLs to use epistemi
Most machine learning (ML) methods produce predictions that are hard or impossible to understand. The black box nature of predictive models obscures potential learning bias and makes it difficult to recognize and trac...
Most machine learning (ML) methods produce predictions that are hard or impossible to understand. The black box nature of predictive models obscures potential learning bias and makes it difficult to recognize and trace problems. Moreover, the inability to rationalize model decisions causes reluctance to accept predictions for experimental design. For ML, limited trust in predictions presents a substantial problem and continues to limit its impact in interdisciplinary research, including early-phase drug discovery. As a desirable remedy, approaches from explainable artificial intelligence (XAI) are increasingly applied to shed light on the ML black box and help to rationalize predictions. Among these is the concept of counterfactuals (CFs), which are best understood as test cases with small modifications yielding opposing prediction outcomes (such as different class labels in object classification). For ML applications in medicinal chemistry, for example, compound activity predictions, CFs are particularly intuitive because these hypothetical molecules enable immediate comparisons with actual test compounds that do not require expert ML knowledge and are accessible to practicing chemists. Such comparisons often reveal structural moieties in compounds that determine their predictions and can be further investigated. Herein, we adapt and extend a recently introduced concept for the systematic generation of molecular CFs to multi-task predictions of different classes of protein kinase inhibitors, analyze CFs in detail, rationalize the origins of CF formation in multi-task modeling, and present exemplary explanations of predictions. For a kinase inhibitor correctly predicted with a multi-task machine learning model (shown on an orange background), counterfactuals with small chemical changes (shown in red) were generated that were predicted to be active against other kinases.
作者：
An, Joon-YongKim, YujinKorea Univ
Coll Hlth Sci Sch Biosyst & Biomed Sci Seoul South Korea Korea Univ
Dept Integrated Biomed & Life Sci Seoul South Korea Korea Univ
L HOPE Program Community Based Total Learning Hlth Seoul South Korea
Human germline mutations—genetic changes in an egg or sperm—are inherited, present in every cell, and play roles in early development (1), whereas somatic mutations arise postzygotically and may or may not affect de...
Human germline mutations—genetic changes in an egg or sperm—are inherited, present in every cell, and play roles in early development (1), whereas somatic mutations arise postzygotically and may or may not affect developmental trajectories. Neurons accumulate hundreds to thousands of somatic mutations throughout development, with distinct mutational processes and rates occurring at various developmental stages (2). Whole-genome sequencing studies characterizing somatic mosaicism in early human brain development indicate that the mutation rate is relatively low during early pregastrulation (the first 2 weeks after fertilization) but increases substantially during late neurogenesis in the prenatal period (starting at 22 weeks of gestation), primarily owing to oxidative damage (3). On page 217 of this issue, Maury et al. (4) report that somatic mutations in the brains of individuals with schizophrenia occurred during neogenesis . This suggests that intrauterine factors might influence mutational mechanisms and brain development in utero.
Many countries have adopted integrative approaches to science, technology, engineering, and mathematics (STEM) education. As preservice teachers with majors in mathematics and science are prospective targets for addre...
详细信息
Many countries have adopted integrative approaches to science, technology, engineering, and mathematics (STEM) education. As preservice teachers with majors in mathematics and science are prospective targets for addressing the demand for STEM education, they are expected to develop professional identities as STEM teachers, in addition to professional identities as general and disciplinary teachers. However, research investigating the extent to which preservice teachers with these majors identify as STEM teachers is scant. Moreover, the disciplinary specificity of preservice teachers' professional identities remains an issue in literature. This study involves survey research examining the degree to which preservice teachers with majors in mathematics (42), general science (44), and advanced science (47) identify as general, disciplinary, and STEM teachers. We collected data using Likert-type questionnaires focusing on three components-motivation, self-efficacy, and self-image-as proxies of professional identity. We analyzed the data using Friedman tests with Conover's post-hoc tests. The results show that preservice teachers similarly self-identify as general and disciplinary teachers regardless of their majors but see themselves significantly less as STEM teachers. When comparing across the components, they imagine themselves significantly less as STEM teachers than they were motivated and self-efficacious to become STEM teachers. The results suggest that the disciplinary specificity of preservice teachers' professional identities largely manifests in the context of STEM education. The siloed approach to teacher education may not adequately prepare preservice teachers to develop professional identities as STEM teachers. Effective and ongoing support is needed for preservice teachers to identify as STEM teachers.
暂无评论