PQSCT: Pseudo-Siamese BERT for Concept Tagging With Both Questions and Solutions

The global outbreak of the new coronavirus epidemic has promoted the development of intelligent education and the utilization of online learning systems. In order to provide students with intelligent services such as cognitive diagnosis and personalized exercises recommendation, a fundamental task is the concept tagging for exercises, which extracts knowledge index structures and knowledge representations for exercises. Unfortunately, to the best of our knowledge, existing tagging approaches based on exercise content either ignore multiple components of exercises, or ignore that exercises may contain multiple concepts. To this end, in this paper, we present a study of concept tagging. First, we propose an improved pre-trained BERT for concept tagging with both questions and solutions (QSCT). Specifically, we design a question-solution prediction task and apply the BERT encoder to combine questions and solutions, ultimately obtaining the final exercise representation through feature augmentation. Then, to further explore the relationship between questions and solutions, we extend the QSCT to a pseudo-siamese BERT for concept tagging with both questions and solutions (PQSCT). We optimize the feature fusion strategy, which integrates five different vector features from local and global into the final exercise representation. Finally, we conduct extensive experiments on real-world datasets, which clearly demonstrate the effectiveness of our proposed models for concept tagging. IEEE

Conceptualising Circular economy performance with non-traditional valuation methods: Lessons for a post-Pandemic recovery

The circular economy (CE) has gained importance in the post-COVID-19 pandemic recovery. Businesses, while realising the CE benefits, have challenges in justifying and evaluating the CE benefits using available performance measurement tools, specifically when considering sustainability and other non-traditional benefits. Given the rising institutional pressures for environmental and social sustainability, we argue that organisations can evaluate their CE implementation performance using non-market-based environmental goods valuation methods. Further, the effectiveness of the CE performance measurement model can be enhanced to support supply chain sustainability and resilience through an ecosystem of multi-stakeholder digital technologies that include a range of emerging technologies such as blockchain technology, the internet-of-things (IoT), artificial intelligence, remote sensing, and tracking technologies. Accordingly, a CE performance measurement model (CEPMM) is conceptualised and exemplified using seven COVID-19 disruption scenarios to provide insights that can be addressed through CE practices. Analyses and implications are presented along with areas for future research.

How Media Literacy, Trust of Experts and Flu Vaccine Behaviors Associated with COVID-19 Vaccine Intentions.

PURPOSE: To assess how previous experiences and new information contributed to COVID-19 vaccine intentions. DESIGN: Online survey (N = 1264) with quality checks. SETTING: Cross-sectional U.S. survey fielded June 22-July 18, 2020. SAMPLE: U.S. residents 18+; quotas reflecting U.S. Census, limited to English speakers participating in internet panels. MEASURES: Media literacy for news content and sources, COVID-19 knowledge; perceived usefulness of health experts; if received flu vaccine in past 12 months; vaccine willingness scale; demographics. ANALYSIS: Structural equation modelling. RESULTS: Perceived usefulness of health experts (b = .422, P < .001) and media literacy (b = .162, P < .003) predicted most variance in vaccine intentions (R-squared=31.5%). A significant interaction (b = .163, P < .001) between knowledge (b = -.132, P = .052) and getting flu shot (b = .185, P < .001) predicted additional 3.5% of the variance in future vaccine intentions. An increase in knowledge of COVID-19 associated with a decrease in vaccine intention among those declining the flu shot. CONCLUSION: The interaction result suggests COVID-19 knowledge had a positive association with vaccine intention for flu shot recipients but a counter-productive association for those declining it. Media literacy and trust in health experts provided strong counterbalancing influences. Survey-based findings are correlational; thus, predictions are based on theory. Future research should study these relationships with panel data or experimental designs.

A Novel Graph Indexing Approach for Uncovering Potential COVID-19 Transmission Clusters

The COVID-19 pandemic has caused the society lockdowns and a large number of deaths in many countries. Potential transmission cluster discovery is to find all suspected users with infections, which is greatly needed to fast discover virus transmission chains so as to prevent an outbreak of COVID-19 as early as possible. In this article, we study the problem of potential transmission cluster discovery based on the spatio-temporal logs. Given a query of patient user q and a timestamp of confirmed infection tq, the problem is to find all potential infected users who have close social contacts to user q before time tq. We motivate and formulate the potential transmission cluster model, equipped with a detailed analysis of transmission cluster property and particular model usability. To identify potential clusters, one straightforward method is to compute all close contacts on-the-fly, which is simple but inefficient caused by scanning spatio-temporal logs many times. To accelerate the efficiency, we propose two indexing algorithms by constructing a multigraph index and an advanced BCG-index. Leveraging two well-designed techniques of spatio-temporal compression and graph partition on bipartite contact graphs, our BCG-index approach achieves a good balance of index construction and online query processing to fast discover potential transmission cluster. We theoretically analyze and compare the algorithm complexity of three proposed approaches. Extensive experiments on real-world check-in datasets and COVID-19 confirmed cases in the United States validate the effectiveness and efficiency of our potential transmission cluster model and algorithms. © 2023 Association for Computing Machinery.

A Practical and Empirical Comparison of Three Topic Modeling Methods using a COVID-19 Corpus: LSA, LDA, and Top2Vec

This study was prepared as a practical guide for researchers interested in using topic modeling methodologies. This study is specially designed for those with difficulty determining which methodology to use. Many topic modeling methods have been developed since the 1980s namely, latent semantic indexing or analysis (LSI/LSA), probabilistic LSI/LSA (pLSI/pLSA), naïve Bayes, the Author-Recipient-Topic (ART), Latent Dirichlet Allocation (LDA), Topic Over Time (TOT), Dynamic Topic Models (DTM), Word2Vec, Top2Vec and \variation and combination of these techniques. For researchers from disciplines other than computer science may find it challenging to select a topic modeling methodology. We compared a recently developed topic modeling algorithm-Top2Vec- with two of the most conventional and frequently-used methodologies-LSA and LDA. As a study sample, we used a corpus of 65,292 COVID-19-focused s. Among the 11 topics we identified in each methodology, we found high levels of correlation between LDA and Top2Vec results, followed by LSA and LDA and Top2Vec and LSA. We also provided information on computational resources we used to perform the analyses and provided practical guidelines and recommendations for researchers. © 2023 IEEE Computer Society. All rights reserved.

The COVID-19 pandemic and global food security: a bibliometric analysis and future research direction

PurposeThis study aims to analyse the nature and trends in the knowledge discovery process on COVID-19 and food insecurity using a comprehensive bibliometric analysis based on the indexing literature in the Scopus database.Design/methodology/approachData were extracted from Scopus using the keywords COVID-19 and food security to ensure extensive coverage. A total of 840 research papers on COVID-19 and food security were analysed using VOSviewer and RStudio software.FindingsThe findings of the bibliometric analysis in terms of mapping of scientific research across countries and co-occurrence of research keywords provide the trends in research focus and future directions for food insecurity research during times of uncertainty. Based on this analysis, the focus of scientific research has been categorised as COVID-19 and food supply resilience, COVID-19 and food security, COVID-19 and public health, COVID-19 and nutrition, COVID-19 and mental health and depression, COVID-19 and migration and COVID-19 and social distancing. A thematic map was created to identify future research on COVID-19 and food security.Practical implicationsThis analysis identifies potential research areas such as food supply and production, nutrition and health that may help set future research agendas and devise policy supports for better managing food insecurity during uncertainty.Originality/valueThis analysis provides epistemological underpinnings for knowledge generation and acquisition on COVID-19 and food insecurity.

Drowsiness Detection System for Masked Face Based on Deep Neural Network and Haar Cascade

Nowadays, technology is growing rapidly followed by modernization. Face detection technology is one technology that has been developed and applied in various sectors such as biometrics recognition systems, retrieval systems, database indexing in digital video, security systems with restricted area access control, video conferencing, and human interaction systems. Eye detection is a further development of face detection in which the image of a human face was detected to be processed by detecting the location of both eyes on the face. Nowadays, the eye detection system can be used as a means of developing more complex applications and can be applied directly in the aspect of technology that uses eye detection like, eye state detection system, drowsiness and fatigue detection system, safety driving support systems or driver assistance system. In this study we propose drowsiness detection system utilizing current novel classification model such as Deep Neural Network (DNN), combined with Haar Cascade. The DNN is utilized to detect face, while Haar Cascade is utilized for detecting the eyes and its state on the detected face. In this study, due to Covid19 pandemic, we focused on developing the classifiers for detecting the face with mask. Apart from that, our proposed classifiers are also capable of identifying non-masked faces. The experimental result showed that by utilizing DNN and Haar Cascade, our proposed system could reach accuracy, precision, recall, and f1 measure as much as 81%, 88%, 80%, and 84%, respectively. © 2022 IEEE.

An Ensemble Learning Approach to Predict Employees' Preference for E-Working in the Post-Pandemic World

The Covid-19 pandemic has forced a large segment of the global workforce to shift to e-working. The pandemic has convinced many organizations that e-working has benefits for a successful business. As a result, it is critical to identify employees' suggestions and evaluate their motivation to continue the e-working concept in the post-pandemic world. The study was conducted by randomly surveying employees using various Machine Learning algorithms, including Naive Bayes, Decision Tree, Random Forest, Multilayer Perceptron (MLP), Support Vector Machine (SVM) and logistic regression. The ensembling algorithm uses 66% of the percentage split method in the Waikato Environment for Knowledge Analysis (WEKA) tool. Accuracy, precision, recall, /-measure values and error rates were used to compare the results. The ensemble learning algorithm shows the best results with 90% accuracy, making it easier to predict employees' preference for e-working and accordingly take decisions.

Non-Sequential Indexing of Videos using Linguistic Computation

Since the outbreak of Corona Virus Disease(COVID-19), the education sector has seen a drift from traditional in-person teaching methods to virtually-assisted learning. This new trend has paved its path for students to easily gain access to a variety of educational instructors across the globe. But online education comes with its own potential and challenges. Factors like high availability, flexibility, and affordability of the online learning platforms add to the effective deliverance of the content in this progressive present-day online learning. Some key disadvantages are lack of powerful conveyance of content to listeners and sequential navigation of videos. Linearly searching for specific topics through long videos is a common problem that students face, while learning from the internet. This research study proposes a novel approach to promote the application of non-sequential navigation of videos by identifying key-topics and automatically generating timestamps. The model utilizes Natural Language Processing (NLP) and Optical Character Recognition (OCR) techniques for determining the key topics from the video. Timestamps are identified for the keywords before they are uttered, using a novel algorithm for audio indexing. Finally, timestamps are successfully generated for every keyword. Through this study, the objective of non-sequential navigation of videos using a new audio-indexing algorithm is achieved. © 2022 IEEE

The Impact of Leadership on Digital Learning in Health

Leaders have a key responsibility in providing employees with the necessary knowledge and training so that they can carry out their work. At the same time, advances in technology, changes in demographics, increased demands on health services and rapid developments of health professions are all elements that contribute to a particularly enhanced demand for upgrade of employees' competences. The role of leadership in meeting these demands and developing competence in health is limited. This is particularly true for competence development through the use of digital tools. Through a structured search in the Social Citation Index and the Science Citation index (in Web of Science) we review past research and develop key insights that address how leadership can be linked to digital learning within health. In particular, we use a relevant and extensive set of search terms in the areas of nursing and health, leadership, knowledge development and digital learning. A key finding from this search was the lack of existing research, which suggests that more research and broader structured searches are needed. This is particularly imminent following the covid-19 pandemic, which has demanded the digitalisation of many fields where education and health have had to undertake considerable changes. We identified three main core stakeholders relative to whom leadership is essential in understanding its impact on digital learning: patients, students and health professionals. Further, results pointed to learning effects as well as barriers and enablers of effect as key dimensions that leaders need to understand and consider. Underlying any effect of leadership and digital learning initiatives are modern tools of technology, including the right information, system and support that enhances resource efficiency. Finally, the leadership effect on learning is context dependent and related to culture, motivation, reflection, behaviour and digital competence. © 2022, Academic Conferences and Publishing International Limited. All rights reserved.

Opinion classification at subtopic level from COVID vaccination-related tweets.

Coronavirus disease 2019 (Covid-19) is a contiguous disease which affected a large volume of population with a high mortality rate across the globe. For dealing with the recent spread of COVID-19, one of the prime measures was to vaccinate people in full extent. People across the globe have diverse opinion regarding the vaccination process, its side effect and effectiveness. Such opinions get located into different micro-blogging sites including twitter. Opinion mining through analyzing public sentiments of such micro-blogs is a common method for detection of public responses. This paper focuses on classifying the public opinions expressed related to COVID-19 vaccination at sub topic level. The procedure tries to find out different keywords regarding positive, negative and neutral sentences. From those keywords, different related query set was constructed using Rocchio query expansion algorithm for positive, negative and neutral sentiments. Later Extended query set is used to form subtopic using LDA algorithm to identify the nature of the tweets. The proposed LDA model came across with 0.56 coherence score with twenty subtopics, which is fair enough to classify the tweets in different classes. This trained model is finally used to classify the tweets in real time with Apache Kafka framework regarding different subtopic based on positive, negative or neutral sentiment.

Researching vulnerable multilinguals: Developing an inclusive research practice

Focus of the seminar While the COVID-19 pandemic has resulted in many international and virtual opportunities that may have not been possible previously, we were convinced that a face-to-face event could have been more beneficial and impactful (i.e., networking and exchange of best practice). Dr. Ateek's presentation ‘Language analysis for determination of origin (LADO) and whether it works' explored how LADO is used as a gatekeeper by the Home Office with a focus on the perspectives of asylum seeker-participants who went through the process. In the presentation ‘Striving for inclusivity in an exclusionary environment – conducting research with refugees and asylum seekers in the UK', Dr. Reynolds reported and reflected on her own efforts to work ethically, responsibly, and reflexively with asylum seekers and refugees during her linguistic ethnographic doctoral study of communication in refugee and asylum legal advice meetings in the UK context.

Navigating the cellular landscape in tissue: Recent advances in defining the pathogenesis of human disease.

Over the past decade, our understanding of human diseases has rapidly grown from the rise of single-cell spatial biology. While conventional tissue imaging has focused on visualizing morphological features, the development of multiplex tissue imaging from fluorescence-based methods to DNA- and mass cytometry-based methods has allowed visualization of over 60 markers on a single tissue section. The advancement of spatial biology with a single-cell resolution has enabled the visualization of cell-cell interactions and the tissue microenvironment, a crucial part to understanding the mechanisms underlying pathogenesis. Alongside the development of extensive marker panels which can distinguish distinct cell phenotypes, multiplex tissue imaging has facilitated the analysis of high dimensional data to identify novel biomarkers and therapeutic targets, while considering the spatial context of the cellular environment. This mini-review provides an overview of the recent advancements in multiplex imaging technologies and examines how these methods have been used in exploring pathogenesis and biomarker discovery in cancer, autoimmune and infectious diseases.

Fuzzy Inference Speed Enhancement for Low Budget Computers using Hash Indices

Fuzzy inference is a powerful tool used in many fields of science nowadays, including medical science. However, for applications where the number of fuzzy rules is very large, the increased computational complexity for systems with limited resources (such as low budget computers and embedded systems) can result in a very slow operation. In this paper, a new method is proposed to accelerate the operation of Fuzzy Inference Systems that is faster than the conventional sequential procedure, primarily for such computer systems. © 2022 IEEE.

Scientific information in times of pandemic analysis of indexing time of journal articles with controlled vocabulary in PubMed

Introduction/Objective: It presents a theme linked to the analysis and verification of delays in indexing scientific journal articles with the terms of the controlled vocabulary Medical Subject Headings (MeSH) in the PubMed search engine. Methodology: Includes bibliographic research for thematic contextualization, and is exploratory and descriptive with practical applicability on subjects related to the COVID-19 pandemic. A search strategy was built on PubMed with 16 terms related to the theme in the MeSH Terms and Text Word fields. The study was published in 2021 in the Portuguese language. The metadata of the 85 publications were exported for analysis in a spreadsheet under the aspects of entry of the publication in PubMed, time until indexing with the MeSH terms and category of the publication. Results: About 89% of the publications, considering the sample of 62 items with MeSH Terms, had a delay in indexing of at least 15 days;and about 11% collapsed in 15 to 135 days. Conclusion: In order for researchers to be able to retrieve the most scientific content on COVID-19, it is essential that they are constructed of searches that contemplate the use of the MeSH controlled vocabulary term combined with the use of the nomenclature variations of the theme in other fields, such as Text Word, since publications may appear in the search engine, but have not yet been indexed, in order to retrieve a greater number of scientific literature already published.

Multi-probe attention neural network for COVID-19 semantic indexing.

BACKGROUND: The COVID-19 pandemic has increasingly accelerated the publication pace of scientific literature. How to efficiently curate and index this large amount of biomedical literature under the current crisis is of great importance. Previous literature indexing is mainly performed by human experts using Medical Subject Headings (MeSH), which is labor-intensive and time-consuming. Therefore, to alleviate the expensive time consumption and monetary cost, there is an urgent need for automatic semantic indexing technologies for the emerging COVID-19 domain. RESULTS: In this research, to investigate the semantic indexing problem for COVID-19, we first construct the new COVID-19 Semantic Indexing dataset, which consists of more than 80 thousand biomedical articles. We then propose a novel semantic indexing framework based on the multi-probe attention neural network (MPANN) to address the COVID-19 semantic indexing problem. Specifically, we employ a k-nearest neighbour based MeSH masking approach to generate candidate topic terms for each input article. We encode and feed the selected candidate terms as well as other contextual information as probes into the downstream attention-based neural network. Each semantic probe carries specific aspects of biomedical knowledge and provides informatively discriminative features for the input article. After extracting the semantic features at both term-level and document-level through the attention-based neural network, MPANN adopts a linear multi-view classifier to conduct the final topic prediction for COVID-19 semantic indexing. CONCLUSION: The experimental results suggest that MPANN promises to represent the semantic features of biomedical texts and is effective in predicting semantic topics for COVID-19 related biomedical articles.

The third wind of language learning strategies research

The researchers noted that whereas most language learning strategies research has accounted for strategies used separately or as part of latent factors, their study participants were found to almost always use strategies in sequences, pairs, and clusters. By analyzing these strategy combinations and their moment-to-moment functions, Cohen and Wang identified six patterns: (1) linear progression in functions (e.g., from metacognitive to cognitive);(2) simultaneous occurrence of functions (e.g., social and affective);(3) linear progression in functions + simultaneous occurrence of functions;(4) bi-directional fluctuation of functions (e.g., back and forth between metacognitive and cognitive);(5) bi-directional fluctuation + simultaneous occurrence of functions;and (6) simultaneous occurrence of functions + micro-fluctuations of functions (see Cohen & Wang, 2018). [...]Gao and Hu drew on an adapted activity system model based on earlier work by Engeström, who expanded on Vygotsky's idea of mediation (see Gao & Hu, 2020). The system has three mediating resources: community (e.g., a social group performing similar actions with similar goals), rules (e.g., time and academic requirements), and division of labor (e.g., roles and power relationships with the community).

Constructing Multiple-Objective Portfolio Selection for Green Innovation and Dominating Green Innovation Indexes

Green innovation investments have rapidly grown since 2000. Green innovation indexes play important roles and are typically constructed by screening and indexing. However, Nobel Laureate Markowitz emphasizes portfolio selection instead of security selection and accentuates that “A good portfolio is more than a long list of good stocks.” Moreover, the screening-indexing strategies ignore that investors can take green innovation as an additional objective and thus gain additional utility. We consequently construct 3-objective portfolio selection for green innovation in addition to variance and expected return. An efficient frontier of portfolio selection then extends to an efficient surface which is a panorama of the optimal variance, expected return, and expected green innovation. Investors thus fully envisage the trade-offs and enjoy the freedom of choosing preferred portfolios on the surface. In contrast, the screening-indexing strategies inflexibly leave investors with only one point (i.e., the green innovation index). As the originality, we prove in a theorem that there typically exists a curve on the efficient surface so all portfolios on the curve dominate the green innovation index. We test the dominance by component stocks of China Securities Index 300 and obtain affirmative results out of sample. The results still hold in robustness tests. At last, we classify green innovation into categories, further model the categories by general k-objective portfolio selection, and still illustrate the dominance. Consequently, investors can consider and control each category.