Large-Scale Analysis of New Employee Network Dynamics

The COVID-19 pandemic has accelerated digital transformations across industries, but also introduced new challenges into workplaces, including the difficulties of effectively socializing with colleagues when working remotely. This challenge is exacerbated for new employees who need to develop workplace networks from the outset. In this paper, by analyzing a large-scale telemetry dataset of more than 10,000 Microsoft employees who joined the company in the first three months of 2022, we describe how new employees interact and telecommute with their colleagues during their "onboarding"period. Our results reveal that although new hires are gradually expanding networks over time, there still exists significant gaps between their network statistics and those of tenured employees even after the six-month onboarding phase. We also observe that heterogeneity exists among new employees in how their networks change over time, where employees whose job tasks do not necessarily require extensive and diverse connections could be at a disadvantaged position in this onboarding process. By investigating how web-based people recommendations in organizational knowledge base facilitate new employees naturally expand their networks, we also demonstrate the potential of web-based applications for addressing the aforementioned socialization challenges. Altogether, our findings provide insights on new employee network dynamics in remote and hybrid work environments, which may help guide organizational leaders and web application developers on quantifying and improving the socialization experiences of new employees in digital workplaces. © 2023 ACM.

Semi-Supervised Modified-UNet for Lung Infection Image Segmentation

Automatic lung infection segmentation in computed tomography (CT) scans can offer great assistance in radiological diagnosis by improving accuracy and reducing time required for diagnosis. The biggest challenges for deep learning (DL) models in segmenting infection region are the high variances in infection characteristics, fuzzy boundaries between infected and normal tissues, and the troubles in getting large number of annotated data for training. To resolve such issues, we propose a Modified U-Net (Mod-UNet) model with minor architectural changes and significant modifications in the training process of vanilla 2D UNet. As part of these modifications, we updated the loss function, optimization function, and regularization methods, added a learning rate scheduler and applied advanced data augmentation techniques. Segmentation results on two Covid-19 Lung CT segmentation datasets show that the performance of Mod-UNet is considerably better than the baseline U-Net. Furthermore, to mitigate the issue of lack of annotated data, the Mod-UNet is used in a semi-supervised framework (Semi-Mod-UNet) which works on a random sampling approach to progressively enlarge the training dataset from a large pool of unannotated CT slices. Exhaustive experiments on the two Covid-19 CT segmentation datasets and on a real lung CT volume show that the Mod-UNet and Semi-Mod-UNet significantly outperform other state-of-theart approaches in automated lung infection segmentation. IEEE

Multi2Claim: Generating Scientific Claims from Multi-Choice Questions for Scientific Fact-Checking

In automated scientific fact-checking, machine learning models are trained to verify scientific claims given evidence. A major bottleneck of this task is the availability of large-scale training datasets on different domains, due to the required domain expertise for data annotation. However, multiple-choice question-answering datasets are readily available across many different domains, thanks to the modern online education and assessment systems. As one of the first steps towards addressing the fact-checking dataset scarcity problem in scientific domains, we propose a pipeline for automatically converting multiple-choice questions into fact-checking data, which we call Multi2Claim. By applying the proposed pipeline, we generated two large-scale datasets for scientific-fact-checking: Med-Fact and Gsci-Fact for the medical and general science domains, respectively. These two datasets are among the first examples of large-scale scientific-fact-checking datasets. We developed baseline models for the verdict prediction task using each dataset. Additionally, we demonstrated that the datasets could be used to improve performance measured by weighted F1 on existing fact-checking datasets such as SciFact, HEALTHVER, COVID-Fact, and CLIMATE-FEVER. In some cases, the improvement in performance was up to a 26% increase. The generated datasets are publicly available. © 2023 Association for Computational Linguistics.

Optimizing Medical Healthcare Services Using Big Data Analytics

Big data is a very large size of datasets which come from many different sources and are in a wide variety of forms. Due to its enormous potential, big data has gained popularity in recent years. Big data enables us to investigate and reinvent numerous fields, including the healthcare industry, education, and others. Big data specifically in the healthcare sector comes from a variety of sources, including patient medical information, hospital records, findings from physical exams, and the outcomes of medical devices. Covid19 recently, one of the most neglected areas to concentrate on has come under scrutiny due to the pandemic: healthcare management. Patient duration of stay in a hospital is one crucial statistic to monitor and forecast if one wishes to increase the effectiveness of healthcare management in a hospital, even if there are many use cases for data science in healthcare management. At the time of admission, this metric aids hospitals in identifying patients who are at high Length of Stay namely LS risk (patients who will stay longer). Once identified, patients at high risk for LS can have their treatment plans improved to reduce LS and reduce the risk of infection in staff or visitors. Additionally, prior awareness of LS might help with planning logistics like room and bed allotment. The aim of the suggested system is to precisely anticipate the length of stay for each patient on an individual basis so that hospitals can use this knowledge for better functioning and resource allocation using data analytics. This would contribute to improving treatments and services. © 2023 IEEE.

Computerized Face Mask Detection System Using Deep CNN and Transfer Learning

Since the COVID-19 pandemic is on the rise again with hazardous effects in China, it has become very crucial for global individuals and the authorities to avoid spreading of the virus. This research aims to identify algorithms with high accuracy and moderate computing complexity at the same time (although conventional machine learning works on low computation power, we have rather used CNN for our research work as the accuracy of CNN is drastically greater than the former), to identify the proper enforcement of face masks. In order to find the best Neural Network architecture we used many deep CNN Methodologies to solve classification problem in regards of masked and non masked image dataset. In this approach we applied different model architectures, like VGG16, Resnet50, Resnet101 and VGG19, on a large dataset to train on and compared the model on the basis of accuracy in which VGG16 came out to be the best. VGG16 was further tuned with different optimizers to determine the one best fit of the model. VGG16 gave an ideal accuracy of 99.37% with the best fit optimizer over a real life data set. © 2023 IEEE.

Explainable AI for COVID-19 prognosis from early Chest X-ray and clinical data in the context of the COVID-CXR international hackathon

Although many AI-based scientific works regarding chest X-ray (CXR) interpretation focused on COVID-19 diagnosis, fewer papers focused on other relevant tasks, like severity estimation, deterioration, and prognosis. The same holds for explainable decisions to estimate COVID-19 prognosis as well. The international hackathon launched during Dubai Expo 2020, aimed at designing machine learning solutions to help physicians formulate COVID-19 patients' prognosis, was the occasion to develop a machine learning model capable of predicting such prognoses and justifying them through interpretable explanations. The large hackathon dataset comprised subjects characterized by their CXR and numerous clinical features collected during triage. To calculate the prognostic value, our model considered both patients' CXRs and clinical features. After automatic pre-processing to improve their quality, CXRs were processed by a Deep Learning model to estimate the lung compromise degree, which has been considered as an additional clinical feature. Original clinical parameters suffered from missing values that were adequately handled. We trained and evaluated multiple models to find the best one and fine-tune it before the inference process. Finally, we produced novel explanations, both visual and numerical, to justify the model predictions. Ultimately, our model processes a CXR and several clinical data to estimate a patient's prognosis related to the COVID-19 disease. It proved to be accurate and was ranked second in the final rankings with 75%, 73.9%, and 74.4% in sensitivity, specificity, and balanced accuracy, respectively. In terms of model explainability, it was ranked first since it was agreed to be the most interpretable by health professionals. © 2023 SPIE.

Learning Topical Structured Interfaces from Medical Research Literature

Accessing large-scale structured datasets such as WDC or CORD-191 is very challenging. Even if one topic (e.g. COVID-19 vaccine efficacy) is of interest, all topical tables in different sources/papers have hundreds of different schemas, depending on the authors, which significantly complicates both finding and querying them. Here we demonstrate a scalable Meta-profiler system, capable of constructing a structured standardized interface to a topic of interest in large-scale (semi-)structured datasets. This interface, that we call Meta-profile represents a multi-dimensional meta-data summary for a selected topic of interest, accumulating all differently structured representations of the topical tables in the dataset. Such Meta-profiles can be used as a rich visualization as well as a robust structural query interface simplifying access to large-scale (semi-)structured data for different user segments, such as data scientists and end users. © 2023 Owner/Author.

Decision region analysis to deconstruct the subgroup influence on AI/ML predictions

Assessing the generalizability of deep learning algorithms based on the size and diversity of the training data is not trivial. This study uses the mapping of samples in the image data space to the decision regions in the prediction space to understand how different subgroups in the data impact the neural network learning process and affect model generalizability. Using vicinal distribution-based linear interpolation, a plane of the decision region space spanned by the random 'triplet' of three images can be constructed. Analyzing these decision regions for many random triplets can provide insight into the relationships between distinct subgroups. In this study, a contrastive self-supervised approach is used to develop a 'base' classification model trained on a large chest x-ray (CXR) dataset. The base model is fine-tuned on COVID-19 CXR data to predict image acquisition technology (computed radiography (CR) or digital radiography (DX) and patient sex (male (M) or female (F)). Decision region analysis shows that the model's image acquisition technology decision space is dominated by CR, regardless of the acquisition technology for the base images. Similarly, the Female class dominates the decision space. This study shows that decision region analysis has the potential to provide insights into subgroup diversity, sources of imbalances in the data, and model generalizability. © 2023 SPIE.

Deep Convolutional Neural Network Based Automatic COVID-19 Detection from Chest X-ray Images

To minimize the rate of death from COVID-19 and stop the disease from spreading early detection is vital. The normal RT-PCR tests for COVID-19 detection take a long time to complete. In contrast to this test, Covid-19 can be quickly detected using various machine-learning technologies. Previous studies only had access to smaller datasets, as COVID-19 data was not readily available back then. Since COVID-19 is a dangerous virus, the model needs to be robust and trustworthy, and the model must be trained on a large and diverse dataset. To overcome that problem, this study combines six publicly available Chest X-ray datasets to produce a larger and more diverse balanced dataset with a total of 68,424 images. In this study, we develop a CNN model that primarily entails two steps: (a) feature extraction and (b) classification, which are used to identify COVID-19 positive cases from X-ray images. The accuracy of this proposed model is 97.58%, which is higher than most state-of-the-art models. © 2022 IEEE.

Addressing Issues of Cross-Linguality in Open-Retrieval Question Answering Systems For Emergent Domains

Open-retrieval question answering systems are generally trained and tested on large datasets in well-established domains. However, low-resource settings such as new and emerging domains would especially benefit from reliable question answering systems. Furthermore, multilingual and cross-lingual resources in emergent domains are scarce, leading to few or no such systems. In this paper, we demonstrate a cross-lingual open-retrieval question answering system for the emergent domain of COVID-19. Our system adopts a corpus of scientific articles to ensure that retrieved documents are reliable. To address the scarcity of cross-lingual training data in emergent domains, we present a method utilizing automatic translation, alignment, and filtering to produce English-to-all datasets. We show that a deep semantic retriever greatly benefits from training on our English-to-all data and significantly outperforms a BM25 baseline in the cross-lingual setting. We illustrate the capabilities of our system with examples and release all code necessary to train and deploy such a system1 © 2023 Association for Computational Linguistics.

Towards improved thermal comfort predictions and higher energy savings: building energy model of an open-plan office based on indoor CO2 and temperature controls

In recent work, a Hierarchical Bayesian model was developed to predict occupants' thermal comfort as a function of thermal indoor environmental conditions and indoor CO2 concentrations. The model was trained on two large IEQ field datasets consisting of physical and subjective measurements of IEQ collected from over 900 workstations in 14 buildings across Canada and the US. Posterior results revealed that including measurements of CO2 in thermal comfort modelling credibly increases the prediction accuracy of thermal comfort and in a manner that can support future thermal comfort prediction. In this paper, the predictive model of thermal comfort is integrated into a building energy model (BEM) that simulates an open-concept mechanically-ventilated office space located in Vancouver. The model predicts occupants' thermal satisfaction and heating energy consumption as a function of setpoint thermal conditions and indoor CO2 concentrations such that, for the same thermal comfort level, higher air changes per hour can be achieved by pumping a higher amount of less-conditioned fresh air. The results show that it is possible to reduce the energy demand of increasing fresh air ventilation rates in winter by decreasing indoor air temperature setpoints in a way that does not affect perceived thermal satisfaction. This paper presents a solution for building managers that have been under pressure to increase current ventilation rates during the COVID-19 pandemic. © 2022 17th International Conference on Indoor Air Quality and Climate, INDOOR AIR 2022. All rights reserved.

Context Aware Automatic Subjective and Objective Question Generation using Fast Text to Text Transfer Learning

Online learning has gained a tremendous popularity in the last decade due to the facility to learn anytime, anything, anywhere from the ocean of web resources available. Especially the lockdown all over the world due to the Covid-19 pandemic has brought an enormous attention towards the online learning for value addition and skills development not only for the school/college students, but also to the working professionals. This massive growth in online learning has made the task of assessment very tedious and demands training, experience and resources. Automatic Question generation (AQG) techniques have been introduced to resolve this problem by deriving a question bank from the text documents. However, the performance of conventional AQG techniques is subject to the availability of large labelled training dataset. The requirement of deep linguistic knowledge for the generation of heuristic and hand-crafted rules to transform declarative sentence into interrogative sentence makes the problem further complicated. This paper presents a transfer learning-based text to text transformation model to generate the subjective and objective questions automatically from the text document. The proposed AQG model utilizes the Text-to-Text-Transfer-Transformer (T5) which reframes natural language processing tasks into a unified text-to-text-format and augments it with word sense disambiguation (WSD), ConceptNet and domain adaptation framework to improve the meaningfulness of the questions. Fast T5 library with beam-search decoding algorithm has been used here to reduce the model size and increase the speed of the model through quantization of the whole model by Open Neural Network Exchange (ONNX) framework. The keywords extraction in the proposed framework is performed using the Multipartite graphs to enhance the context awareness. The qualitative and quantitative performance of the proposed AQG model is evaluated through a comprehensive experimental analysis over the publicly available Squad dataset. © 2023, International Journal of Advanced Computer Science and Applications. All Rights Reserved.

Artificial Intelligence based prediction system for General Medicine

COVID-19 is highly infectious and has been extensively spread worldwide, with approximately 651 million definite cases crosswise the globe including Pakistan. At that era of pandemic where patients are not able to approach a doctor for even the routine checkups, in such curial situation even normal disease checkups are ignored by many families due to pandemic situations, those diseases may lead to be a perilous disease are results of it. Human disorders portray scenarios that even disturb or permanently cutoff the essential functions of a body parts. Consequently, the aim is to transform raw health data potential into actionable insights to applying the promising outcomes of Body Sensor Network (BSN) and State-of-Art Artificial Intelligence (AI) techniques to get proper medicine allocation to the particular health state of patient. In this paper the different techniques of Deep Learning and Machine Learning introduced to predict the actual medicine for the specific health state of patient according to data from the BSN. Experiments have been conducted on large dataset which shepherd it into 16 states of patient's health which will allotted to AI model to predict the medicine accordingly to the health state of patient. Experimental results show the 87.46% by Random Forest, 92.74% by K-Nearest Neighbors, 74.57% by Naive Bayes, 94.41% by Extreme Gradient Boost, 84.88% by Multi-Layer Perceptron in terms of precision of model training in event of classification. © 2023 IEEE.

Analyzing the Evolution of Disinformation Content on Facebook - a Pilot Study

Disinformation spread on social media generates a truly massive amount of content on a daily basis, much of it not quite duplicated but repetitive and related. In this paper, we present an approach for clustering social media posts based on topic modeling in order to identify and formalize an underlying structure in all the noise. This would be of great benefit for tracking evolving trends, analyzing large-scale campaigns, and focusing efforts on debunking or community outreach. The steps we took in particular include harvesting through CrowdTangle huge collection of Facebook posts explicitly identified as containing disinformation by debunking experts, following those links back to the people, pages and groups where they were shared then collecting all posts shared on those channels over an extended period of time. This generated a very large textual dataset which was used in the topic modeling experiments attempting to identify the larger trends in the available data. Finally, the results were transformed and collected in a Knowledge Graph for further study and analysis. Our main goal is to investigate different trends and common patterns in disinformation campaigns, and whether there exist some correlations between some of them. For instance, for some of the most recent social media posts related to COVID-19 and political situation in Ukraine. © 2022 Copyright for this paper by its authors.

COVID-19 prediction with Chest X-Ray images using CNN

Coronavirus has outbreak as an epidemic disease, created a pandemic situation for the public health across the Globe. Screening for the large masses is extremely crucial to control disease for the people in a neighborhood. Real-time-PCR[18] is the general diagnostic approach for pathological examination. However, the increasing figure of false results from the test has created a way in choosing alternative procedures. COVID-19 patient's X-rays images of chest has emerged as a significant approach for screening the COVID-19 disease. However, accuracy depends on the knowledge of a radiologist. X-Ray images of lungs may be proper assistive tool for diagnosis in reducing the burden of the doctor. Deep Learning techniques, especially Convolutional Neural Networks (CNN), have been shown to be effective for classification of images in the medical field. Diagnosing the COVID-19 using the four types of Deep-CNN models because they have pre-trained weights. Model needs to pre-trained on the ImageNet database in simplifying the large datasets. CNN-based architectures were found to be ideal in diagnosing the COVID-19 disease. The model having an efficiency of 0.9835 in accuracy, precision of 0.915, sensitivity of 0.963, specificity with 0.972, 0.987 F1 Score and 0.925 ROC AUC. © 2023 IEEE.

Detection of Sars-Cov-2 from human chest CT images in Multi-Convolutional Neural Network's environment

This article focuses on the detection of the Sars-Cov2 virus from a large-scale public human chest Computed Tomography (CT) scan image dataset using a customized convolutional neural network model and other convolutional neural network models such as VGG-16, VGG-19, ResNet 50, Inception v3, DenseNet, XceptionNet, and MobileNet v2. The proposed customized convolutional neural network architecture contains two convolutional layers, one max pooling layer, two convolutional layers, one max pooling layer, one flatten layer, two dense layers, and an activation layer. All the models are applied on a large-scale public human chest Computed Tomography (CT) scan image dataset. To measure the performance of the various convolutional neural network models, different parameters are used such as Accuracy, Error Rate, Precision, Recall, and F1 score. The proposed customized convolutional neural network architecture's Accuracy, Error Rate, Precision Rate, Recall, and F1 Score are 0.924, 0.076, 0.937, 0.921, and 0.926 respectively. In comparison with other existing convolutional neural network strategies, the performance of the proposed model is superior as far as comparative tables and graphs are concerned. The proposed customized convolutional neural network model may help researchers and medical professionals to create a full-fledged computer-based Sars-Cov-2 virus detection system in the near future. © 2023 IEEE.

Ontology-based semantic data interestingness using BERT models

The COVID-19 pandemic has generated massive data in the healthcare sector in recent years, encouraging researchers and scientists to uncover the underlying facts. Mining interesting patterns in the large COVID-19 corpora is very important and useful for the decision makers. This paper presents a novel approach for uncovering interesting insights in large datasets using ontologies and BERT models. The research proposes a framework for extracting semantically rich facts from data by incorporating domain knowledge into the data mining process through the use of ontologies. An improved Apriori algorithm is employed for mining semantic association rules, while the interestingness of the rules is evaluated using BERT models for semantic richness. The results of the proposed framework are compared with state-of-the-art methods and evaluated using a combination of domain expert evaluation and statistical significance testing. The study offers a promising solution for finding meaningful relationships and facts in large datasets, particularly in the healthcare sector. © 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

COVID-19 Identification From Lung CT Scans in a Low-Resource Setting Using a Regularized 3D Convolutional Neural Network

Right at the end of 2019, the world saw an outbreak of a new type of SARS (severe acute respiratory syndrome) disease, SARS-Cov-2, or COVID-19. Even in 2022, around 1 million people worldwide are getting infected with the virus every day. To date, more than 6 million people have died as a result of the virus. To tackle the pandemic, the first step is to successfully detect the virus among the mass population. The most popular method is the RT-PCR test, which, unfortunately, is not always conclusive. The physicians thus suggest lung CT tests for the patients for clinical relevance. But the problem with lung CT scans for the detection of coronavirus is that the COVID-19 infected scan is very similar to community-affected pneumonia (CAP) infected scan, and the results in many cases get wrongly interpreted. In addition, the virus is always mutating into different strains, and the severity and infection pattern slightly change with each mutation. Because of this rapid mutation, a large and balanced dataset of lung CT scans is not always available. In this work, we systematically evaluate the accuracy of a deep 3D convolutional neural network (CNN) on a small-scale and highly imbalanced dataset of lung CT scans (the SPGC COVID 2021 dataset). Our experiments show that it can outperform previous state-of-the-art 3D CNN models with proper regularization, an appropriate number of dense layers, and a weighted loss function. Our research, therefore, suggests an effective solution for identifying COVID-19 in lung CT scans using deep learning for small and highly imbalanced datasets. © 2022 IEEE.

Twitter User Sentiments Analysis: Health System Cyberattacks Case Study

Social media, such as Twitter, allow people to interact with ongoing events and share their sentiments. Therefore, people use social media to report and express their emotions about events they are experiencing. Furthermore, some officials take advantage of the popularity of social media to keep the public informed, especially during emergent events. Researchers have covered sentiment analysis on Twitter in many fields, such as movie reviews, stocks, politics, health, and sports. However, there is a research gap in studying the public's concerns on social media when a cybersecurity breach occurs and how people's sentiment changes over time. To fill the gap, The researchers selected the cyberattacks against Universal Health Services (UHS) during the late days of September 2020 and collected a large dataset of related tweets over five weeks. Live-streaming tweets and historical ones both were compiled. The focus while gathering tweets was in the context of cyberattacks on UHS using keywords and hashtags such as Universal Health System, UHS cyberattack, UHS Ransome, UHS security breach, and UHS locked. Then, the researchers determined tweets' sentiment classification on this developing event using deep learning of Long Short-Term Memory (LSTM) and Artificial Neural Networks (ANN) and their accuracies. Furthermore, the researchers performed exploratory data analysis for the dataset supplying information about how sentiment has changed over time to compare the sentiment per week since the start of these cyberattacks on UHS. This study is the first to provide an analysis of the public's sentiment toward a significant cybersecurity breach on a healthcare provider dealing with COVID-19 based on a large-scale dataset extracted from social media feeds. © 2023 IEEE.

SCN-SAM: A Modified Self-Cure Network for Facial Expression Recognition under Face Masks

Due to COVID-19, wearing masks has become more common. However, it is challenging to recognize expressions in the images of people wearing masks. In general facial recognition problems, blurred images and incorrect annotations of images in large-scale image datasets can make the model's training difficult, which can lead to degraded recognition performance. To address this problem, the Self-Cure Network (SCN) effectively suppresses the over-fitting of the network to images with uncertain labeling in large-scale facial expression datasets. However, it is not clear how well the SCN suppresses the uncertainty of facial expression images with masks. This paper verifies the recognition ability of SCN on images of people wearing masks and proposes a self-adjustment module to further improve SCN (called SCN-SAM). First, we experimentally demonstrate the effectiveness of SCN on the masked facial expression dataset. We then add a self-adjustment module without extensive modifications to SCN and demonstrate that SCN-SAM outperforms state-of-the-art methods in synthetic noise-added FER datasets. © 2023 Global IT Research Institute (GiRI).