Determinants of students' satisfaction with digital classroom services: moderating effect of students' level of study

Purpose: Like every other sector, educational institutions have also been suffering immensely due to COVID-19 pandemic. Many educational institutions are now adopting digital classroom services. However, an online platform with the need for appropriate technology and infrastructure from the students' perspective poses a severe challenge to developing countries like Bangladesh. The paper aims to figure out the relevant factors that affect the extent of student satisfaction with digital classroom services at the school and tertiary levels. Design/methodology/approach: It is a quantitative study of 450 students from Bangladesh who encountered online classes during the pandemic of COVID-19. An equal number of students from all levels, including schools, colleges and tertiary stages, participated in the survey. Exploratory and confirmatory factor analyses are used to interpret the data. Structural equation modeling using AMOS graphic software is incorporated to test the study's hypothesis. Findings: Among all the four determinants of student satisfaction during this critical era, all levels look satisfied with the three underlying influences: technological, convenience and resource-related factors. However, school-level students found the digital classroom services abrasive with Internet connectivity and technical structures during online classes and exams. Research limitations/implications: A comprehensive study can assess the difference between private and public university students in this regard. In addition, the impact of gender and/or location (rural/urban area) can be assessed by using the same model of the study. Practical implications: Having the experience of the students' satisfaction level during this pandemic, the government, educational institutions and other stakeholders can take away the findings of the results to have a better plan for Internet-based education at every level. Originality/value: The study is unique to see the readiness of developing nations such as Bangladesh to focus on the sudden uncertainty like a pandemic in introducing the digital education platform. The study can add value to achieving the country's sustainable development goal of becoming a digitally enabled regional education hub. © 2023, Md Abdul Momen, Seyama Sultana, Md. Anamul Hoque, Shamsul Huq Bin Shahriar and Abu Sadat Muhammad Ashif.

The FBProphet forecasting model to evaluate the spread of COVID-19 pandemic: A machine learning approach

The pandemic due to the COVID-19 virus causes financial disruption almost in all the countries in the world. It breaks the supply chain management system, especially in the food, energy and industrial sectors. This COVID-19 pandemic hits the world over different periods of time in a year and continues from the year 2020. It is important to predict the severity of the infection rate and death rate to prepare and take necessary actions for the future. But it is still difficult to forecast the outbreaks for the long term with higher accuracy. Here, a machine learning (ML) algorithm, FBProphet, is employed to forecast COVID-19 outbreaks in Bangladesh as a time series forecasting. This model predicts the daily and cumulative infection and death rates with a high and low error rate. Due to the seasonality feature of the FBProphet model, it can predict the different waves of outbreaks with higher accuracy. Furthermore, cross-validation of the predicted results has been performed to ensure the accuracy of the results. The effort and the outcomes of this computational study will help the decision makers of this country to take necessary actions for the future which can save lives and prevent economic disruption.

COVID-19 Detection Using Feature Extraction and Semi-Supervised Learning from Chest X-ray Images

The world has experienced the very first pandemic of 21st century, called the COVID-19 which is caused by a deadly virus named Coronavirus. In this regard, one of the very first strategy to minimize the number of affected patients and reduce casualties is to diagnose COVID-19 at an early stage. Currently, PCR test is primarily utilized for the diagnosis of COVID-19. However, PCR test requires a huge number of expensive test kits as well as trained experts. Therefore, chest X-Ray imaging technique (including Machine Learning) has been considered as an alternative for COVID-19 diagnosis among the researchers. This particular method is faster, less expensive and will allow the authorities to manage the COVID-19 diagnosis system in a cost-effective way. Machine learning techniques have been proven to be significantly efficient and accurate for image classification problems. On the other hand, One of the most utilized techniques in machine learning is supervised learning which is highly convenient and helping the experts to diagnose and make informed decisions about COVID-19. Supervised learning in image classification requires vast amount of radiography images with notable accuracy which can be a peculiar issue in medical domain. In order to address the problem, we have investigated a distinct approach for COVID-19 Diagnosis with a nominal dataset. In this work, We have studied the effectiveness of Semi-Supervised Learning (SSL) for COVID-19 diagnosis from chest X-ray images. We have investigated a prepossessing technique by extracting and combining local phase image feature into multi-feature image to train our SSL model in teacher/student archetype. Our study have shown that by using 17.0% of the total dataset for training, the SSL model achieve 93.45% accuracy. We also provide comparative metrics of SSL approach against other fully supervised techniques.

Investigating the Robustness of Deep Neural Network Based COVID-19 Detection Models Against Universal Adversarial Attacks

Corona viruses are a type of virus with a large family which can cause several terrible and devastating infectious diseases like middle east respiratory syndrome and severe acute respiratory syndrome. The first task of the authority is to screen as many people as possible to detect COVID-19 patients which arises the challenge of rapid screening. Although polymerase chain reaction(PCR) tests are primarily used for the COVID-19 test but because of it's high false negative results and need of experts leading to an alternative diagnostic system based on radiological images like chest X-ray. Moreover, computer aided diagnosis systems from radiography images has significantly been advanced during the last decade with promising efficiency which can overcome the need of both time and experts. In this case, machine learning(ML) and deep learning(DL) based screening techniques can provide automated, fast and reliable results. Therefore, many researchers have proposed several deep neural network(DNN) models for rapid screening of COVID-19 using chest X-ray images. Nevertheless, the vulnerability issue DNN models are overlooked or poorly evaluated in the COVID-19 screening. DNN models are remarkably vulnerable to perturbation which is addressed universal adversarial perturbation (UAP). UAP can falsely influence a DNN model and can eventually lead to going wrong in most of the classification problems. Here, we experimented and evaluated the performance of several DNN based automated COVID-19 diagnostic models, and investigated the robustness of these models against two types of adversarial attack:non targeted and targeted. We showed that DNN based COVID-19 detection models are highly vulnerable to adversarial attack and it is substantially important to be aware of the risk factors of DNN models before deploying for real life applications. © 2021 IEEE.

COVID-19 Impact on Poultry Production and Distribution Networks in Bangladesh

The COVID-19 pandemic has severely affected numerous economic sectors across the world, including livestock production. This study investigates how the pandemic has impacted the poultry production and distribution network (PDN), analyses stakeholders' changing circumstances, and provides recommendations for rapid and long-term resilience. This is based on a literature review, social media monitoring, and key informant interviews (n = 36) from across the poultry sector in Bangladesh. These included key informants from breeder farms and hatcheries, pharmaceutical suppliers, feed companies, dealers, farmers, middlemen, and vendors. We show that the poultry sector was damaged by the COVID-19 pandemic, partly as a result of the lockdown and also by rumors that poultry and their products could transmit the disease. This research shows that hardly any stakeholder escaped hardship. Disrupted production and transportation, declining consumer demand and volatile markets brought huge financial difficulties, even leading to the permanent closure of many farms. We show that the extent of the damage experienced during the first months of COVID-19 was a consequence of how interconnected stakeholders and businesses are across the poultry sector. For example, a shift in consumer demand in live bird markets has ripple effects that impact the price of goods and puts pressure on traders, middlemen, farmers, and input suppliers alike. We show how this interconnectedness across all levels of the poultry industry in Bangladesh makes it fragile and that this fragility is not a consequence of COVID-19 but has been revealed by it. This warrants long-term consideration beyond the immediate concerns surrounding the COVID-19 pandemic. © Copyright © 2021 Sattar, Mahmud, Mohsin, Chisty, Uddin, Irin, Barnett, Fournie, Houghton and Hoque.

Clustering-Based Pattern Analysis on COVID-19 Using K-Means Algorithm to Predict the Death Cases in Bangladesh

In December 2019, the COVID-19 rapidly infected virus disease was an outbreak in China. Woefully, the number of infected and death cases is increasing day after day. Still, there is no effective vaccine and antibiotic. As a result, our fighting starts with an invisible enemy. Up to July 31, 2020 the total number of death cases is 3111 (Male: 2446 and Female: 665), infected cases are 237,661, and recovered cases are 135,136 in Bangladesh. The most infected city is Dhaka with 47.83% death cases. Among our overall population, senior citizens and infants are more valuable assets for the nation. In this research work, the K-means algorithm was implemented to discover a pattern from collected data with the help of a data mining technique to predict the age level for death cases on COVID-19. Data were collected from 8 March 2020 to 31 July 2020 from authentic recognized sources. In this research work, R language is used for clustering on the K-means algorithm. The result analysis of this research shows that the most infected age range is above 60 years in this period, and they have a high possibility of death cases in the near future. © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.