The prediction of coronavirus disease 2019 outbreak on Bangladesh perspective using machine learning: a comparative study

Abstract

Coronavirus disease 2019 (COVID-19) has made a huge pandemic situation in many countries of the world including Bangladesh. If the increase rate of this threat can be forecasted, immediate measures can be taken. This study is an effort to forecast the threat of present pandemic situation using machine learning (ML) forecasting models. Forecasting was done in three categories in the next 30 days range. In our study, multiple linear regression performed best among the other algorithms in all categories with R2 score of 99% for first two categories and 94% for the third category. Ridge regression performed great for the first two categories with R2 scores of 99% each but performed poorly for the third category with R2 score of 43%. Lasso regression performed reasonably well with R2 scores of 97%, 99% and 75% for the three categories. We also used Facebook Prophet to predict 30 days beyond our train data which gave us healthy R2 scores of 92% and 83% for the first two categories but performed poorly for the third category with R2 score of 34%. Also, all the models’ performances were evaluated with a 40-day prediction interval in which multiple linear regression outperformed other algorithms.

Stability of classification performance on an adaptive neuro fuzzy inference system for disease complication prediction

“…One of the algorithms that combine the neural network and the fuzzy system is called ANFIS. This research supports the evidence that the application of intelligent systems for classification in the medical field is very appropriate, especially for machine learning [34]- [36].…”

Support system of self-assessment and gap analysis for new normal tourism standards

“…GWL3.4 "Having reports on the amount of monthly waste used for recycling" relates to WMD1. 3 The GWL and NWL, WMD and mapping rules are stored in a separate sheet as shown in Table 1 and automatically imported into the system's knowledge base, which is used as a decision-making table in the system in research studies. Stored outside the system, further knowledge can be updated from the experts in that knowledge domain separately from the program.…”

“…Red/Failed, 61-80% is Light Green/Passed and more than 80% is Dark Green/Excellent. Calculate the score of Each GWL in the WL under the GNP standard [26] in (3):…”

“…The global tourism outlook for 2020 reflects a decline in tourism revenue from $910 billion to $1.2 trillion, as well as a decline in global GDP from 1.5% to 2.8% [2]. However, the studies have reported that during the COVID-19 pandemic, nature and ecosystems have been recovering from greenhouse gas emissions resulting in travel restrictions and changes in human living habits [3], [4]. Although the decline in tourism contributes to revenue in the tourism industry, it has helped the ecosystem recover from destruction and deteriorating conditions.…”

Virtual analysis of machine learning models for diseases prediction in muskmelon

“…Step 6: at this stage, various classification algorithms are applied to the dataset. These algorithms, including support vector machines (SVM), decision trees, back propagation networks, XGBoost, and random forest [36] [40] are carefully chosen for their ability to handle classification tasks effectively. This diversity allows us to explore a range of methods, maximizing the chances of finding the most suitable one for our specific problem.…”