Back

Forecasting COVID-19 New Cases Using Transformer Deep Learning Model

Patil, S.; Mollaei, P.; farimani, A. B.

2023-11-03 health informatics
10.1101/2023.11.02.23297976 medRxiv
Show abstract

Making accurate forecasting of COVID-19 cases is essential for healthcare systems, with more than 650 million cases as of 4 January,1 making it one of the worst in history. The goal of this research is to improve the precision of COVID-19 case predictions in Russia, India, and Brazil, a transformer-based model was developed. Several researchers have implemented a combination of CNNs and LSTMs, Long Short-Term Memory (LSTMs), and Convolutional Neural Networks (CNNs) to calculate the total number of COVID-19 cases. In this study, an effort was made to improve the correctness of the models by incorporating recent advancements in attention-based models for time-series forecasting. The resulting model was found to perform better than other existing models and showed improved accuracy in forecasting. Using the data from different countries and adapting it to the model will enhance its ability to support the worldwide effort to combat the pandemic by giving more precise projections of cases.

Matching journals

The top 11 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.