Dr. Santosh Ray, Head /Associate Professor, KIC IT Department, has collaborated on a project that aims to predict confirmed COVID-19 cases.
“Knowing the accurate number of COVID patients in advance will help hospitals and local governments to prepare themselves well to serve COVID patients in an effective manner,” commented Dr. Ray. “Deep learning models usually make inaccurate predictions if they are trained with insufficient data. Therefore, in the circumstances where sufficient data is not available especially during the first few months of COVID pandemic, classical machine learning models trained on different types of data such as times series data, social media data, etc. may be combined to predict the number of confirmed cases. ”
The project was led by Dr. Amir Ahmad, Associate Professor of Information Systems and Security in the College of Information Technology at United Arab Emirates University (UAEU), in collaboration with international researchers.
The team conducted a meta-analysis of associated COVID-19 case-prediction models to propose performance improvements, by leveraging machine learning to predict COVID-19 cases, and ultimately address the global pandemic.
Predicting COVID-19 cases is crucial to controlling the pandemic. Accurate predictions enable concerned decision-makers to optimize planning and carry out effective countermeasures.
However, according to Dr. Amir Ahmad, current prediction models are limited, and these limitations should be addressed.
The team published a paper on their work in the journal Archives of Computational Methods in Engineering.
Read more: Innovation@UAE Magazine, Issue 1, English, December 2020 by UAE Ministry of Education.