Back

On Modeling of COVID-19 for the Indian Subcontinent using Polynomial and Supervised Learning Regression

Neve, D.; Patel, H.; Dhiman, H. S.

2020-10-16 public and global health
10.1101/2020.10.14.20212563 medRxiv
Show abstract

COVID-19, a recently declared pandemic by WHO has taken the world by storm causing catastrophic damage to human life. The novel cornonavirus disease was first incepted in the Wuhan city of China on 31st December 2019. The symptoms include fever, cough, fatigue, shortness of breath or breathing difficulties, and loss of smell and taste. Since the devastating phenomenon is essentially a time-series representation, accurate modeling may benefit in identifying the root cause and accelerate the diagnosis. In the current analysis, COVID-19 modeling is done for the Indian subcontinent based on the data collected for the total cases confirmed, daily recovered, daily deaths, total recovered and total deaths. The data is treated with total confirmed cases as the target variable and rest as feature variables. It is observed that Support vector regressions yields accurate results followed by Polynomial regression. Random forest regression results in overfitting followed by poor Bayesian regression due to highly correlated feature variables. Further, in order to examine the effect of neighbouring countries, Pearson correlation matrix is computed to identify geographic cause and effect.

Matching journals

The top 7 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.