id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-288303-88c6qsek	Paul, S. K.	On nonlinear incidence rate of Covid-19	2020-10-21		.txt	text/plain	7358	483	53	Classical Susceptible-Infected-Removed model with constant transmission rate and removal rate may not capture real world dynamics of epidemic due to complex influence of multiple external factors on the spread. Local interpretations of the model using perturbation method reveals local influence of different features on transmission rate which in turn is used to generate a set of generalized global interpretations. We experimented with available data of Covid-19 across multiple regions of USA and the model achieved 7.95% and 0.19% mean absolute percent error in terms of new infection cases in each locality and cumulative total infection cases across the country in a 10-day prediction period respectively. A long-term forecast using the trained model and modified recovery rate to satisfy disease-free equilibrium criteria reveals rapid damping of active infection cases to reach the baseline. The model predicts the transmission rate for a future time period for each pixel which in turn is used to calculate daily new infection cases ∆ + ( ) using equation 11.	./cache/cord-288303-88c6qsek.txt	./txt/cord-288303-88c6qsek.txt