A couple of professors from IIT Kanpur have come up with an extremely flawed defense of the UP Model of dealing with the disaster suffered by the state especially during the second wave of the pandemic. The data science engineers of our union have seen the report and the shortcoming of the entire report is pretty transparent. Let us have a look at some of these issues.
Firstly, the paper on the much hyped ‘SUTRA’ model which adds a fraction of ‘Removed’ cases to ‘Undetected’ cases in order to try and show that it has correctly accounted for undetected cases uses various parameters like contact rate, detection rate, removal rate and reach to account for these changes has already mentions that all these parameters are subject to change during the pandemic.
For example, as per SUTRA model, the second wave was never imminent. However, towards the end of April, when they realised that the contact rate has increased from 0.25 to 0.38, they simply updated it and said that they have correctly predicted the second wave curve to peak around 8th of May. The paper itself mentions that it can only make predictions as long as the parameters are not changing as its motivation, so any prediction of a wave based on unforeseen events is beyond the scope of the model.
Moreover, the bias in the visualisation of data is also to be noted. Since the second wave had a much higher peak than any other, smaller peaks would only seem like a flat curve in the visualisation over a timeline which includes the second wave.
The model also mentions that it has only taken into account government statistics on cases and deaths, which largely downplays the impact of the pandemic. There is a complete denial of horrific images of dead bodies floating in River Ganga and the estimates of deaths by any statistician for India has been at least 5 times greater than what has been reported if not way more.
To further cover up the utter negligence of both the Uttar Pradesh and central governments, it focuses its analysis on the 4 lakh returning migrant workers to the state after the onset of the second wave to account for undetected cases which might have changed the parameter. It completely ignores the participation of crores of people in the Panchayat elections around mid-April and the returnees from Kumbh congregation where around 1 crore people had assembled and no protocol was put up in the state for the returnees. Also, a number of fake COVID negative reports for people participating in Kumbh were reported later on.
The entire report is full of methodical flaws and is a poor act in trying to cover up the inactions of the government when countless people of the state and the country perished. It has been released strategically before the upcoming Vidhansabha elections in the state.
AIITEU strongly condemns the attempt of using the name of a prestigious institute of the country to serve the diabolical political agenda of the ruling party in the state and the centre by a simple misuse of data and data models. We also call upon the IT sector employees to sign the undertaking by academicians and researchers against it.
This form is collecting signatures endorsing the following statement. For this statement in Hindi, please refer to this document. This statement is meant to be read in conjunction with a detailed critique of the IIT Kanpur report prepared by IITK Citizen’s Forum and Hamara Manch IITK.