Relation between Population Density and spread:
Higher the number of people per area of a region, more is the ease of spread due to community transmission. This hypothesis holds true for the most populous states of the US like New York, New Jersey, where there is a constant movement of people. Given the time taken to identify the infected person, they would interact with more people than in sparsely populated area.
Relation between Urbanization and spread:
Urban cities have sophisticated infrastructure, healthcare and transport system. At any given point of time in a day, buses, trains, carpool are densely packed with large number of people. Urbanization is closely associated with more employment for both skilled and non skilled workers. This in turn oversees a high rate of migration into Urban cities for livelihood
Relation between Literacy and spread:
Literacy rate plays a major role in measuring effective transfer of latest health advisories and preventive measures from Health organizations and Government. Staying updated in terms of constantly changing norms and comprehending the information is equally important to be aligned with the nations war against covid-19. In that case, population with less literacy is often prone to be less aware of the changing laws
Global impact of the pandemic motivated us to try and find patterns and trends affecting the spread of the virus, which allows us for predicting pandemic prone regions. This will help us to track down the most vulnerable places and allows us to apply the norms at the right time.
We developed a interactive web application which closely analysed the patterns between Human activities and Covid-19 cases. It comprises of hotspot predictor which takes population density, urbanization and literacy as chief factors. The hotspot can predict if any particular city/state in india could be a potential hotspot. In determining the hotspot we have given population density the highest priority followed by the urbanisation and literacy rate. We used the binary numbers concept and placed the most significant factor like population density in the higher scale of 2 power n -1 where n is the number of factors we take into account. and next priority factor will take the scale of 2 power n -2 and so on. When the entered values of these factors exceed a certain threshold we give them binary value of 1 or else 0 if less than threshold i.e. like 1 0 1 value which tells density is high , urbanisation is lower than threshold and literacy rate is fine and the corresponding 2 power values will be calculated to give the final value . After we went through/analyzed the data we could come up with this idea to position the most dominant factor with highest priortiy and highest power value .
When the user enter his city's details about three factors we try to predict if it is going to be potential hotspot or not.
Data Resources:
We utilized the open data sources of data.census.gov for US dataset having key indicators like Population Density, Commuting methods of working class, Education Attainment.
Tools :
For Frontend we've used Angular CLI 6.1.5, node js and Node package manager
For the Backend, we've used Java, Spring Webapp, Spring JPA and Hibernate
For persistence we used MySQL.All of the dataset we used are present in the AWS RDS instance.
Challenges :
1) Aggregating data from various sources to justify the above mentioned factors
2) Could not completely leverage Petabytes of Space data resources
Achievements :
1) Web application covering all the various factors
2) Easy to interpret graphical information
3) Fair prediction model of hotspots in India
Improvements:
1) If we had more time we would have developed the hotspot predictor to more on the global level and with better accuracy
India Dataset : https://censusindia.gov.in/
USA Dataset : https://data.census.gov/cedsci/
Information on Covid cases in India , USA and Croatia : Google Search, https://www.worldometers.info/coronavirus