This project helps the government of Malaysia make use of their public dataset on dengue (mosquito-borne disease) to predict future outbreaks using machine learning and GIS tools.
- Exploratory analysis
categorical - by state, by city
quantitative - time series, statistics
Graphical - histograms, heat maps
- add external data
- rainfall, by state
- weather / temp
- humidity
- population
Questions:
what percent of population has been infected?
what are the variables correlated with incidence?
what is the probability incidence
what is the error in the test set