dengAI

Predicting Disease Spread

Introduction

Dengue fever is a mosquito-borne life threatening disease that occurs mostly in tropical and subtropical regions of the world. Since the disease is carried by mosquitoes, the transmission dynamics of dengue are related to climate variables such as temperature and precipitation which might help in reproduction of mosquitoes. Although the relation between those are complex and many researchers say that the climate change has a considerable effect on distributional shifts of reported cases. Therefore, if there is a way to estimate the relationship between dengue dynamics and climate changes, the research initiatives and resource allocation to help fight life-threatening pandemics can be improved. The main consideration of this project is to predict the possible number of dengue cases for San Juan, Puerto Rico and Iquitos in Peru using data mining aspects. This task is available as a data science competition in www.drivedata.org website.

About data

/data directory contains the data related to this model, followings are the attributes that they provided.

1. San Juan as sj or Iquitos as iq
2. year : Year of the recorded value
3. weekofyear : Week of the year of recorded value
4. week_start_date : First date of the week
5. total_cases : Total dengue cases reported
6. ndvi_se : Normalized difference vegetation index for south east side from city centroid
7. ndvi_sw : Normalized difference vegetation index for south west side from city centroid
8. ndvi_ne : Normalized difference vegetation index for north east side from city centroid
9. ndvi_nw : Normalized difference vegetation index for north west side from city centroid
10. precipitation_amt_mm : Total precipitation amount (PERSIANN)
11. reanalysis_sat_precip_amt_mm
12. reanalysis_air_temp_k : Mean air temperature (NCEP)
13. reanalysis_avg_temp_k : Average air temperature (NCEP)
14. reanalysis_dew_point_temp_k : Mean dew point temperature (NCEP)
15. reanalysis_max_air_temp_k : Maximum air temperature (NCEP)
16. reanalysis_min_air_temp_k : Minimum air temperature (NCEP)
17. reanalysis_precip_amt_kg_per_m2 : Total precipitation amount (NCEP)
18. reanalysis_relative_humidity_percent : Mean relative humidity (NCEP)
19. reanalysis_specific_humidity_g_per_kg : Mean specific humidity (NCEP)
20. reanalysis_tdtr_k : Diurnal temperature range (NCEP)
21. station_avg_temp_c : Average temperature (GHCN)
22. station_diur_temp_rng_c : Diurnal temperature range (GHCN)
23. station_max_temp_c : Maximum air temperature (GHCN)
24. station_min_temp_c : Minimum air temperature (GHCN)
25. station_precip_mm : Total precipitation (GHCN)

Getting Started..

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

Linux pc with python3 installed 
keras with Tensorflow backend
scikit-learn toolkit for python
Other python libraries like numpy, matplotlib(for visualizations), pandas, skimage etc..

Please refer to the documentations of above tools and install latest versions that they supported. We recommand to install them in a seperate python virtual environment for convenience. Please note that If you are willing to run prophet.py you should install prophet library as follows.

# bash
$ pip install fbprophet

For more details, refer to this: https://facebook.github.io/prophet/docs/installation.html

Do simple Inference

To run Bayesian Ridge model

python3 final_regr.py

If you want to use ipython notebook

final_regr.ipynb

To run Time Series prediction

python3 prophet.py

Note: Our final presentation can be found on final_presentation.pdf

Hope you enjoy well....