Unusuala1l2e3x4 / Research-Spring2021

Investigating how carbon emissions, particulate matter, and climate variables/indices impact mortality from chronic respiratory disease. Working with pollutant, climate, mortality, population, and geographic datasets. Modeling with Random Forest regression.

Citation

A. He and T. Munasinghe, "Chronic Respiratory Disease: Risk Modeling Potential and Limitations," in 2021 IEEE International Conference on Big Data (Big Data), Orlando, FL, USA, 2021 pp. 1045-1053.

doi: 10.1109/BigData52589.2021.9672074

keywords: {microorganisms;temperature;pulmonary diseases;big data;water pollution;data models;spatiotemporal phenomena}

url's:

• https://ieeexplore.ieee.org/document/9672074
• https://doi.ieeecomputersociety.org/10.1109/BigData52589.2021.9672074

Research-Spring2021

[Undergraduate Research Project]

To get/construct GFED datasets:

• Get original data:
  • Download original GFED data from https://www.geo.vu.nl/~gwerf/GFED/GFED4/, place in folder "GFED4s"
• Construct GFED data with appropriate datasets multiplied by cell area:
  • Copy all .hdf5 files from the original GFED data (from folder "GFED4s") into new folder "GFED4s_timesArea"
  • Run "code/multiply_by_area_gfed4s.py"

To get 2000-2018 Surface PM2.5 datasets (monthly, 0.01 deg resolution)

• Corresponding article: https://pubs.acs.org/doi/full/10.1021/acs.est.0c01764
• More info + instructions: https://sites.wustl.edu/acag/datasets/surface-pm2-5/

To get mortality data (death counts by county, month):

• CDC WONDER - Underlying Cause of Death - Chronic lower respiratory diseases - (saved query): https://wonder.cdc.gov/controller/saved/D76/D133F078

To write all data files (after downloading original data files from source):

1. Run "adjust_sup_deaths_data_by_pop.py"
2. Run "read_acag_pm2-5.py", "read_gfed4s.py"
3. Run "write_county_month_pm2-5.py"
4. Run, in any order:
   • Run "write_county_month_gfed.py"
   • Run "write_county_month_clim.py"
   • Run "write_county_month_median-income.py"
5. To write AQI data (optional)
   • Run "impute_county_month_AQI.py"
   • Run "write_county_month_AQI_main.py"

If in any case, you encounter error message where a directory does not exist, create it in the path described

To run Random Forest and RFECV (recursive feature elimination and cross-validated selection)

• To tune/test hyperparameters and static combinations of features, run "random_forest.py"
  • Set hyperparameters by editing "param_grid" variable
  • Set combinations of features by editing "columns_list" variable
• To perform feature selection, run "random_forest_RFECV.py"
  • Adjust starting features by editing "columns" variable
  • Note: not for tuning hyperparameters due to runtime; hyperparameters ("param_grid" variable) can be set for a single iteration of RFECV