TY - JOUR
T1 - Two-step light gradient boosted model to identify human west nile virus infection risk factor in Chicago
AU - Wan, Guangya
AU - Allen, Joshua
AU - Ge, Weihao
AU - Rawlani, Shubham
AU - Uelmen, John
AU - Mainzer, Liudmila Sergeevna
AU - Smith, Rebecca Lee
N1 - The first author was supported by the Students Pusing Innovation Internship (SPIN) at the National Center for Supercomputing Applications. https://spin.ncsa.illinois.edu/ The project was funded by the NCSA Center-Directed Discretionary Research (CDDR). The authors would like to thank the HAL cluster and support team for providing the computational resources to complete the work. The author would also like to acknowledge the efforts of the NCSA Industry Group for supporting the work. The authors would like to thank Dr. Christina Fliege for her editorial suggestions on this manuscript. The authors would like to thank Mr. Mingyu Yang for his help in retrieving and preprocessing the census data.
PY - 2024/1/5
Y1 - 2024/1/5
N2 - West Nile virus (WNV), a flavivirus transmitted by mosquito bites, causes primarily mild symptoms but can also be fatal. Therefore, predicting and controlling the spread of West Nile virus is essential for public health in endemic areas. We hypothesized that socioeconomic factors may influence human risk from WNV. We analyzed a list of weather, land use, mosquito surveillance, and socioeconomic variables for predicting WNV cases in 1-km hexagonal grids across the Chicago metropolitan area. We used a two-stage lightGBM approach to perform the analysis and found that hexagons with incomes above and below the median are influenced by the same top characteristics. We found that weather factors and mosquito infection rates were the strongest common factors. Land use and socioeconomic variables had relatively small contributions in predicting WNV cases. The Light GBM handles unbalanced data sets well and provides meaningful predictions of the risk of epidemic disease outbreaks.
AB - West Nile virus (WNV), a flavivirus transmitted by mosquito bites, causes primarily mild symptoms but can also be fatal. Therefore, predicting and controlling the spread of West Nile virus is essential for public health in endemic areas. We hypothesized that socioeconomic factors may influence human risk from WNV. We analyzed a list of weather, land use, mosquito surveillance, and socioeconomic variables for predicting WNV cases in 1-km hexagonal grids across the Chicago metropolitan area. We used a two-stage lightGBM approach to perform the analysis and found that hexagons with incomes above and below the median are influenced by the same top characteristics. We found that weather factors and mosquito infection rates were the strongest common factors. Land use and socioeconomic variables had relatively small contributions in predicting WNV cases. The Light GBM handles unbalanced data sets well and provides meaningful predictions of the risk of epidemic disease outbreaks.
UR - http://www.scopus.com/inward/record.url?scp=85181630549&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85181630549&partnerID=8YFLogxK
U2 - 10.1371/journal.pone.0296283
DO - 10.1371/journal.pone.0296283
M3 - Article
C2 - 38181002
AN - SCOPUS:85181630549
SN - 1932-6203
VL - 19
JO - PloS one
JF - PloS one
IS - 1 January
M1 - e0296283
ER -