首页|Swiss Tropical and Public Health Institute Reports Findings in Machine Learning (Modelling Europe-wide fine resolution daily ambient temperature for 2003-2020 using machine learning)
Swiss Tropical and Public Health Institute Reports Findings in Machine Learning (Modelling Europe-wide fine resolution daily ambient temperature for 2003-2020 using machine learning)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News - New research on Machine Learning is th e subject of a report. According to newsreporting originating from Allschwil, S witzerland, by NewsRx correspondents, research stated, “To improveour understan ding of the health impacts of high and low temperatures, epidemiological studies requirespatiotemporally resolved ambient temperature (Ta) surfaces. Exposure a ssessment over various Europeancities for multi-cohort studies requires high re solution and harmonized exposures over larger spatiotemporalextents.”Our news editors obtained a quote from the research from Swiss Tropical and Publ ic Health Institute,“Our aim was to develop daily mean, minimum and maximum amb ient temperature surfaces with a 1x 1 km resolution for Europe for the 2003-202 0 period. We used a two-stage random forest modellingapproach. Random forest wa s used to (1) impute missing satellite derived Land Surface Temperature(LST) us ing vegetation and weather variables and to (2) use the gap-filled LST together with land useand meteorological variables to model spatial and temporal variati on in Ta measured at weather stations.To assess performance, we validated these models using random and block validation. In addition toglobal performance, an d to assess model stability, we reported model performance at a higher granularity (local). Globally, our models explained on average more than 81 % and 93 % of the variability in theblock validation sets for LST a nd Ta respectively. Average RMSE was 1.3, 1.9 and 1.7 ℃ for mean,min and max a mbient temperature respectively, indicating a generally good performance. For Ta models,local performance was stable across most of the spatiotemporal extent, but showed lower performance inareas with low observation density. Overall, mod el stability and performance were lower when using blockvalidation compared to random validation.”
NETL
NSTL
