Title :
Can Topic Modeling Shed Light on Climate Extremes?
Author :
Cheng Tang ; Monteleoni, Claire
Abstract :
Understanding changes in climate extremes is an urgent challenge. Topic modeling techniques from natural language processing can help scientists learn climate patterns from data. The authors´ work extracts global climate patterns from multivariate climate data, modeling relations between variables via latent topics and discovering the probability of each climate topic appearing at different geographical locations.
Keywords :
climatology; data analysis; geophysics computing; natural language processing; climate extremes; climate pattern probability; geographical locations; global climate pattern extraction; latent topics; multivariate climate data; natural language processing; topic modeling techniques; Atmospheric modeling; Computational modeling; Data models; Hidden Markov models; Meteorology; Tensile stress; climate extremes; climate informatics; latent Dirichlet allocation; machine learning; scientific computing; topic models; unsupervised learning;
Journal_Title :
Computing in Science Engineering
DOI :
10.1109/MCSE.2015.128
