You are here

Keyword: machine learning

Reclassifying the wildland–urban interface using fire occurrences for the United States

Science Spotlights Posted on: July 31, 2020
The wildland–urban interface (WUI) occurs at the intersection of houses and undeveloped wildlands, where fire is a safety concern for communities. Previous definitions of the WUI do not explicitly account for differences in fire risk, but data are now available to use objective measures of fire occurrence to refine the definition by assessing the housing densities where fires actually occurred. 

Reclassifying the wildland-urban interface using fire occurrences for the United States

Publications Posted on: July 22, 2020
The wildland–urban interface (WUI) occurs at the intersection of houses and undeveloped wildlands, where fire is a safety concern for communities, motivating investment in planning, protection, and risk mitigation.

Mapping causes of disturbance in U.S. forests

Science Spotlights Posted on: June 18, 2020
A collaborative project between USFS FIA, NASA, and several universities has developed a new, national map attributing the cause and timing of forest canopy cover losses occurring between 1986 and 2010 across the conterminous United States. The models separated areas of stable forest from areas experiencing persisting forest cover loss (e.g. conversion of forest to other land uses), temporary forest cover loss due to abrupt changes (fire, removals, wind), or long gradual declines (stress from insects and disease).

Modelling post-fire tree mortality: Can random forest improve discrimination of imbalanced data?

Publications Posted on: May 10, 2020
Predicting post-fire tree mortality is a major area of research in fire-prone forests, woodlands, and savannas worldwide. Past research has relied overwhelmingly on logistic regression analysis (LR) that predicts post-fire tree status as a binary outcome (i.e. living or dead). One of the most problematic issues for LR (or any classification problem) occurs when there is a class imbalance in the training data.

Random forests for classification in ecology

Publications Posted on: September 10, 2019
Classification procedures are some of the most widely used statistical methods in ecology. Random forests (RF) is a new and powerful statistical classifier that is well established in other disciplines but is relatively unknown in ecology.

Function modeling improves the efficiency of spatial modeling using big data from remote sensing

Publications Posted on: July 19, 2017
Spatial modeling is an integral component of most geographic information systems (GISs). However, conventional GIS modeling techniques can require substantial processing time and storage space and have limited statistical and machine learning functionality. To address these limitations, many have parallelized spatial models using multiple coding libraries and have applied those models in a multiprocessor environment.