{"@type": "dcat:Dataset", "accessLevel": "public", "accrualPeriodicity": "irregular", "bureauCode": ["026:00"], "contactPoint": {"@type": "vcard:Contact", "fn": "Elizabeth Foughty", "hasEmail": "mailto:elizabeth.a.foughty@nasa.gov"}, "description": "UNDERSTANDING SEVERE WEATHER PROCESSES THROUGH\r\nSPATIOTEMPORAL RELATIONAL RANDOM FORESTS\r\n\r\nAMY MCGOVERN, TIMOTHY SUPINIE, DAVID JOHN GAGNE II, NATHANIEL TROUTMAN,\r\nMATTHEW COLLIER, RODGER A. BROWN, JEFFREY BASARA, AND JOHN K. WILLIAMS\r\n\r\nAbstract. Major severe weather events can cause a significant loss of life and property. We\r\nseek to revolutionize our understanding of and ability to predict such events through the mining\r\nof severe weather data. Because weather is inherently a spatiotemporal phenomenon, mining\r\nsuch data requires a model capable of representing and reasoning about complex spatiotemporal\r\ndynamics, including temporally and spatially varying attributes and relationships. We introduce\r\nan augmented version of the Spatiotemporal Relational Random Forest, which is a Random Forest\r\nthat learns with spatiotemporally varying relational data. Our algorithm maintains the strength\r\nand performance of Random Forests but extends their applicability, including the estimation of\r\nvariable importance, to complex spatiotemporal relational domains. We apply the augmented\r\nSpatiotemporal Relational Random Forest to three severe weather data sets. These are: predicting\r\natmospheric turbulence across the continental United States, examining the formation of tornadoes\r\nnear strong frontal boundaries, and understanding the translation of drought across the southern\r\nplains of the United States. The results on such a wide variety of real-world domains demonstrate\r\nthe extensive applicability of the Spatiotemporal Relational Random Forest. Our long-term goal\r\nis to significantly improve the ability to predict and warn about severe weather events.", "distribution": [{"@type": "dcat:Distribution", "description": "UNDERSTANDING SEVERE WEATHER PROCESSES THROUGH SPATIOTEMPORAL RELATIONAL RANDOM FORESTS", "downloadURL": "https://c3.nasa.gov/dashlink/static/media/publication/Paper_17_.pdf", "format": "PDF", "mediaType": "application/pdf", "title": "Paper 17 .pdf"}], "identifier": "DASHLINK_239", "issued": "2010-10-13", "keyword": ["ames", "dashlink", "nasa"], "landingPage": "https://c3.nasa.gov/dashlink/resources/239/", "modified": "2025-04-01", "programCode": ["026:029"], "publisher": {"@type": "org:Organization", "name": "Dashlink"}, "title": "UNDERSTANDING SEVERE WEATHER PROCESSES THROUGH SPATIOTEMPORAL RELATIONAL RANDOM FORESTS"}