@techreport{
 title = {Multi-class ROC Analysis},
 type = {techreport},
 source = {Weather},
 city = {Tucson, AZ},
 institution = {The University Of Arizona},
 id = {11ff9e4b-6753-3dce-9349-19a2d8da3c8c},
 created = {2009-08-19T05:38:36.000Z},
 file_attached = {true},
 profile_id = {7edb4dd2-367e-3b20-8e26-a7e1d668fa33},
 group_id = {3f9481eb-d6d1-360a-a41c-cd46dd294d2e},
 last_modified = {2011-04-23T07:06:28.000Z},
 read = {false},
 starred = {false},
 authored = {false},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wandishin},
 abstract = {Receiver Operating Characteristic (ROC) curves have become a common analysis tool for evaluating forecast discrimination: the ability of a forecast system to distinguish between events and non-events.  As is implicit in that statement, application of the ROC curve is limited to forecasts involving only two possible outcomes, such as rain and no rain.  However, many forecast scenarios exist for which there are multiple possible outcomes, such as rain, snow, and freezing rain.  An extension of the ROC curve to multi-class forecast problems is explored.  The full extension involves high-dimensional hyper-surfaces that cannot be visualized and that present other problems.  Therefore, several different approximations to the full extension are introduced using both artificial and actual forecast data sets.  These approximations range from sets of simple two-class ROC curves to sets of three-dimensional ROC surfaces.  No single approximation is superior for all forecast problems, thus the specific aims in evaluating the forecast must be considered.},
 bibtype = {techreport},
 author = {Wandishin, Matthew S. and Mullen, Steven J.}
}

{"_id":"CpwYJXtp4JWgE5GRP","bibbaseid":"wandishin-mullen-multiclassrocanalysis","downloads":0,"creationDate":"2016-03-15T05:23:31.117Z","title":"Multi-class ROC Analysis","author_short":["Wandishin, M., S.","Mullen, S., J."],"year":null,"bibtype":"techreport","biburl":null,"bibdata":{"title":"Multi-class ROC Analysis","type":"techreport","source":"Weather","city":"Tucson, AZ","institution":"The University Of Arizona","id":"11ff9e4b-6753-3dce-9349-19a2d8da3c8c","created":"2009-08-19T05:38:36.000Z","file_attached":"true","profile_id":"7edb4dd2-367e-3b20-8e26-a7e1d668fa33","group_id":"3f9481eb-d6d1-360a-a41c-cd46dd294d2e","last_modified":"2011-04-23T07:06:28.000Z","read":false,"starred":false,"authored":false,"confirmed":"true","hidden":false,"citation_key":"Wandishin","abstract":"Receiver Operating Characteristic (ROC) curves have become a common analysis tool for evaluating forecast discrimination: the ability of a forecast system to distinguish between events and non-events. As is implicit in that statement, application of the ROC curve is limited to forecasts involving only two possible outcomes, such as rain and no rain. However, many forecast scenarios exist for which there are multiple possible outcomes, such as rain, snow, and freezing rain. An extension of the ROC curve to multi-class forecast problems is explored. The full extension involves high-dimensional hyper-surfaces that cannot be visualized and that present other problems. Therefore, several different approximations to the full extension are introduced using both artificial and actual forecast data sets. These approximations range from sets of simple two-class ROC curves to sets of three-dimensional ROC surfaces. No single approximation is superior for all forecast problems, thus the specific aims in evaluating the forecast must be considered.","bibtype":"techreport","author":"Wandishin, Matthew S. and Mullen, Steven J.","bibtex":"@techreport{\n title = {Multi-class ROC Analysis},\n type = {techreport},\n source = {Weather},\n city = {Tucson, AZ},\n institution = {The University Of Arizona},\n id = {11ff9e4b-6753-3dce-9349-19a2d8da3c8c},\n created = {2009-08-19T05:38:36.000Z},\n file_attached = {true},\n profile_id = {7edb4dd2-367e-3b20-8e26-a7e1d668fa33},\n group_id = {3f9481eb-d6d1-360a-a41c-cd46dd294d2e},\n last_modified = {2011-04-23T07:06:28.000Z},\n read = {false},\n starred = {false},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Wandishin},\n abstract = {Receiver Operating Characteristic (ROC) curves have become a common analysis tool for evaluating forecast discrimination: the ability of a forecast system to distinguish between events and non-events. As is implicit in that statement, application of the ROC curve is limited to forecasts involving only two possible outcomes, such as rain and no rain. However, many forecast scenarios exist for which there are multiple possible outcomes, such as rain, snow, and freezing rain. An extension of the ROC curve to multi-class forecast problems is explored. The full extension involves high-dimensional hyper-surfaces that cannot be visualized and that present other problems. Therefore, several different approximations to the full extension are introduced using both artificial and actual forecast data sets. These approximations range from sets of simple two-class ROC curves to sets of three-dimensional ROC surfaces. No single approximation is superior for all forecast problems, thus the specific aims in evaluating the forecast must be considered.},\n bibtype = {techreport},\n author = {Wandishin, Matthew S. and Mullen, Steven J.}\n}","author_short":["Wandishin, M., S.","Mullen, S., J."],"urls":{"Paper":"http://bibbase.org/service/mendeley/7edb4dd2-367e-3b20-8e26-a7e1d668fa33/file/026023ba-9a6a-01db-11a2-d53c359affc5/Multi-class_ROC_Analysis.pdf.pdf"},"bibbaseid":"wandishin-mullen-multiclassrocanalysis","role":"author","downloads":0},"search_terms":["multi","class","roc","analysis","wandishin","mullen"],"keywords":[],"authorIDs":[]}