@article{
 title = {A Response to Webb and Ting's On the Application of ROC Analysis to Predict Classification Performance Under Varying Class Distributions},
 type = {article},
 year = {2005},
 identifiers = {[object Object]},
 keywords = {class skew,classi cation,classi er evaluation,roc},
 pages = {33-38},
 volume = {58},
 websites = {http://www.springerlink.com/index/10.1007/s10994-005-5256-4},
 month = {1},
 id = {7fd06c5f-1e23-3c3f-a568-32d6adb56296},
 created = {2009-08-18T04:22:48.000Z},
 file_attached = {true},
 profile_id = {7edb4dd2-367e-3b20-8e26-a7e1d668fa33},
 last_modified = {2017-07-27T21:18:51.867Z},
 tags = {ROC,class skew 1.,classification,classifier evaluation},
 read = {true},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Fawcett2005a},
 private_publication = {false},
 abstract = {In an article in this issue,Webb and Ting criticize ROC analysis for its inability to handle certain changes in class distributions. They imply that the ability of ROC graphs to depict performance in the face of changing class distributions has been overstated. In this editorial response, we describe two general types of domains and argue that Webb and Ting’s concerns apply primarily to only one of them. Furthermore, we show that there are interesting real-world domains of the second type, in which ROC analysis may be expected to hold in the face of changing class distributions.},
 bibtype = {article},
 author = {Fawcett, Tom and Flach, Peter A.},
 journal = {Machine Learning},
 number = {1}
}

{"_id":"2xaiwXj7FksKnZBRz","bibbaseid":"fawcett-flach-aresponsetowebbandtingsontheapplicationofrocanalysistopredictclassificationperformanceundervaryingclassdistributions-2005","downloads":0,"creationDate":"2016-03-15T04:09:55.070Z","title":"A Response to Webb and Ting's On the Application of ROC Analysis to Predict Classification Performance Under Varying Class Distributions","author_short":["Fawcett, T.","Flach, P., A."],"year":2005,"bibtype":"article","biburl":null,"bibdata":{"title":"A Response to Webb and Ting's On the Application of ROC Analysis to Predict Classification Performance Under Varying Class Distributions","type":"article","year":"2005","identifiers":"[object Object]","keywords":"class skew,classi cation,classi er evaluation,roc","pages":"33-38","volume":"58","websites":"http://www.springerlink.com/index/10.1007/s10994-005-5256-4","month":"1","id":"7fd06c5f-1e23-3c3f-a568-32d6adb56296","created":"2009-08-18T04:22:48.000Z","file_attached":"true","profile_id":"7edb4dd2-367e-3b20-8e26-a7e1d668fa33","last_modified":"2017-07-27T21:18:51.867Z","tags":"ROC,class skew 1.,classification,classifier evaluation","read":"true","starred":false,"authored":"true","confirmed":"true","hidden":false,"citation_key":"Fawcett2005a","private_publication":false,"abstract":"In an article in this issue,Webb and Ting criticize ROC analysis for its inability to handle certain changes in class distributions. They imply that the ability of ROC graphs to depict performance in the face of changing class distributions has been overstated. In this editorial response, we describe two general types of domains and argue that Webb and Ting’s concerns apply primarily to only one of them. Furthermore, we show that there are interesting real-world domains of the second type, in which ROC analysis may be expected to hold in the face of changing class distributions.","bibtype":"article","author":"Fawcett, Tom and Flach, Peter A.","journal":"Machine Learning","number":"1","bibtex":"@article{\n title = {A Response to Webb and Ting's On the Application of ROC Analysis to Predict Classification Performance Under Varying Class Distributions},\n type = {article},\n year = {2005},\n identifiers = {[object Object]},\n keywords = {class skew,classi cation,classi er evaluation,roc},\n pages = {33-38},\n volume = {58},\n websites = {http://www.springerlink.com/index/10.1007/s10994-005-5256-4},\n month = {1},\n id = {7fd06c5f-1e23-3c3f-a568-32d6adb56296},\n created = {2009-08-18T04:22:48.000Z},\n file_attached = {true},\n profile_id = {7edb4dd2-367e-3b20-8e26-a7e1d668fa33},\n last_modified = {2017-07-27T21:18:51.867Z},\n tags = {ROC,class skew 1.,classification,classifier evaluation},\n read = {true},\n starred = {false},\n authored = {true},\n confirmed = {true},\n hidden = {false},\n citation_key = {Fawcett2005a},\n private_publication = {false},\n abstract = {In an article in this issue,Webb and Ting criticize ROC analysis for its inability to handle certain changes in class distributions. They imply that the ability of ROC graphs to depict performance in the face of changing class distributions has been overstated. In this editorial response, we describe two general types of domains and argue that Webb and Ting’s concerns apply primarily to only one of them. Furthermore, we show that there are interesting real-world domains of the second type, in which ROC analysis may be expected to hold in the face of changing class distributions.},\n bibtype = {article},\n author = {Fawcett, Tom and Flach, Peter A.},\n journal = {Machine Learning},\n number = {1}\n}","author_short":["Fawcett, T.","Flach, P., A."],"urls":{"Paper":"https://bibbase.org/service/mendeley/7edb4dd2-367e-3b20-8e26-a7e1d668fa33/file/4fba5fee-3fe3-6fb2-63a3-c9b194aef9a1/2005-A_Response_to_Webb_and_Tings_On_the_Application_of_ROC_Analysis_to_Predict_Classification_Performance_Under_Varying.pdf.pdf","Website":"http://www.springerlink.com/index/10.1007/s10994-005-5256-4"},"bibbaseid":"fawcett-flach-aresponsetowebbandtingsontheapplicationofrocanalysistopredictclassificationperformanceundervaryingclassdistributions-2005","role":"author","keyword":["class skew","classi cation","classi er evaluation","roc"],"downloads":0},"search_terms":["response","webb","ting","application","roc","analysis","predict","classification","performance","under","varying","class","distributions","fawcett","flach"],"keywords":["class skew","classi cation","classi er evaluation","roc","class skew 1.","classification","classifier evaluation"],"authorIDs":[]}