?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=A+neural+network+cost+function+for+highly+class-imbalanced+data+sets&rft.creator=Twomey%2C+D&rft.creator=Gorse%2C+D&rft.description=We+introduce+a+new+cost+function+for+the+training+of+a+neural+network+classifier+in+conditions+of+high+class+imbalance.+This+function%2C+based+on+an+approximate+confusion+matrix%2C+represents+a+balance+of+sensitivity+and+specificity+and+is+thus+well+suited+to+problems+where+cost+functions+such+as+the+mean+squared+error+and+cross+entropy+are+prone+to+overpredicting+the+majority+class.+The+benefit+of+the+new+measure+is+shown+on+a+set+of+common+class-imbalanced+datasets+using+the+Matthews+Correlation+Coefficient+as+an+independent+scoring+measure.&rft.publisher=i6doc.com&rft.date=2018-01-01&rft.type=Proceedings+paper&rft.language=eng&rft.source=+++++In%3A++ESANN+2018+-+Proceedings+26th+European+Symposium+on+Artificial+Neural+Networks%2C+Computational+Intelligence+and+Machine+LearningESANN+2017+-+Proceedings+25th+European+Symposium+on+Artificial+Neural+Networks%2C+Computational+Intelligence+and+Machine+Learning.++(pp.+pp.+207-212).++i6doc.com%3A+Bruges%2C+Belgium.+(2018)+++++&rft.format=text&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F10115865%2F1%2Fesann_2017_DTDG.pdf&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F10115865%2F&rft.rights=open