APPrOVE: Application-oriented Validation and Evaluation of Supervised Learners
Blekinge Institute of Technology, School of Computing2010 (English)Conference paper (Refereed) Published
Learning algorithm evaluation is usually focused on classification performance. However, the characteristics and requirements of real-world applications vary greatly. Thus, for a particular application, some evaluation criteria are more important than others. In fact, multiple criteria need to be considered to capture application-specific trade-offs. Many multi-criteria methods can be used for the actual evaluation but the problems of selecting appropriate criteria and metrics as well as capturing the trade-offs still persist. This paper presents a framework for application-oriented validation and evaluation (APPrOVE). The framework includes four sequential steps that together address the aforementioned problems and its use in practice is demonstrated through a case study.
Place, publisher, year, edition, pages
London: IEEE press , 2010.
classification, evaluation, supervised learning
Software Engineering Computer Science
IdentifiersURN: urn:nbn:se:bth-7758Local ID: oai:bth.se:forskinfoBF98093116D46606C12577640038B637ISBN: 978-1-4244-5164-7OAI: oai:DiVA.org:bth-7758DiVA: diva2:835419
IEEE Intelligent Systems