Abstract | ||
---|---|---|
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. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1109/IS.2010.5548402 | IEEE Conf. of Intelligent Systems |
Keywords | Field | DocType |
learning (artificial intelligence),program verification,software performance evaluation,application oriented validation and evaluation,application specific trade off,learning algorithm evaluation,supervised learner,classification,evaluation,supervised learning,software engineering,computer science | Data mining,Computer science | Conference |
ISBN | Citations | PageRank |
978-1-4244-5164-7 | 0 | 0.34 |
References | Authors | |
0 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Niklas Lavesson | 1 | 148 | 21.83 |
Paul Davidsson | 2 | 315 | 53.19 |