Online F-Measure Optimization

The F-measure is an important and commonly used performance metric for bi-nary prediction tasks. By combining precision and recall into a single score, itavoids disadvantages of simple metrics like the error rate, especially in cases ofimbalanced class distributions. The problem of optimizing the F-measure, thatis, of developing learning algorithms that perform optimally in the sense of thismeasure, has recently been tackled by several authors. In this paper, we studythe problem of F-measure maximization in the setting of online learning. Wepropose an efficient online algorithm and provide a formal analysis of its conver-gence properties. Moreover, first experimental results are presented, showing thatour method performs well in practice.