Threshold optimisation for multi-label classifiers

TitleThreshold optimisation for multi-label classifiers
Publication TypeJournal Article
Year of Publication2013
AuthorsPillai, I, Fumera, G, Roli, F
JournalPattern Recognition
Start Page2055
Date Published07/2013
Keywordsdoc00, doc01, F-measure, multi-label categorization, Precision-Recall, thresholding

Many multi-label classifiers provide a real-valued score for each class. A well known design approach consists of tuning the corresponding decision thresholds by optimising the performance measure of interest. We address two open issues related to the optimisation of the widely used F measure and precision-recall (P-R) curve, with respect to class-related decision thresholds, on a given data set. (i) We derive properties of the micro-averaged F, which allow its global maximum to be found by an optimisation strategy with a low computational cost. So far, only a suboptimal threshold selection rule and a greedy algorithm with no optimality guarantee were  known. (ii) We rigorously define the macro and micro P-R curves, analyse a previously suggested strategy for computing them, based on maximising F, and develop two possible implementations, which can be also exploited for optimising related performance measures. We evaluate our algorithms on five data sets related to three different application domains.

Citation Key 1435
pillai_PR2013_Thresholding.pdf445.97 KB