Active Learning for Age Regression in Social Media

Abstract

Large-scale annotated corpora are a prerequisite for developing high-performance age regression models. However, such annotated corpora are sometimes very expensive and time-consuming to obtain. In this paper, we aim to reduce the annotation effort for age regression via active learning. The key idea of our active learning approach is first to divide the whole feature space into several disjoint feature subspaces and then leverage them to learn a committee of regressors. Given the committee of regressors, we apply a query by committee (QBC) method to select unconfident samples in the unlabeled data for manual annotation. Empirical studies demonstrate the effectiveness of the proposed approach to active learning for age regression.