Popularity Prediction for Social Media over Arbitrary Time Horizons
Daniel Haimovich, Dima Karamshuk, Thomas Leeper, Evgeniy Riabenko, Milan Vojnovic
ACM Conference on Knowledge Discovery and Data Mining (KDD)
In entity matching, a fundamental issue while training a classifier to label pairs of entities as either duplicates or non-duplicates is the one of selecting informative examples. Although active learning presents an attractive solution to this problem, previous approaches minimize the misclassification rate (0-1 loss) of the classifier, which is an unsuitable metric for entity matching due to class imbalance (i.e., many more non-duplicate pairs than duplicate pairs).
To address this, a recent work proposes to maximize recall of the classifier under the constraint that its precision should be greater than a specified threshold. However, the proposed technique requires the labels of all n input pairs in the worst-case. Our main result is an active learning algorithm that approximately maximizes recall of the classifier under precision constraint with provably sub-linear label complexity (under certain distributional assumptions). Our algorithm uses as a black-box any active learning approach that minimizes 0-1 loss.
We show that label complexity of our algorithm is at most log(n) times the label complexity of the black-box, and also bound the difference in the recall of classifier learnt by our algorithm and the recall of the optimal classifier satisfying the precision constraint. We provide an empirical evaluation of our algorithm on several real-world matching data sets that demonstrates the effectiveness of our approach.
Daniel Haimovich, Dima Karamshuk, Thomas Leeper, Evgeniy Riabenko, Milan Vojnovic
Liqi Yan, Qifan Wang, Yiming Cu, Fuli Feng, Xiaojun Quan, Xiangyu Zhang, Dongfang Liu
Barlas Oğuz, Kushal Lakhotia, Anchit Gupta, Patrick Lewis, Vladimir Karpukhin, Aleksandra Piktus, Xilun Chen, Sebastian Riedel, Wen-tau Yih, Sonal Gupta, Yashar Mehdad