On Missing Labels, Long-tails and Propensities in Extreme Multi-label Classification
Erik Schultheis, Marek Wydmuch, Rohit Babbar, Krzysztof Dembczyński
2022Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining20 citationsDOIOpen Access PDF
Abstract
The propensity model introduced by Jain et al has become a standard approach for dealing with missing and long-tail labels in extreme multi-label classification (XMLC). In this paper, we critically revise this approach showing that despite its theoretical soundness, its application in contemporary XMLC works is debatable. We exhaustively discuss the flaws of the propensity-based approach, and present several recipes, some of them related to solutions used in search engines and recommender systems, that we believe constitute promising alternatives to be followed in XMLC.
Topics & Concepts
SoundnessComputer scienceRecommender systemArtificial intelligenceMachine learningData miningProgramming languageMachine Learning and Data ClassificationMachine Learning and AlgorithmsText and Document Classification Technologies