Gandalf: Learning Label-label Correlations in Extreme Multi-label Classification via Label Features
Loading...
Access rights
openAccess
Journal Title
Journal ISSN
Volume Title
A4 Artikkeli konferenssijulkaisussa
This publication is imported from Aalto University research portal.
View publication in the Research portal
View/Open full text file from the Research portal
Other link related to publication
View publication in the Research portal
View/Open full text file from the Research portal
Other link related to publication
Date
2024-08-25
Department
Major/Subject
Mcode
Degree programme
Language
en
Pages
12
Series
Abstract
Extreme Multi-label Text Classification (XMC) involves learning a classifier that can assign an input with a subset of most relevant labels from millions of label choices. Recent works in this domain have increasingly focused on a symmetric problem setting where both input instances and label features are short-text in nature. Short-text XMC with label features has found numerous applications in areas such as query-to-ad-phrase matching in search ads, title-based product recommendation, prediction of related searches. In this paper, we propose Gandalf, a novel approach which makes use of a label co-occurrence graph to leverage label features as additional data points to supplement the training distribution. By exploiting the characteristics of the short-text XMC problem, it leverages the label features to construct valid training instances, and uses the label graph for generating the corresponding soft-label targets, hence effectively capturing the label-label correlations. Surprisingly, models trained on these new training instances, although being less than half of the original dataset, can outperform models trained on the original dataset, particularly on the PSP@k metric for tail labels. With this insight, we aim to train existing XMC algorithms on both, the original and new training instances, leading to an average 5% relative improvements for 6 state-of-the-art algorithms across 4 benchmark datasets consisting of up to 1.3M labels. Gandalf can be applied in a plug-and-play manner to various methods and thus forwards the state-of-the-art in the domain, without incurring any additional computational overheads. Code has been open-sourced at www.github.com/xmc-aalto/InceptionXML.Description
Publisher Copyright: © 2024 Copyright held by the owner/author(s).
Keywords
co-occurrence matrix, correlation graph, data augmentation, extreme classifiers, label-label correlations, multi-label classification
Other note
Citation
Kharbanda, S, Gupta, D, Schultheis, E, Banerjee, A, Hsieh, C J & Babbar, R 2024, Gandalf: Learning Label-label Correlations in Extreme Multi-label Classification via Label Features . in KDD 2024 - Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . ACM, pp. 1360-1371, ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Barcelona, Spain, 25/08/2024 . https://doi.org/10.1145/3637528.3672063