Extreme Multi-label Text Classification (XMTC) is a key task of finding the most relevant labels from a large label set for a document. Although some deep learning-based methods have shown great success in XMTC, they still suffer from the following drawbacks. First, although several methods have improved the precision by clustering labels and combining several sub-models to train and predict for one dataset, they were not ideal in terms of computational efficiency. Second, most of those methods need a low dimensional bottleneck layer before the output layer to compress the feature representations to fit the GPU memory, which results in information loss of original features. In this paper, we proposed a novel two-stage XMTC framework with candidate Retrieving and deep Ranking (XRR) to address those drawbacks. In the retrieving stage, we designed two retrieval strategies, including an aligning Point Mutual Information (aPMI) method, and a Unified Label-Semantic Embedding (ULSE) method, to extract hundreds of candidates from massive labels. In the ranking stage, we presented a deep ranking model using a pre-trained transformer to distinguish the true labels from candidates. Extensive experiments show that XRR outperforms the state-of-the-art methods on five widely used multi-label datasets.

Jie Xiong

Li Yu

Xi Niu

Youfang Leng

Information Sciences

信息科学将发表原创、创新和创造性的研究成果。将不时出版少量及时的教程和调查贡献。该杂志旨在为研究人员、开发人员、管理人员、战略规划人员、研究生和其他对信息、知识方面的最新研究活动感兴趣的人提供服务工程和智能系统。假定读者对信息科学有共同的兴趣，但在工程、数学、统计学、物理学、计算机科学、细胞生物学、分子生物学、管理科学、认知科学、神经生物学、行为科学和生物化学等领域具有不同的背景。

The journal "Information Sciences" is dedicated to publishing original, innovative, and creative research findings. We occasionally publish timely tutorials and survey contributions. This journal aims to serve researchers, developers, managers, strategic planners, graduate students, and others interested in the latest research activities in the fields of information and knowledge, particularly in service engineering and intelligent systems. We assume that readers share a common interest in information science, but come from diverse backgrounds in engineering, mathematics, statistics, physics, computer science, cell biology, molecular biology, management science, cognitive science, neurobiology, behavioral science, and biochemistry.

《Information Sciences》期刊致力于发表原创、创新和创造性的研究成果。我们不定期出版一些及时的教程和调查贡献。该期刊旨在为研究人员、开发人员、管理人员、战略规划人员、研究生及其他对信息和知识领域最新研究活动感兴趣的人士提供服务，尤其是工程和智能系统方面的研究。我们假定读者对信息科学有共同的兴趣，但在工程、数学、统计学、物理学、计算机科学、细胞生物学、分子生物学、管理科学、认知科学、神经生物学、行为科学和生物化学等领域具有不同的背景。

Category	Quartile
计算机科学	1
计算机科学, 计算机信息系统	1

XRR: Extreme multi-label text classification with candidate retrieving and deep ranking

ivySCI AI Smartly Parses PDF, Answers Researchers' Questions, and Helps You Understand Papers in Seconds

Journal Info

Category	Quartile
COMPUTER SCIENCE, INFORMATION SYSTEMS	1