Hierarchical Multi-Label Classification of Scientific Documents

TL;DR

This paper introduces SciHTC, a large hierarchical multi-label dataset for scientific paper classification, and proposes a multi-task learning model that improves classification performance, opening new research avenues.

Contribution

The paper presents a new large-scale dataset for hierarchical multi-label classification of scientific papers and a multi-task learning approach that enhances classification accuracy.

Findings

Achieved a Macro-F1 score of 34.57% with the proposed model.

Provided a new dataset with 186,160 papers and 1,233 categories.

Established strong baselines for hierarchical multi-label classification.

Abstract

Automatic topic classification has been studied extensively to assist managing and indexing scientific documents in a digital collection. With the large number of topics being available in recent years, it has become necessary to arrange them in a hierarchy. Therefore, the automatic classification systems need to be able to classify the documents hierarchically. In addition, each paper is often assigned to more than one relevant topic. For example, a paper can be assigned to several topics in a hierarchy tree. In this paper, we introduce a new dataset for hierarchical multi-label text classification (HMLTC) of scientific papers called SciHTC, which contains 186,160 papers and 1,233 categories from the ACM CCS tree. We establish strong baselines for HMLTC and propose a multi-task learning approach for topic classification with keyword labeling as an auxiliary task. Our best model achieves a Macro-F1 score of 34.57% which shows that this dataset provides significant research opportunities on hierarchical scientific topic classification. We make our dataset and code available on Github.