Citation Recommendation based on Argumentative Zoning of User Queries

TL;DR

This paper proposes a multi-task learning model that leverages argumentative zoning to improve citation recommendation accuracy, supported by an annotated PubMed Central corpus.

Contribution

It introduces a novel approach combining argumentative zoning with citation recommendation and provides an annotated dataset for this task.

Findings

Enhanced citation recommendation performance with argumentative zoning information

Created an annotated corpus from PubMed Central for this purpose

Demonstrated the effectiveness of multi-task learning in this context

Abstract

Citation recommendation aims to locate the important papers for scholars to cite. When writing the citing sentences, the authors usually hold different citing intents, which are referred to citation function in citation analysis. Since argumentative zoning is to identify the argumentative and rhetorical structure in scientific literature, we want to use this information to improve the citation recommendation task. In this paper, a multi-task learning model is built for citation recommendation and argumentative zoning classification. We also generated an annotated corpus of the data from PubMed Central based on a new argumentative zoning schema. The experimental results show that, by considering the argumentative information in the citing sentence, citation recommendation model will get better performance.