Bag of Tricks for Citation Intent Classification via SciBERT

Motrichenko, Dmitry; Nedumov, Yaroslav; Skorniakov, Kirill

doi:10.1109/ispras53967.2021.00022

Cited by 2 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we align with the common practice of using the terms "citation purpose", "citation function" and "citation intent" interchangeably, as we observed this trend among authors. Citation context analysis involves various tasks, including citation function analysis [1,13], citation provenance analysis [13], citation intent identification [14][15][16], and citation sentiment classification [5,[17][18][19][20][21][22][23]. However, our focus in this research is primarily on two pivotal citation analysis tasks: citation intent classification and citation sentiment (polarity) analysis.…”

Section: Literature Surveymentioning

confidence: 99%

“…The author employed SciBERT to experiment with two citation intent classification datasets viz. ACL-ARC ‡ (6 classes, 1941 instances) and SciCite § (3 classes, 11020 instances) and obtained a macro F1 of 72% and 85.5% for ACL-ARC and SciBERT respectively [14]. They considered hand generated features suggested by [7] of a vector of length of 56 and generated a vector of same size for each paragraph for SciCite dataset.…”

Section: Literature Surveymentioning

confidence: 99%

See 1 more Smart Citation

Multi-sentence and multi-intent classification using RoBERTa and graph convolutional neural network

Ravi,

Singh,

Gautam

2024

Preprint

View full text Add to dashboard Cite

Citation analysis has garnered significant attention in academia, particularly in the realm of scientometrics analysis. Most studies related to citation analysis focus on quantitative aspects, assigning equal weight to every citation regardless of its placement within the paper. However, understanding the distribution of citation weight across different sections of a research article is crucial for citation analysis and impact assessment. Therefore, the analysis of citation intent becomes a pivotal task in determining the qualitative importance of a citation within a scientific article. In this context, we undertook two essential tasks related to citation analysis: citation length analysis and citation intent analysis. Through citation length analysis, we identified the optimal number of citation sentences to consider around a cited sentence. Simultaneously, citation intent analysis aimed to categorize citations into seven distinct types, namely background, motivation, uses, extends, similarities, differences, and future work. For the latter task, we introduced two novel architectures based on graph neural networks, namely CiteIntentRoBERTaGCN and CiteIntentRoBERTaGAT. The performance of these proposed models was evaluated on five multi-intent datasets curated from 1,200 research papers, considering different context lengths. The results demonstrated that the proposed models achieved state-of-the-art performance.

show abstract