Citation recommendation based on argumentative zoning of user queries

2025-01-30

J. Informetrics

AI Summary

This paper addresses the mismatch between citation recommendations and users’ argumentative intents in academic search. To this end, it proposes the first citation recommendation method integrating argument mining. Its core innovation lies in adopting the argumentative zoning paradigm to identify fine-grained argumentative roles—such as claim, evidence, and counterargument—within user queries, thereby enabling semantic intent–driven, precise matching. Methodologically, the approach employs a BERT-based argument role tagging model, leverages graph neural networks to capture structured relationships among documents and argumentative roles, and introduces a multi-objective learning-to-rank framework that jointly optimizes relevance and argumentative consistency. Evaluated on the ACL Anthology and PubMed QA benchmarks, the method achieves a 12.7% improvement in Recall@5 and a 23.4% gain in argumentative consistency score, significantly outperforming current state-of-the-art methods.

Technology Category

• Natural Language Processing: Discourse, Pragmatics & Argument Mining
• Knowledge Representation and Reasoning: Argumentation
• Data Mining & Knowledge Management: Intelligent Query Processing

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Community question answering

Problem

• Citation Recommendation
• Argumentation Analysis
• Relevance Improvement

Innovation

• Multi-task Learning Model
• Argument Mining
• Citation Recommendation