Text-Driven Neural Collaborative Filtering Model for Paper Source Tracing

TL;DR

This paper introduces a neural collaborative filtering approach utilizing SciBERT for text feature extraction to identify key references in scholarly articles, demonstrating improved performance in a citation source tracing challenge.

Contribution

The paper presents a novel framework combining NCF and SciBERT for automated citation source tracing, outperforming baselines in a competitive challenge.

Findings

Achieved a MAP score of 0.37814, ranking 11th in the challenge.

Demonstrated the effectiveness of text-driven neural collaborative filtering.

Outperformed baseline models in citation source tracing accuracy.

Abstract

Identifying significant references within the complex interrelations of a citation knowledge graph is challenging, which encompasses connections through citations, authorship, keywords, and other relational attributes. The Paper Source Tracing (PST) task seeks to automate the identification of pivotal references for given scholarly articles utilizing advanced data mining techniques. In the KDD CUP OAG-Challenge PST track, we design a recommendation-based framework tailored for the PST task. This framework employs the Neural Collaborative Filtering (NCF) model to generate final predictions. To process the textual attributes of the papers and extract input features for the model, we utilize SciBERT, a pre-trained language model. According to the experimental results, our method achieved a score of 0.37814 on the Mean Average Precision (MAP) metric, outperforming baseline models and ranking 11th among all participating teams. The source code is publicly available at https://github.com/MyLove-XAB/KDDCupFinal.