ASIS&T 2014 Annual Meeting 
Seattle, WA | October 31 - November 5, 2014

 
Citation Role Labeling via Local, Pairwise, and Global Features

Chun Guo1, Yingying Yu2, Azadeh Sanjari1, Xiaozhong Liu1
1
Indiana University Bloomington, School of Informatics and Computing, United States of America; 2College of Transportation Management, Dalian Maritime University, Dalian, China

Tuesday, Nov. 4, 4:00pm


Summary

Citation relationship between scientific publications has been successfully used for bibliometrics, information retrieval and data mining tasks, and citation-based recommendation algorithms are well documented. While previous studies investigated citation relations from various viewpoints, most of them share the same assumption that, if paper1 cites paper2 (or author1 cites author2), they are connected, regardless of citation importance, sentiment, reason, topic, or motivation. However, this assumption is oversimplified. In this study, we propose a novel method to automatically label the massive citations in the scientific repository, a.k.a. citation role labeling, by employing citing and cited paper context. Unlike earlier studies, we employed both local features, i.e., citation textual context, pairwise features, i.e., citing and cited paper similarities, and global features, i.e., citing and cited paper proximity on the heterogeneous graph. Evaluation result shows pairwise and global features, if properly used, can be very helpful to enhance the citation role labeling performance, especially while full-text data is not readily available.