The Importance of Fine-Grained Cue Phrases in Scientific Citations

Scientific citations play a crucial role in maintaining the network of relationships among mutually relevant articles within a research field. Customarily, authors include citations in their papers to indicate works that are foundational in their field, background for their own work, or representative of complementary or contradictory research. But, determining the nature of the exact relationship between a citing and cited paper is often difficult to ascertain. To address this problem, the aim of formal citation analysis has been to categorize and, ultimately, automatically classify scientific citations. In previous work, Garzone and Mercer (2000) presented a system for citation classification that relied on characteristic syntactic structure to determine citation category. In this present work, we extend this idea to propose that fine-grained cue phrases within citation sentences may provide just such a stylistic basis for categorization. 1 The Citation Problem: Automating Classification 1.1 The Purpose of Citations Scientific citations play a crucial role in maintaining the network of relationships among articles within a research field by linking together works whose methods and results are in some way mutally relevant. Customarily, authors include citations in their papers to indicate works that are foundational in their field, background for their own work, or representative of complementary or contradictory research. A researcher may then use the presence of citations to locate articles she needs to know about when entering a new field or to read in order to keep track of progress in a field where she is already wellestablished. But, determining the nature of the exact relationship between a citing and cited paper, whether a particular article is relevant and, if so, in what way, is often difficult to ascertain. To address this problem, the aim of citation analysis studies has been to categorize and, ultimately, automatically classify scientific citations. An automated citation classifier could be used, for example, in scientific indexing systems to provide additional information to help users navigating a digital library of scientific articles. In previous work, Garzone and Mercer [10] presented a system for citation classification that relied on a citation sentence’s characteristic syntactic structure to assist in determining citation category. In this present work, we extend this idea to propose that fine-grained cue phrases within citation sentences may provide just such a stylistic basis for categorization.