Purpose and Polarity of Citation: Towards NLP-based Bibliometrics

Amjad Abu-Jbara, Jefferson Ezra and Dragomir Radev

Bibliometric measures are commonly used to estimate the popularity and the impact of published research. Existing bibliometric measures provide "quantitative" indicators of how good a published paper is. This does not necessarily reflect the "quality" of the work presented in the paper. For example, when h-index is computed for a researcher, all incoming citations are treated equally, ignoring the fact that some of these citations might be negative. In this paper, we propose using NLP to add a ``qualitative'' aspect to biblometrics. We analyze the text that accompanies citations in scientific articles (which we term citation context). We propose supervised methods for identifying citation text and analyzing it to determine the purpose (i.e. author intention) and the polarity (i.e. author sentiment) of citation.