• Summary
• Related Documents
  • Active Bibliography
  • Co-citation
• Version History

Abstract:

We explore the use of natural language processing technologies to assist in content and communication analysis, and argue that there is significant synergy between the goals of this social science analysis and the aims and capabilities of computational linguistics research. We discuss specific technologies that can be deployed for use in social science analysis, and describe the key components of a proposed system in which the use of such technologies can result in a significant benefit to the social science researcher interested in analyzing and formalizing the meaning in documents. Social scientists often analyze textual data for indicators of the source, purpose, and consequences of communications. In media and political analysis, for instance, texts are scrutinized for evidence of thematic trends and framing, or the packaging of information with the intent of creating a particular interpretation [1]. The methodology of content analysis has been developed for systematic analysis of the characteristics of messages [1] in support of identification and categorization of texts or text segments relative to the core questions of communication theory: “Who says what, to whom, why, to what extent, and with what effect?”. This methodology includes both qualitative analysis through the coding of document segments in terms of previously established data theories and quantitative analysis of word and code frequencies. It is a methodology that can clearly benefit from automation, and indeed tools known collectively as Computer-Assisted Qualitative Data Analysis

Citations