User-centered text mining (invited tutorial)

AXEL J. SOTO; EVANGELOS MILIOS

Halifax

Conferencia; 28th Canadian Conference on Artificial Intelligence, Canadian AI 2015; 2015

CANVAC

Historically,  text  mining  methods  for  extracting  "knowledge" from text have increased in sophistication by the incorporation of both statistical learning and symbolic natural language processing. However, in scenarios where a domain user aims to make sense of document collections and derive insights from them,  domain  knowledge  is necessary  to inform  the analysis,  and  text  mining needs to be complemented by text visualization and user-driven interaction with the analytic process. In this tutorial we introduce Visual Text Analytics as a multi-disciplinary field of research. We cover conceptual and practical methods and tools, review state-of-the-art  systems  that  integrate  text  mining  with  visualization  and  user  interaction, and identify promising research directions for making text mining more user-centered and accessible to users with an interest in domain-specific applications.