Text Analytics – Phrase and Theme Extraction

This white paper examines theme extraction in the context of text analytics. Theme extraction helps to define the context and content of a conversation providing a highly valuable combination of context scored noun phrases. This paper focuses on the extraction of these nouns and noun phrase themes, specifically those nouns which are not easy to get to via entity extraction and will detail four computational techniques for extracting phrases: clustering, N-grams, noun phrase extraction and themes. Phrase themes provide an excellent view of the context of conversations, and are useful on all lengths of content – from tweets up to hundred-page secondary research reports.