Does it Matter What They Said? A Text Mining Analysis of the State of the Union Addresses of USA Presidents

Abstract

Text data are a major type of data in the modern society. Literatures have pointed out that more than 80% of data are in text form. It is important to study the insights from text data in addition to quantitative data. The development of text mining techniques started in the early 80's. The methodology has become much more mature in the recent years. In this article, we conduct a case study using text mining technique to analyze the patterns of the president's State of the Union address in USA. The speeches analyzed include the recent four USA presidents, Bush (1989-1992), Clinton (1993 - 2000), G.W. Bush (2001-2008), and Obama (2009-2011). Using two different text mining techniques, we identify six clusters from the 23 speeches using one technique and obtain seven topics based on the other technique.