Computational text analysis

Computational text analysis (CTA) comprises techniques for measuring the content of texts with the help of computer algorithms. The methods are discussed under various labels, such as text-as-data, automated content analysis, natural language processing, or text mining. The defining characteristic of a CTA technique is that once it is initially configured, the computational system performs the measurements independently without requiring any manual intervention or effort. The strength of CTA lies in its scalability, enabling the measurement of characteristics across vast amounts of text. As a result, CTA has seen widespread application in communication, related social sciences, and the digital humanities, with the increasing availability of digital or digitized, machine-readable texts.We start this chapter with an overview of the historical development of CTA. We then systematize CTA along two dimensions: the representations of texts for the computational analysis and the supervision of the measurement process. While doing so, we provide some examples of popular techniques. The chapter ends with an outlook into the near future.