ABSTRACT
The multi-sentence unit of analysis that interests discourse scientists has traditionally made quantitative analyses elusive, imprecise, or even impossible due to a poverty of data sources. However, the growth of big data resources has opened a whole new methodological toolbox for discourse researchers. The growth of big data has also led to methodological developments in networks and complex systems, which are starting to see increased use in discourse analysis. The chapter highlights several broad clusters of approaches that capitalize on these advances in especially fruitful and productive ways. Discourse science has a deep tradition of analyzing naturally occurring linguistic sources to explore higher-order cognitive phenomena. A topic of research that benefits of applications as well as cognitive and computational theories of discourse processing is involved with identification of discourse relations, such as causal and temporal relations between neighboring sentences in a text. More "deep" semantic tools have grown from techniques in data science and cognitive modeling.
