B(eo)W(u)LF: Facilitating recurrence analysis on multi-level language

Discourse analysis may seek to characterize not only the overall composition of a given text or corpora but also the dynamic patterns within the data. Patterns of interest may occur at multiple levels, from character to sentence to corpus. Researchers may be interested in the way that sentence structures recur between participants or how affect words cluster in a single text. Recurrence analyses are an ideal tool for such investigations, but linguistic data must often be transformed prior to being analyzed. This technical report introduces a data format called the by-­‐word long-­‐form or B(eo)W(u)LF. Inspired by the long-­‐form data format required for mixed-­‐effects modeling, B(eo)W(u)LF structures linguistic data into an expanded matrix encoding any number of researchers-­‐specified markers. While we do not necessarily claim to be the first to use methods along these lines, we have created a series of tools utilizing Python and MATLAB to enable such discourse analyses. We demonstrate this analysis on 319 lines of the Old English epic poem, Beowulf, translated into modern English (Appendix 1). At the end of this report, we provide the original text, scripts adapted to the text, and (for brevity's sake) a portion of the final result. The sample text is a single file, but if a corpus is saved in multiple individual files, these scripts can be modified with " if " statements to streamline the process. Text Preparation These scripts require that text data be stored in a plain text file (.txt or .csv). If the corpus comprises separate text files, it's highly recommended that each file be transcribed or formatted identically. The data will be run through with a series of regular expressions in Python to quickly and automatically reformat the text (Appendix 2). Therefore, if individual files are formatted differently, the cleanup script will have to be tweaked for each format type. The analyses to be performed will dictate the cleanup choices. In our example, we remove commas, colons, and semicolons, but we are interested in keeping the end-­‐ of-­‐sentence punctuations (e.g., periods, question marks). We may choose to do