Find research datasets worth reusing

Data Files for Tresoldi/Robinson article on spelling variation in Canterbury Tales manuscripts

<p>This data is at <a href="https://github.com/peterrobinson/CTSpellingArticle2024">https://github.com/peterrobinson/CTSpellingArticle2024</a>. It is contained in three folders, each folder corresponding to one of the three sets of data used in this analysis, as follows:</p> <p><strong>&nbsp;</strong></p> <ol> <li> <p>&ldquo;sorted by regularization&rdquo;. This folder contains spelling data and results derived from the regularization process, where (for example) spellings of forms regularized to &ldquo;goode&rdquo; are distinguished from spellings of forms regularized to &ldquo;god&rdquo;;</p> </li> <li> <p>&ldquo;sorted by part of speech&rdquo;. This folder contains spelling data and results derived from a lemmatization and part-of-speech identification process, where (for example) spellings of forms lemmatized&nbsp; to &ldquo;goode&rdquo; singular adjective are distinguished from spellings of forms regularized to &ldquo;goode&rdquo; plural adjective, and forms lemmatized to &ldquo;gode&rdquo; singular noun nominative case are distinguished from &ldquo;gode&rdquo; singular noun oblique case (as in &ldquo;to gode&rdquo;).</p> </li> <li> <p>&ldquo;all spellings unsorted&rdquo;. This folder contains spelling data as undifferentiated counts of &ldquo;bags of words&rdquo;: for each witness: so many occurrences of &ldquo;good&rdquo;, so many of &ldquo;goode&rdquo;, so many of &ldquo;god&rdquo;, so many of &ldquo;gode&rdquo;.</p> </li> </ol> <p><strong>&nbsp;</strong></p> <p>Each folder contains the following files (under various names):</p> <p><strong>&nbsp;</strong></p> <ol> <li> <p>A . json file holding all the data, structured according to its categorization. The &ldquo;sorted by part of speech&rdquo; folder contains two .json files, one with spellings organized by headword lemma, the other organized by part-of-speech;</p> </li> <li> <p>Two .nex Nexus files containing all the data. In the &ldquo;sorted by regularization&rdquo; and&nbsp; &ldquo;sorted by part of speech&rdquo; folders one Nexus file groups spellings by variant sites within each line, the other Nexus file groups spellings&nbsp; by words within each line. In the &ldquo;all spellings unsorted&rdquo; folder one Nexus file contains all the spellings organized by spelling; the second holds a Nexus distance matrix with distances created according to the Manhattan distance algorithm;</p> </li> <li> <p>A .dst distance matrix file, containing a distance matrix constructed with distacnes calculated by the Manhattan distance algorithm;</p> </li> <li> <p>A &ldquo;features&rdquo; file, containing a spreadsheet ranking each variant site according to its impact on the analysis&nbsp;</p> </li> <li> <p>Multiple .pdf files visualizing the results of our analysis, with the names reflecting the analysis each contains. Files with names including &ldquo;Splits&rdquo; were created using the SplitsTree algorithm and software <a href="https://www.zotero.org/google-docs/?peaEav">(Huson and Bryant 2006; &ldquo;SplitsTree | Universit&auml;t T&uuml;bingen,&rdquo; n.d.)</a></p> </li> </ol> <p><strong>&nbsp;</strong></p> <p>The &ldquo;sorted by regularization&rdquo; folder also contains a single image file, &ldquo;tiagoplot1.jpg&rdquo;, visualizing the results of PCA analysis on the &ldquo;sorted by regularization&rdquo; data.</p> <p>&nbsp;</p>