A dataset for temporal analysis of files related to the JFK case

<p>This dataset contains the content of the subset of all files with a correct publication date from the 2017 release of files related to the JFK case (retrieved from https://www.archives.gov/research/jfk/2017-release). This content was extracted from the source PDF files using the R OCR libraries tesseract and pdftools.</p> <p>The code to derive the dataset is given as follows:</p> <p>### BEGIN R DATA PROCESSING SCRIPT</p> <p>library(tesseract)<br> library(pdftools)</p> <p>pdfs &lt;- list.files(&quot;/home/STAFF/luczakma/RProjects/JFK/data/files/&quot;)</p> <p>meta &lt;- read.csv2(&quot;/home/STAFF/luczakma/RProjects/JFK/data/jfkrelease-2017-dce65d0ec70a54d5744de17d280f3ad2.csv&quot;,header = T,sep = &#39;,&#39;)</p> <p>meta$Doc.Date &lt;- as.character(meta$Doc.Date)</p> <p>meta.clean &lt;- meta[-which(meta$Doc.Date==&quot;&quot; | grepl(&quot;/0000&quot;,meta$Doc.Date)),]<br> for(i in 1:nrow(meta.clean)){<br> &nbsp; meta.clean$Doc.Date[i] &lt;- gsub(&quot;00&quot;,&quot;01&quot;,meta.clean$Doc.Date[i])<br> &nbsp;<br> &nbsp; if(nchar(meta.clean$Doc.Date[i])&lt;10){<br> &nbsp;&nbsp;&nbsp; meta.clean$Doc.Date[i]&lt;-format(strptime(meta.clean$Doc.Date[i],format = &quot;%d/%m/%y&quot;),&quot;%m/%d/%Y&quot;)<br> &nbsp; }<br> &nbsp;<br> }</p> <p>meta.clean$Doc.Date &lt;- strptime(meta.clean$Doc.Date,format = &quot;%m/%d/%Y&quot;)</p> <p>meta.clean &lt;- meta.clean[order(meta.clean$Doc.Date),]</p> <p>docs &lt;- data.frame(content=character(0),dpub=character(0),stringsAsFactors = F)<br> for(i in 1:nrow(meta.clean)){<br> #for(i in 1:3){<br> &nbsp; pdf_prop &lt;- pdftools::pdf_info(paste0(&quot;/home/STAFF/luczakma/RProjects/JFK/data/files/&quot;,tolower(gsub(&quot;\\s+&quot;,&quot; &quot;,gsub(&quot; &quot;,&quot;&quot;,meta.clean$File.Name[i])))))<br> &nbsp; tmp_files &lt;- c()<br> &nbsp; for(k in 1:pdf_prop$pages){<br> &nbsp;&nbsp;&nbsp; tmp_files &lt;- c(tmp_files,paste0(&quot;/home/STAFF/luczakma/RProjects/JFK/data/tmp/&quot;,k))<br> &nbsp; }<br> &nbsp;<br> &nbsp; img_file &lt;- pdftools::pdf_convert(paste0(&quot;/home/STAFF/luczakma/RProjects/JFK/data/files/&quot;,tolower(gsub(&quot;\\s+&quot;,&quot; &quot;,gsub(&quot; &quot;,&quot;&quot;,meta.clean$File.Name[i])))), format = &#39;tiff&#39;, pages = NULL, dpi = 700,filenames = tmp_files)<br> &nbsp;<br> &nbsp; txt &lt;- &quot;&quot;<br> &nbsp;<br> &nbsp; for(j in 1:length(img_file)){<br> &nbsp;&nbsp;&nbsp; extract &lt;- ocr(img_file[j], engine = tesseract(&quot;eng&quot;))<br> &nbsp;&nbsp;&nbsp; #unlink(img_file)<br> &nbsp;&nbsp;&nbsp; txt &lt;- paste(txt,extract,collapse = &quot; &quot;)<br> &nbsp; }<br> &nbsp;<br> &nbsp; docs &lt;- rbind(docs,data.frame(content=iconv(tolower(gsub(&quot;\\s+&quot;,&quot; &quot;,gsub(&quot;[[:punct:]]|[\n]&quot;,&quot; &quot;,txt))),to=&quot;UTF-8&quot;),dpub=format(meta.clean$Doc.Date[i],&quot;%Y/%m/%d&quot;),stringsAsFactors = F),stringsAsFactors = F)<br> }</p> <p>### END R DATA PROCESSING SCRIPT</p>