This week, I practiced using Gale Digital Scholar Lab. I was impressed by the variety of text analysis tools that Gale Digital Scholar Lab offers. I had not heard about topic modelling or document clustering tools. I was curious to see what type of output the sentiment analysis and ngrams tools would create. I had more success using the various tools on a .txt file version of Moby Dick than I did with the documents I found through a search of Gale’s digital archives.
I was most frustrated by Gale Digital Scholar Lab’s inability to work with texts in Yiddish. I enjoy reading and learning about Yiddish and I was excited to see that Gale had a few Yiddish texts. However, the OCR text for the Yiddish documents was largely gibberish in the Cyrillic alphabet. This meant that any text analysis with the Yiddish documents was virtually meaningless. I was disappointed that the OCR text didn’t even use the Hebrew alphabet.
I will be able to help students use Gale Digital Scholar Lab to analyze large texts, particularly in English. The tools are easy to run, even if the setups are sometimes less than intuitive. I am excited to use Named Entity Recognition in my research in English.
