Warning: Undefined variable $num in /home/shroutdo/public_html/courses/wp-content/plugins/single-categories/single_categories.php on line 126

Warning: Undefined variable $posts_num in /home/shroutdo/public_html/courses/wp-content/plugins/single-categories/single_categories.php on line 127

A lot goes on behind geo-tagging historical texts. A piece by Tilton, Arnold, Rivard titled “Locating Place Names at Scale: Using Natural Language Processing to Identify Geographical Information in Text” explains the process fairly thoroughly and in simple terms. From identifying the location fragment in the source, to trying to find other references, to finding general geographic closeness, the steps the algorithms takes is laid out in an easy to comprehend way. Though very concise, the writing gives us a good understanding of how the goal is accomplished.

There are some lingering questions, however. What if the location found is not what it seems to be? Does the algorithm discard the term if there is too much ambiguity? Depending on the use of the results, this could have some implications, where incorrect locations are tagged and conclusions may be skewed.

However, this article for the most part is very informative, and as SA says “an eye-opener” to consider what goes on behind the scenes of this sort of interpretation of data.