Here in law library-land, we’re all familiar with the concept of “citation chasing”- i.e., finding one good on-point article and then mining its citations, footnotes, and citing sources for other relevant articles. But what if there were a way to let an algorithm look at the relevant article you’d already found and mine it for keywords, ultimately generating you a list of other relevant articles?
Sound like science fiction? It’s not- this is a basic description of JSTOR’s new tool, “Text Analyzer.”
This tool allows you to place large chunks of text (or even the entire text of an article!) into its search box, which will then analyze the text and return other relevant JSTOR articles. If you’ve ever used the “related articles” link on Google Scholar (another great way to citation chase), it’s a similar algorithm. EBSCO also has a similar tool.
However, JSTOR’s Text Analyzer does several things these tools do not. Text Analyzer will also provide a list of what it has identified as relevant keywords in the article, along with importance/prominence. After you’ve run your search, though, you can play with these keywords and elevate their importance, or add or delete keywords in order to re-run your search. Therein lies the true strength of this new tool- not assuming that the search or algorithm gives perfect results right away, but allowing the user to tweak and re-tweak in order to find what they are looking for. It’s still not perfect, but it’s certainly a step in the right direction as far as search engines go, and I’m hoping this pushes other search engines to develop similar and even better tools for searching.