Understanding Legal Search Algorithms


Ever wonder why the same search across different legal databases can produce results with different relevancy rankings? It's because of the unique algorithms each database uses. In order for researchers to better understand how the databases cultivate their specific algorithms, representatives from Bloomberg Law, Fastcase, Lexis Advance, and Westlaw contributed to "Inside the Black Box of Search Algorithms" in the November/December 2019 issue of the AALL Spectrum, a publication of the American Association of Law Libraries. The article also summarizes some "common ingredients in how search relevance is calculated", including TF-IDF (term frequency / inverse document frequency), Boolean search operators, proximity factors, and "rank and file".

This article is available to most Jenkins members on the HeinOnline database. HeinOnline is also available in the library.

More from the blog

You may have noticed that General OneFile and LegalTrac look a little different lately. Read on to learn about the new Gale interface.
Generally you should search PA Legal Forms by Title. Sometimes you'll retrieve a forms package containing the list of forms and instructions but not the forms themselves. Review the list, then go back and search by Category to retrieve the ones you want.