Boolean Search: Will George Boole Rotate in His Grave?

January 12, 2016

Boolean logic is, for most math wonks, the father of Boolean logic. This is a nifty way to talk about sets and what they contain. One can perform algebra and differential equations whilst pondering George and his method for thinking about fruits when he went shopping.

In the good old days of search, there was one way to search. One used AND, OR, NOT, and maybe a handful of other logic operators to retrieve information from structured indexes and content. Most folks with a library science degree or a friendly math major can explain Boolean reasonably well. Here’s an example which might even work on CSA ProQuest (nèe Lockheed Dialog) even today:

CC=77? AND scam?

The systems when fed the right query would reply with pretty good precision and recall. Precision provided info that was supposed to be useful. Recall meant that what should be included was in the result set.

I thought about Boole, fruit, and logic when I read “The Best Boolean and Semantic Search Tool.” Was I going to read about SDC’s ORBIT, ESA Quest, or (heaven help me) the original Lexis system?

Nope.

I learned about LinkedIn. Not one word about Palantir’s injecting Boolean logic squarely in the middle of its advanced data management processes. Nope.

LinkedIn. I thought that LinkedIn used open source Lucene, but maybe the company has invested in Exorbyte, Funnelback, or some other information access system.

The write up stated:

If you use any source of human capital data to find and recruit people (e.g., your ATS/CRM, resume databases, LinkedIn, Google, Facebook, Github, etc.) and you really want to understand how to best approach your talent sourcing efforts, I recommend watching this video when you have the time.

Okay, human resource functions. LinkedIn, right.

But there is zero content in the write up. I was pointed to a video called “Become a LinkedIn Search Ninja: Advanced Boolean Search” on YouTube.

Here’s what I learned before I killed the one hour video:

  1. The speaker is in charge of personnel and responsible for Big Data activities related to human resources
  2. Search is important to LinkedIn users
  3. Profiles of people are important
  4. Use OR. (I found this suggestion amazing.)
  5. Use iterative, probabilistic, and natural language search, among others. (Yep, that will make sense to personnel professionals.)

Okay. I hit the stop button. Not only will George be rotating, I may have nightmares.

Please, let librarians explicitly trained in online search and retrieval explain methods for obtaining on point results. Failing a friendly librarian, ask someone who has designed a next generation system which provides “helpers” to allow the user to search and get useful outputs.

Entity queries are important. LinkedIn can provide some useful information. The tools to obtain that high value information are a bit more sophisticated than the recommendations in this video.

Stephen E Arnold, January 12, 2016

Comments

Comments are closed.

  • Archives

  • Recent Posts

  • Meta