Teaching Notes - Teaching about search

This post is a continuation from my previous posts - (Part 1, Part 2 and Part 3) on teaching a 100-level undergraduate course called Language and Computers. As mentioned earlier, it is a very diverse class, and I use this textbook: Language and Computers by Marcus Dickinson, Chris Brew and Detmar Meurers. This post is about Chapter 4 of that book, called Searching. The chapter primarily dealt with ideas behind searching through unstructured text.

This chapter discussed two topics - basic working of a search engine, and searching with regular expressions. I think this has been the most detailed chapter in terms of the content.

I also spent about a week giving an overview of the topic, and discussing how a search engine works, and another week discussing regular expressions, and doing practice exercises, and concluding the topic discussion. The exercises at the end of this chapter were very useful to keep the class engaged. Along with some of these exercises, I made use of NACLO problem “Pooh’s Encylopedia” to get students thinking about how search works. I also showed or referred to one or two videos on how google works, why are regular expressions not usable in a search engine like google, the video where Chris Manning explains how are regular expressions used Stanford Tokenizer etc., and demonstrated how regular expressions are used in LibreOffice, etc.

What worked: I think the part about how a search engine works, the videos, and Pooh’s encyclopedia exercise worked out well.

In terms of things that did not work:

  • I think asking such an assorted group of students to work with egrep (and with terminal in general) is a bit too much. I think continuing in the text class made it slightly better though. I could have done this better with more planning and preparation on the topic. Also, I should look for a more graphical tool. Perhaps asking to do this with Word/Office instead of egrep is a better idea for GenZ. I should perhaps not be too negative about myself either - 3 of the 6 teams so far opted to talk about regular expressions for their midterm presentations (from a list of 11 problem descriptions covering 4 broad topics). Clearly, some people seem to have understood it reasonably.
  • I think I had a hard time explaining precision and recall. Despite multiple attempts, and despite even using visuals, I got a feeling the class is either disinterested or did not understand.

So, in the next iteration, I should perhaps work on these two issues.

Written on September 29, 2017