Constructional Change and Distributional Semantics
Martin Hilpert · 2021
This brings us to the tenth and last lecture of these Ten Lectures on Diachronic Construction Grammar.I wish I could say that this lecture ties it all together and explains all the remaining questions.That is not really what I am going to do here.What I will try to do is to tie up some loose ends with regard to constructional change and the distributional methods that I have been talking about in earlier lectures.I feel that when you hear about distributional semantics and semantic vector spaces for the first few times, it can be very demanding.In this lecture, I want to take things a little more slowly and present some of the issues that I have already talked about in a little more detail.I also want to give you two more examples of analyses that you can do with this method.Without further ado, let me talk about the motion charts that you have seen in earlier lectures.I was introduced to motion charts by Hans Rosling.A few years ago, I watched a video of a talk that he gave, which had the somewhat sensationalist title Hans Rosling Shows the Best Statistics that You've Ever Seen.