Tutorial on Latent Semantic Analysis and Its Potential Use for Vialogues
In today's EdLab Development & Research meeting, Manav talked about Latent Semantic Analysis (LSA), an approach to analyzing discussions by clustering similar discussions. It is not difficult to grasp the concept of LSA but it is difficult to imagine a real example of how LSA is applied to discussion analysis even after reading this paper
and this website
I found an excellent tutorial
on how to do a mini LSA on search results of book titles on Amazon.com.
The most helpful thing that I learned from this tutorial is that we can use a modified version of the Python code example as shown in the tutorial to construct a visualizable graph on how Vialogues discussions are clustered in a semantic space, similar to the graph below:
Such a graph can help the Vialogue coordinator visualize whether the discussions are focused on the topics (as represented by keywords) related to the video or as defined by the coordinator. Based on the graph, the coordinator can easily know, for example:
- What discussants are interested in in the video?
- Which discussion threads are off topic?
- Which portions of the video need coordination if there are many irrelevant discussion threads?
What do you think? Do you agree LSA could be a great add-on to Vialogues? Will people be interested in using this tool to visualize their Vialogues?
Thanks Manav for introducing this fantastic tool!