In order to make this fair, I decided to create a dummy data set for two conferences. I did not want to use real data as the results may not be pleasing for some people.
So, stage one was to simply evaluate
- the number of speakers at each conference
- the number of cases of the same person speaking at both.
- the number of cases of the same person speaking more than once at a conference
I have been considering a time series approach - to look at each speaker and see if they are going to talk about the same topics as the past or if they have fresh material.
All just a bit of fun. At a later date I will put all of this on github along with some tools to get data from real conference websites
No comments:
Post a Comment