Massimo Poesio's suggestions
(1) Bayesian methods:
- start perhaps with Knight's tutorial "Bayes with tears" at
- then move on to linear regression etc with Gelman and Hill (I've got the book):
- (then if we're really keen we can do David Barber's book "Bayesian Reasoning and Machine Learning". A draft of that is
available at http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=Main.Textbook
- else if we find we lack stats background we can do Harald Baayen's book
(2) Linear algebra
this is fundamental to all work with matrices, vector spaces, etc
I need to find a good book
David Hunter's Suggestions
There is a growing literature – in both computer science and in social psychology – on the analysis of blogs and twitter feeds. Much of this analysis is very simplistic and just amounts to counting word frequencies in order to determine, for example, gross national happiness. I’d be happy to prepare some slides in order to highlight some of the more interesting recent publications in this area, so as to provoke some discussion amongst LAC members. I think this is a topic that would be of interest to many group members, and it’d be worthwhile to see what comments people have.
Also, I definitely approve of Massimo’s suggestions, particularly Bayesian networks and linear algebra.
Sounds all good to me. How about doing Bayesian methods and as we have done in previous years we will then have different topics every now and then in between (e.g. David's suggestion, reviewing of research papers, presentation of conference talks, etc)