[ml] intro machine learning meetup: Thursday 12/19 @ 7:00pm

[Sam Tepper]

Sadly, I will be unable to attend, as I'm away for the holidays. 
Hopefully someone will be able to post notes to fill in myself and
others like me.

I wanted to propose a mini-project to investigate, improve, and create. 
I recently had my attention drawn to the awesome power that is noisebot
<https://noisebridge.net/wiki/Noisebot>.  Apprently it's a stock rbot
<http://ruby-rbot.org/rbot-trac/wiki> implimentation that noisebridge
has on its irc channel, which uses hidden markov models and a whole
bunch of other interesting modules.

Unfortunately, it's written in Ruby, and I'm not sure how easy it would
be to access and change directly.  So I'd recommend people to check it
out, see what it can do and what it can do better, and think about
whether you might be interested in working on something like it, either
by improving the code (Ruby) or creating a new and improved Python/Sage
Noisebot 2.0

Anyway, that's just an idea, but I thought it might be nice to check out
what Noisebridge machine learning implementations we are already using,
especially one as multi-faceted and public-facing as this.

Cheers,
Sam

On 12/12/2013 06:50 PM, Mike Schachter wrote:
> Next Thursday, in the Church classroom, I will be teaching an
> introductory class in machine learning. Our goal will be
> straightforward: given a dataset, what are the first things to be
> done? We will go over:
>
> 1) How to read and represent data in python
> 2) How to create and plot features from the data
> 3) How to use histograms to look at the distribution of features
> 4) What is regression? (If there is time)
>
> No software prerequisites will be necessary, because we will be using
> SAGE (https://cloud.sagemath.com/). All you need is a web browser,
> your brain, and it's support system.
>
> I am synthesizing a dataset from two sources:
>
> The SIPRI database of international arms transfers:
> http://portal.sipri.org/publications/pages/transfer/splash
>
> The PRIO database of international armed conflicts:
> http://www.prio.no/Data/Armed-Conflict/
>
> With such a rich dataset, we can start to examine fun things like the
> relationship between international arms transfers and worldwide armed
> conflicts.
>
>  mike
>
>
>
> _______________________________________________
> Noisebridge-discuss mailing list
> Noisebridge-discuss at lists.noisebridge.net
> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.noisebridge.net/pipermail/ml/attachments/20131218/6234c504/attachment-0003.html>