joeroot
Joined 16 years ago
Karma
63
Submissions
29
Member since
Oct 16
Recent submissions
Comment · 12 years ago
Echo this, they went out of their way to come through for us in an emergency. Fantastic service.
Comment · 12 years ago
Vice is a fairly compelling case. They started with trashy "Do"s and "Don't"s, and now have reporters embedded within ISIS.
Comment · 12 years ago
or Forego! https://github.com/ddollar/forego
Comment · 12 years ago
Enough to get by! I worked on this throughout university, and now at my startup. I was slightly blasé however! The corpus is tiny (173 episodes: http://www.chakoteya.net/ds9/episodes.htm), so a topic model is unlikely to yield anything valuable. There are probably around 10-15 arcs, and simple clustering could be better -- but this is purely hypothetical. In this case, it's simply curiosity.
If you're interested in tools, Mallet (http://mallet.cs.umass.edu/) is a fairly good place to start, and the original LDA paper by Blei, Ng & Jordan (http://machinelearning.wustl.edu/mlpapers/paper_files/BleiNJ...) is a great academic starting point.
Comment · 12 years ago
I see, apologies! I thought you had pulled in all episodes with a score > 7.
I wonder how easily arcs can be identified. I'll try running the transcripts through a topic model this evening.
Comment · 12 years ago
I like the idea, however I'd disagree with the heuristic used. DS9 was really the first Star Trek series to introduce story arcs, and as such, skipping the wrong episodes leads to a confusing experience. I'd argue that in order to avoid missing out on important chunks of the arc, hand curation is really the only viable option (maybe text-clustering of plot narratives might work?).
Personally, as a huge DS9 fan, I think that you should watch every episode. Each episode adds colour and depth to the series' characters, and in my opinion makes it the most rewarding Star Trek series.
Comment · 12 years ago
Streak (http://www.streak.com/) has been a good start for us.
Upsides are that it's flexible enough to match most pipeline progressions, exports to CSV, supports mail merge, and allows you to easily track emails.
Our main issue has been that in person meetings and calls are difficult to log and keep track of, and as we've progressed towards those its become less and less useful.
If most of your interactions happen via email however, Streak is a great flexible (and currently free) tool.
Comment · 13 years ago
Right now no. We use intuition to extract these. We're taking notes though!
Comment · 13 years ago
Definitely. You can sign up without Twitter, it might just take a bit longer to get to the bottom of what interests you!
Comment · 13 years ago
When you signup we spend some time trying to understand you by exploring your Twitter profile (followers, links, retweets etc.). Once done, we either recommend you an article from a curate pool of articles which we've come across before, or by browsing sources we think will have interesting content which you might not be looking at already.
We then send you an email with a link to the content and an explanation behind our reason. If the user replies or reads the link, we keep going. If they don't we stop! Ultimately to do this well, we want to have a conversation with the subscriber (almost like bibliotherapy).
For example, were someones Twitter data points indicative of an interest in Programming Languages and Education, we might send them http://worrydream.com/#!/LearnableProgramming, with a brief explanation on why. If they reply saying they don't like long form, old content, then we'd make sure to adjust the next article accordingly. Hope that helps!
Comment · 13 years ago
Cheers Gigablah!
Comment · 13 years ago
He noticed. 3 years of lurking down the drain.
Comment · 13 years ago
No. We go through a user's followers by hand; identify common influencers and interests; cross reference these against interesting content from a range of sources. We will eventually look for scalable aspects, however right now we're trying to figure out what works best on an individual level.
joeroot.com github.com/joeroot twitter.com/joeroot