Fixes to README #2
Conversation
|
Oh, Haskell port would be wonderful. I've been wanting on learning Haskell for a while now. |
|
Have you thought about including stemming with this? You already do stop-words (nice!), but stemming would be a cool addition to that. I use this free one: http://tartarus.org/~martin/PorterStemmer/ |
|
I have! but unfortunately didn't have time to implement it. What you are seeing right now is the minimum viable product to get a project running (http://cooln.es). :) You seemed to be knowledge-able in machine learning. I'd love to have more contributors :) |
|
Haha...I've just started a machine-learning project so I don't know much yet. A small team of us had done most of what your project does when a friend pointed out BayesOnRedis. Now we can chuck most of our code! ;) Our corpus is going to be pretty big, so we needed something fast. After experimenting with MySQL and Redis, Redis is orders of magnitude faster (one benchmark took a 5hr training process down to 10 minutes!). That's why we are going with Haskell for now because it's very fast but keeps the high level of abstraction. Is cooln.es your site? Pretty sweet! |
|
Cooln.es is indeed my site, I was in the mood of learning on how to write large-ish scale web scrapers and categorizing content. Writing it in Haskell sounds like a miles of fun. Good luck! |
Redis is now at redis.io and I corrected some grammar in the README. I want to do a Haskell port of this when I have the time. Thanks for this awesome lib!