Twss.js • Hacker News Board

Re: Twss.js

#21

Post by driverdan » Thu, Jan 12, 2012, 4:29 PM UTC

An interesting (and funny) exercise.

For those interested in neural networks and Bayesian classifiers check out the brain.js library: http://harthur.github.com/brain/

It works in both node and the browser.

Re: Twss.js

#22

Post by sc00ter » Thu, Jan 12, 2012, 4:37 PM UTC

tibbon wrote: post #5

Now I'm just waiting for someone to make a Twitter bot that randomly samples tweets and responds to them with this...

Already done (and undone) it would seem:

https://github.com/DanielRapp/twss.js/commit/d025aa46eef8559...

Re: Twss.js

#23

Post by hawtshot » Thu, Jan 12, 2012, 4:39 PM UTC

dangrossman wrote: post #20

phreeza wrote: post #6

Has anyone thrown this on a web server with a simple interface?

Sure, why not -- http://pulse.w3counter.com:8080/

It seems to be returning a lot of false positives, at least with the default options. "Good morning" = true, "How are you?" = true

Re: Twss.js

#24

Post by Rapp » Thu, Jan 12, 2012, 4:42 PM UTC

burgerbrain wrote: post #16

mertd wrote: post #12

The training data is pretty funny. I suppose he collected it from an online TWSS thread.

@Rapp You seem to have been hellbanned 132 days ago.

@burgerbrain

Thank you for letting me know! Is there anything I can do about it?

Re: Twss.js

#25

Post by mckoss » Thu, Jan 12, 2012, 4:44 PM UTC

MRI's have shown that humans are able to do this because of a dedicated site in the brain called "Scott's region". Once activated, this linguistic region is constantly searching for linguistic cues, surfacing signals to our conscious thoughts when the cues are strong enough.

Re: Twss.js

#26

Post by dangrossman » Thu, Jan 12, 2012, 4:58 PM UTC

hawtshot wrote: post #23

dangrossman wrote: post #20

Earlier quoted context omitted.

Sure, why not -- http://pulse.w3counter.com:8080/

It seems to be returning a lot of false positives, at least with the default options. "Good morning" = true, "How are you?" = true

I tried adding dropdowns to change the algorithm and threshold, but changing to knn crashes out ("ReferenceError: trainingPrompt is not defined"), so scrapped that and just left the demo running the defaults.

Re: Twss.js

#27

Post by svdad » Thu, Jan 12, 2012, 5:14 PM UTC

hawtshot wrote: post #23

dangrossman wrote: post #20

Earlier quoted context omitted.

Sure, why not -- http://pulse.w3counter.com:8080/

It seems to be returning a lot of false positives, at least with the default options. "Good morning" = true, "How are you?" = true

"Capitalism without failure isn't capitalism."

That's what she said.

Re: Twss.js

#28

Post by donohoe » Thu, Jan 12, 2012, 5:18 PM UTC

I'm still looking for a classifier that will take a phrase, determine if and what the "In Soviet Russia X Y you" response would be.

Anyone?

Re: Twss.js

#29

Post by Zelphyr » Thu, Jan 12, 2012, 5:47 PM UTC

While it seems on the surface like a waste of time (albeit amusing one), I actually expect this is a great project to learn from because of its use of Bayesian classifiers.

In other words, I'm TOTALLY going to be using this on my next project.

Re: Twss.js

#30

Post by aithleyadeno » Thu, Jan 12, 2012, 6:17 PM UTC

goo.gl/39SLa