Show HN: My Leeds Hack Day project, a language detection API(polyglossy.com)
polyglossy.com
Show HN: My Leeds Hack Day project, a language detection API
http://polyglossy.com/
3 comments
That's pretty cool! One of my friends suggested to use dictionaries and bloom filters but I've wanted to build probabilistic language models.
Do you have any accuracy stats? I'm guessing my approach might work better in some cases because the models include frequency information too. Did you experience significant accuracy loss when adding new languages? Anyway, I'll run it over my test data and compare.
Do you have any accuracy stats? I'm guessing my approach might work better in some cases because the models include frequency information too. Did you experience significant accuracy loss when adding new languages? Anyway, I'll run it over my test data and compare.
Do you have any accuracy stats?
No, but as you have noted, the method has the intrinsic property of being less accurate with fewer words and more accurate the longer the text. As my anticipated use was for documents over 10-20 words, this was OK. I expect the other techniques I outlined that I'm switching to to yield more accurate results across the board.
No, but as you have noted, the method has the intrinsic property of being less accurate with fewer words and more accurate the longer the text. As my anticipated use was for documents over 10-20 words, this was OK. I expect the other techniques I outlined that I'm switching to to yield more accurate results across the board.
...and these are my results: http://imgur.com/9U6II.png
I've only kept English, French and German for WhatLanguage. It looks like the Markov chains approach is indeed more accurate for shorter phrases but the dictionary approach gets slightly better for longer sentences.
Uhhh..what?
http://imgur.com/LH2Hi.png
http://imgur.com/LH2Hi.png
Sorry, only supports English, German and French at the moment
"Je t'aime" is wrongly classified as English. (Although understandable, since it's short)
How does this work? You claim it has "zero language knowledge," so how does it classify? Did it start out with nothing, and then train it on a corpus?
EDIT: nevermind, found your slides (http://polyglossy.com/presentation.pdf [pdf]) How big of a corpus did you use?
How does this work? You claim it has "zero language knowledge," so how does it classify? Did it start out with nothing, and then train it on a corpus?
EDIT: nevermind, found your slides (http://polyglossy.com/presentation.pdf [pdf]) How big of a corpus did you use?
I've used around 40KB of text per language for training.
wouldn't something like 2-gram bayes be more accurate & faster?
I don't know. I haven't really done NLP before. My guess is that it would be slower, but not by a significant amount. It might be more accurate, so I could give it a shot.
I don't think 2-gram bayes uses any NLP at all. To my understanding, it's just like naive bayes, but can use word combinations to improve accuracy. In terms of complexity, bayes classification is not much more than O(N).
Any chance of posting the source?
Yeah, I'll probably push it to github later this week.
...and here it is: https://github.com/lgeek/polyglossy (with a slight delay because I've worked on improving the accuracy)
Good luck with your project, just wanted to say Google Translate API also has language detection, and gives you a degree of confidence.
http://www.google.com/uds/samples/language/detect.html
It uses a slightly weird technique, though. Dictionary based and using a bloom filter for memory efficiency. Going forward, though, I plan to rewrite it to use a combination of n-grams and language "fingerprints."