StackExchange Tagger

shagunsodhani.in

22 points·shagunsodhani··6 comments

Looks interesting! I hope they don't compare their results with the tags on the page to determine how good their tagger is, because there are plenty of questions that the authors tag incorrectly.

Great work. It might be a good idea to add usage instructions to the GitHub readme. A few questions:

— Are you releasing the dataset? Is there any place I can get more details about the dataset? I'd love to see how well a simple RNNLM does.

— Is it an end-to-end pipeline? As in, if I have a SE question url or text, can I just pass that in and it'll automatically do all the preprocessing and give me the predicted tag?

— Is the trained model available somewhere?

Thanks for the suggestion @abhshkdz! I will add the setup/usage instructions. As for your questions, - You can use a sample from http://blog.stackoverflow.com/tags/cc-wiki-dump/ - The end-to-end pipeline is missing. But that's easy to add. Will update the project soon. - The project was done some time back and we do not have the trained model with us now.

I see, okay. Thanks!

Link to supporting paper: https://docs.google.com/viewer?a=v&pid=sites&srcid=ZGVmYXVsd...

Sorry, But I can't access this link. Looks like it is dynamic generated, but I got to access from here: https://sites.google.com/site/sanketmehtaiitr/home/stack-exc...

Reply on news.ycombinator.com