To automatically assigns tags for questions posted on a forum such as StackOverflow or Quora.
To automatically assigns tags for questions posted on a forum such as StackOverflow or Quora :
A Question can be clasified in to two ways for such forums :
QUESTION
|__________> Category
|__________> Tags
Here Tags, may be seen as a sub list of categories.
For a clear vision lets take example :
Question : What is the step by step guide to invest in share market in india?
--> It may come under category like "share market", "investment"
--> But it may have tags like "step by step guide","guide",'"india","market"
Basically tags are to attract more corwd towards the problem so that it can be addressed in a broad way.
->> TAGS are something that are already available in the text itself and just need to explored with some processing.
->> While doing so i noticed that methods like LDA or NMF may not help you out here. ( Unsupervised Learning .. )
->> Using some basic NLP tweeks may give you a list of tags that can be suggested to users.
->> Not using any kind of ML algorithms here !