Classifying categories on Yahoo! Answer

now I have a seemingly easy but challenging task.I need to develop a data set of questions,and I classify the questions into two categories:

1. Factoid questions: "who is the current president of France."
2. Free questions: "Can you rate the cameras below for me,please?"

now I need to know the percentage of both categories on Yahoo! answer so that I could maintain my data set accordingly,but I don't know a go开发者_运维技巧od way of doing this statistic.Doing manually seems really impossible,does anyone have an idea?I would be really grateful,thanks.

You mean, recognize one from the other? Automatically, without any categorization from the site's end? That's probably going to be impossible.

I think the best you can do is compare some metrics. "Free" questions will probably tend to have more contributions with more text; they would be more heavily discussed if Y!Answers had a discussion system... "Factoid" questions may start with "What is..." more often ... and so on.

Maybe fetch 100 random questions, do a manual check and write down the percentages.