challenge-prompt

QUORA ML PROBLEM: ANSWERED
As Quora gets many questions every day, a challenge we have is to figure out good, interesting and meaningful questions 
from the bad. What questions are valid ones that can be answered? What questions attract reputable answers that then get
upvoted? Can you tell which questions will likely get answers quickly, so that we can surface them in real-time to our 
users?

For this task, given Quora question text and topic data, predict whether a question gets an upvoted answer within 1 day.


QUORA ML PROBLEM: INTEREST
Quora uses machine learning algorithms to try to generate interesting news feeds and digest emails for people. Before 
a question gets an answer, we'd like to be able to make use of all the available information we have to be able to 
predict how interesting or relevant the question is to people. Ideally, we'd like to be able to tell this in real-time
as soon as a few people have viewed it, by measuring the people following the question as a proxy of interest. Can you 
tell what questions will get the most followers?

For this task, given Quora question text and topic data for questions with 0 visible answers, predict the ratio of 
viewers to followers.


QUORA ML PROBLEM: VIEWS
Not all of the questions on Quora are as interesting or appealing to people on the web. Can you tell what questions 
can organically attract the most viewers? What about questions that eventually become viral? Which questions are 
timeless and can sustain traffic?

For this task, given Quora question text, topic data, number of answers and number of people promoted to, predict the 
number of views per day in age of the question.