Which pennkey are you using to turn in your code via turnin? *
Your answer
Part 1: Gathering articles
What techniques did you use to gather your urls? Where did your articles come from? *
Your answer
If we are trying to build a comprehensive database of gun violence incidents across the country, what problems might arise from the way we gathered articles in this assignment? Are there any potential biases in any of these methods? *
Your answer
Parts 2 & 3: Re-evaluating your classifier's performance
How many workers contributed to the task you posted? *
Your answer
What was the estimated hourly rate for the workers? *
Your answer
How do you think you could increase their hourly rate without increasing the amount that you pay them per item? *
Your answer
How many of the articles did you have labeled? (Hopefully it was 500, but it is okay if it was less.) *
Your answer
How many of the articles did they label as "gun-related"? *
Your answer
Based on this, what is the estimated precision of your classifier? *
Your answer
What is precision? How is it computed, and how does it differ from accuracy? *
Describe this in terms of true positives, false positives, true negatives, and false negatives.
Your answer
What is recall? Why can't we compute the recall of your classifier using the data you collected from Crowdflower? *