Hello All,
This is a question about the SSQR #2, Question 7:
Snappy
The following passage will be used to answer questions #6 - #10
Snappy is a crowdsourcing app designed with the purpose of making the world a happier place. The app collects photos users submit of “Happy Moments”. Photos are tagged with the location where they were taken. Users have access to a map where they can zoom in to photos submitted at specific locations. A leaderboard displays the locations with the most “Happiness Moments” and the users who are the “snappiest” in each city due to the volume and rating of “Happiness Moments” they have submitted. Users can rate other “Happiness Moments” using a limited set of emojis and can also leave comments which are moderated using machine learning in an effort to only allow positive comments, by checking each comment for common negative words learned through a training set.
- Which of the following data is LEAST likely to be included in the rankings of users on the leaderboard?
A. The number of ratings a “Happiness Moment” received.
B. The number of comments a “Happiness Moment” received.
C. The location of a “Happiness Moment”.
D. The number of “Happiness Moments” a user has submitted.
Answer B. The number of comments a “Happiness Moment” received.
Could it be described “why” Answer B is the LEAST likely to be included?
Students chose a variety of responses but not Answer B.
Thank you!
Hmm well this is more on the side of data analytics than actual Machine Learning and how to feed it proper data for it to function properly usually the data you want to feed it the most descriptive parts of what it’s looking at “Carrot on the stick approach” to achieve the desired results
Lets break down What the answers are looking at
A: is looking at ratings which plays a crucial role into how it’s interpreted by multiple individuals
B: is looking for a number of comments that doesn’t really tell us much though since it’s only looking at a number value here and not what each comment says
C: the location i would also say plays a unique role especially if the user is interested in a specific area we should up the chances of seeing another picture in a nearby location rather than at random
D: this could also be arguable as mostly unimportant in my eyes at least since i don’t think there would be any significant value in it however this could be used as a metric as to how active they are on the platform since it’s been broken down on an individual level… in my opinion it’d be better to base it off of how many reactions they’ve revived rather than how they reacted but I’d say overall it’s better than B
that’s just my thinking on the matter on what the data means maybe someone can explain it better hope this helps!
1 Like
if you read the specification carefully you will find which answer directly supports the leaderboard of users.
A. A leaderboard displays … the users who are the “snappiest” in each city due to the volume and rating of “Happiness Moments” they have submitted.
B. A leaderboard displays … the users who are the “snappiest” in each city due to the volume and rating of “Happiness Moments” they have submitted.
C. A leaderboard displays … the users who are the “snappiest” in each city due to the volume and rating of “Happiness Moments” they have submitted.
D. A leaderboard displays … the users who are the “snappiest” in each city due to the volume and rating of “Happiness Moments” they have submitted.
One could argue for answer D because the total number of “Happiness Moments” submitted could be split between more than one city. But that is not as obvious as answer B.