How recommendation feeds work quiz

Question 1

What is a recommendation feed?

Accepted Answer

A list of items an app picks for you based on what it predicts you'll like

Answer

A page where you type your password

Answer

A list of all videos ever made, in random order

Answer

A government list of approved websites

Question 2

What does it mean for a recommender to 'learn from examples'?

Accepted Answer

It studies past user choices to find patterns and predict what you'll like

Answer

It memorizes one rule written by a teacher

Answer

It guesses completely at random every time

Answer

It copies answers from another app

Question 3

What is 'collaborative filtering' in simple terms?

Accepted Answer

Finding users similar to you and suggesting what they liked

Answer

Recommending items based on the words in their titles only

Answer

Sorting items alphabetically

Answer

Blocking content the user dislikes

Question 4

What is 'content-based filtering'?

Accepted Answer

Recommending items similar to ones you liked, based on their features

Answer

Recommending only the newest items

Answer

Recommending whatever your friends posted

Answer

Recommending items chosen by a coin flip

Question 5

What is the difference between collaborative filtering and content-based filtering?

Accepted Answer

Collaborative uses similar users; content-based uses item features

Answer

They are exactly the same thing

Answer

Both only use the time of day

Answer

Collaborative ignores data; content-based ignores items

Question 6

What kind of data do recommendation feeds collect about you?

Accepted Answer

Your clicks, watch time, likes, searches, and similar behavior signals

Answer

Only your exact height and weight

Answer

Nothing at all; it just guesses

Answer

Only the color of your phone case

Question 7

What is 'watch time' and why do feeds care about it?

Accepted Answer

How long you actually engage with content, used as an interest signal

Answer

The time of day you log in

Answer

The total number of videos on the app

Answer

How fast your internet is

Question 8

What is an 'implicit signal' versus an 'explicit signal' in recommendations?

Accepted Answer

Explicit = ratings/likes you give; implicit = behavior the app infers

Answer

Explicit = behavior you didn't state; implicit = ratings you give

Answer

They mean the same thing

Answer

Both require typing a review

Question 9

Why might two people see completely different feeds in the same app?

Accepted Answer

Because the feed is personalized to each person's behavior and interests

Answer

Because the app is broken

Answer

Because newer phones get better content

Answer

Because everyone secretly sees the same thing

Question 10

What is a 'cold start' problem in recommendation feeds?

Accepted Answer

When there's too little data about a new user or item to predict well

Answer

When the app runs slowly in winter

Answer

When the server overheats

Answer

When a user logs out

Question 11

How does a new user get recommendations before the app knows much about them?

Accepted Answer

It shows popular/trending content and may ask for your interests

Answer

It refuses to show anything until a year passes

Answer

It shows only black screens

Answer

It copies your friend's exact feed forever

Question 12

What does 'engagement' mean to a recommendation system, and why can chasing it be risky?

Accepted Answer

Engagement means interaction; chasing it can push shocking, low-quality content

Answer

Engagement means file size; it has no downsides

Answer

Engagement means battery use; it improves quality

Answer

Engagement means screen brightness

Question 13

What is a 'filter bubble'?

Accepted Answer

When a feed mostly shows content matching your existing views, hiding other ideas

Answer

A tool that cleans your screen

Answer

A type of internet cable

Answer

A setting that deletes your history

Question 14

What is an 'echo chamber' and how is it related to feeds?

Accepted Answer

A situation where you mostly hear repeated similar opinions, reinforced by feeds

Answer

A room with good acoustics

Answer

A type of microphone

Answer

A feature that mutes sound

Question 15

Why is it important for recommendation feeds to include some variety (diversity)?

Accepted Answer

Variety aids discovery and reduces filter bubbles

Answer

Variety wastes data and should be avoided

Answer

Variety only matters for music

Answer

Variety means showing random ads

Question 16

What is the 'exploration vs. exploitation' tradeoff in recommendations?

Accepted Answer

Exploitation uses known likes; exploration tries new items to learn tastes

Answer

Exploitation tries new items; exploration repeats safe ones

Answer

Both mean deleting old data

Answer

They describe how to mine for gold

Question 17

What is a 'ranking' step in a recommendation feed?

Accepted Answer

Ordering candidate items by how likely you are to enjoy them

Answer

Deleting items you disliked

Answer

Choosing the app's color theme

Answer

Counting how many users exist

Question 18

What are 'candidates' in a recommendation pipeline?

Accepted Answer

A smaller pool of possibly-relevant items chosen before detailed ranking

Answer

People applying for a job at the company

Answer

Items you have already watched

Answer

The app's error messages

Question 19

Why can't a feed just score every single item in its huge catalog for each user?

Accepted Answer

Scoring millions of items per user would be too slow and costly

Answer

It legally cannot store items

Answer

Items disappear after one view

Answer

Users would get bored reading the list

Question 20

What is a 'user profile' or 'user vector' in a recommender?

Accepted Answer

A summary of your interests, often stored as a list of numbers

Answer

Your profile picture only

Answer

Your home address

Answer

A list of your passwords

Question 21

How can representing interests as numbers help match users to content?

Accepted Answer

Numbers let the computer mathematically measure how similar users and items are

Answer

Numbers make the screen brighter

Answer

Numbers hide your data

Answer

Numbers slow the system down on purpose

Question 22

This Python snippet finds the user's top interests. What does it print?

Accepted Answer

The two highest-scoring topics: science and music

Answer

The two lowest-scoring topics

Answer

All topics in random order

Answer

Nothing, it errors

Question 23

In this code, how does the system decide which video to recommend next?

Accepted Answer

It picks the video with the highest predicted score, 'space_doc'

Answer

It picks the first video listed

Answer

It picks a random video

Answer

It picks the lowest score

Question 24

What is a 'feedback loop' in recommendation systems and why can it be a problem?

Accepted Answer

When showing content trains the feed to show more of the same, narrowing it

Answer

A wire that connects two computers

Answer

A feature that gives you cash back

Answer

A way to reset your account

Question 25

What is 'bias' in a recommendation system?

Accepted Answer

When the system unfairly favors or ignores certain content or groups

Answer

A type of computer chip

Answer

A way to speed up the internet

Answer

A button to like a post

Question 26

Why should a recommendation feed be careful about recommending content to children?

Accepted Answer

Children are easily influenced, so feeds need extra safety filters

Answer

Children's feeds don't need any rules

Answer

Children should see the most shocking content

Answer

Feeds should hide everything from children

Question 27

What does 'transparency' mean for recommendation feeds, and why is it valued?

Accepted Answer

Being open about how the feed chooses content, building trust

Answer

Making the app screen see-through

Answer

Hiding all the app's rules

Answer

Showing only ads

Question 28

How can you, as a user, influence what your recommendation feed shows?

Accepted Answer

Through what you watch, like, search, and 'Not interested' choices

Answer

You cannot influence it at all

Answer

Only by buying a subscription

Answer

By changing your phone's wallpaper

Question 29

What is the difference between a personalized feed and a trending/popular list?

Accepted Answer

Personalized is tailored to you; trending shows what's popular for everyone

Answer

They are identical

Answer

Trending is only for one user

Answer

Personalized ignores your behavior

Question 30

Why might a feed sometimes recommend something you have no interest in?

Accepted Answer

Because predictions aren't perfect and it sometimes explores or misreads signals

Answer

Because the app dislikes you

Answer

Because feeds never make mistakes

Answer

Because you typed your password wrong

Test yourself