GoCrackIt

Quiz 6 Questions and Answers

Question No 1

1. Why are recommendation engines becoming popular?

  • Users have lesser time, more options and face an information overload
  • It is mandatory to have recommendation engine as per telecom rules
  • It is better to recommend than ask user to search on mobile phones
  • Users don't know what they want

Answer : Users have lesser time, more options and face an information overload

Internet speeds are becoming faster, prices are lowering and competitors are mushrooming. It is getting very difficult for online players to retain users on their platforms and hence it has become essential to personalize the experience through technology...

Question No 2

2. What kind of information does a Recommendation Engine need for effective recommendations?

  • Users' explicit interactions such as information about their past activity, ratings, reviews
  • Users’ implicit interactions such as device they use for access, clicks on a link, location, and dates
  • Other information about profile, such as gender, age, or income levels
  • All of the above

Answer: All of the above

Recommendation engines use different models to predict what a user might like. For this, they need as much information as possible about each user.

Question No 3

3. What are different Recommendation Engine techniques?

  • Content based filtering
  • Collaborative filtering
  • Knowledge based system
  • All of the above

Answer: All of the above

Collaborative filtering, content filtering, knowledge based filtering and different hybrid approaches are used for building recommendation engines.

Question No 4

4. Which of the below is collaborative filtering based Recommendation Engine?

    Egg
  • Items inspired by your history
  • Items which customers like you viewed earlier

Answer : Items which customers like you viewed earlier

Collaborative filtering is based on how people with similar likes will have similar likes and does not worry about what explicit data the user has given. Hence items which customers like you viewed earlier is the correct answer.

Question No 5

5. What are the challenges in Content Based Filtering?

  • Need to capture significant amount of users' information, which may lead to regulatory and pricing issues
  • Need to have information of all users across different demographics
  • Need to have lower number of categories for content based filtering to be effective
  • Need to have user's social media and digital footprint

Answer: Need to capture significant amount of users' information, which may lead to regulatory and pricing issues

Since recommendations are based out of users’ history, a lot of user information is required which may lead to regulatory and privacy issues and also this will work best when product categories are lesser so that you can show more accurate similar products.

Question No 6

6. What are the three pillars of Netflix's Recommendation Engine?

  • History of films and TV Series, History of User on Netflix, Taggers who tag content
  • History of films and TV Series, History of User on Netflix, Machine Learning Algorithm
  • History of User on Netflix, History of films and TV Series, Taggers who tag content
  • History of User on Netflix, Taggers who tag content, Machine Learning Algorithm

Answer: History of User on Netflix, Taggers who tag content, Machine Learning Algorithm

History of User on Netflix, Taggers who tag content, Machine Learning Algorithm. Netflix has been one of the pioneers of recommendation engines.

Question No 7

7. Which of the following businesses would be least likely to use Recommendation Engine?

  • Facebook
  • WhatsApp
  • Instagram
  • Ola

Answer: WhatsApp

WhatsApp. Well, thats what the founders had agreed to : No Ads, No Games, No Gimmicks. And for these reasons, WhatsApp does not need to use a Recommendation Engine :)

Question No 8

8. What percentage of Amazon's sales are due to its recommendation system?

  • Around 10%
  • Around 35%
  • Around 55%
  • Around 80%

Answer: Around 35%

Its around 35%. You can read more here :

Question No 9

9. Which of the following are not used for filtering in a Recommendation Engine?

  • Sine Similarity
  • Cosine Similarity
  • Jaccard Similarity
  • Euclidean Distance

Answer: Sine Similarity

Cosine SImilarity, Jaccard Similarity, Euclidean Distance and others are algorithms used to build recommendation systems. As of now there is nothing such as sine similarity.

Question No 10

10. For an ecommerce website, which of the following is explicit data?

  • Order history
  • Page views
  • Cart events
  • Product feedback

Answer: Order history

Explicit data is something which ecommerce websites and other businesses specifically ask from the users, while implicit data is collected through their behaviour on the website.