Data Analytics - Papers - Shmueli et al. 2016 - - Association Rules and Collaborative Filtering In Data Mining for Business Analytics

9 important questions on Data Analytics - Papers - Shmueli et al. 2016 - - Association Rules and Collaborative Filtering In Data Mining for Business Analytics

What is 'Item Based Collaborative Filtering'?

When items are bought by the same user, one can filter based on this data

What is 'Association Rule Discovery'?

The goal is to identify item clusters in transaction type databases. Answering the question; Which groups of products tend to be purchased together?

What are two shortcoming of Association Rules?

  1. The profusion/abundanc of rules that are generated
  2. Rare combinations tend to be ignored because they do not meet the minimum support requirement
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

How can you overcome the shortcoming of not meeting the minimum support requirement of Association Rules?

Use higher level hierarchies as the items, so instead of using the titels of books, you can use the Type of books (horror, thriller, fiction, etc)

What is the main advantage of Association Rules?

It generates clear simple rules of the form 'IF X is purchased THEN Y is also likely to be purchased.

It is therefore Transparant and Easy to Understand.

What are the two stages of the process creating an Association Rule

  1. A set of candidate rules bases on frequent itemsets is generated
  2. From these rules, the rules that indicate the strongest association between items are selected


The user also indicates a minimum support and confidence values, and the Lift Ratio is used to compare the efficiency of the rule. (by detecting a real association and comparing this to a random combination)

Explain the difference between the use of the association rule and item based collaborative filtering with an example;


If the rule is "IF milk, THEN cookies and cornflakes," then the association rules would recommend
cookies and cornflakes to a milk purchaser, while item-based collaborative filtering would recommend
the most popular single item purchased with milk.

What is an important component for success of collaborative filtering?

That users provide feedback on the recommendations provided and have sufficient information on each item

What are two disadvantages of Collaborative Filtering that are mentioned in the paper?

  1. They cannot generate recommendations for new users or new items
  2. User based Collaborative Filtering becomes computationally challenging with huge numbers of users - alternatives such as items-based methods or dimension reduction are then preffered

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo