liveBook · Manning

Confidence

As explained in Advanced Algorithms and Data Structures

Confidence is the conditional frequency with which one product is purchased when another product is purchased.

Confidence measures how often one product appears when another product is purchased. For products X and Y, compute the number of purchases containing both and divide by the number containing Y; use the result to quantify their association. section 5.1

Use confidence to support a simple product-grouping and recommendation workflow:

  • Count purchases in which products occur together. section 5.1
  • Divide the joint-purchase count by the purchases containing the comparison product. section 5.1
  • Set a confidence threshold for merging strongly associated product groups, then recommend the strongest associations. section 5.1

Product recommendations

An online store can identify frequent pairings such as laptops with external disks or tennis rackets with tennis balls. When a customer adds a related item, it can suggest a strong associated product. section 5.1

How Advanced Algorithms and Data Structures covers it

How deep it goes
Defines the conditional-frequency calculation and applies thresholds, but stops short of detailed recommender-system implementation.
Examples
Uses laptops with external disks, tennis rackets with tennis balls, and follow-up product suggestions.
What you can do after
Apply confidence by counting co-purchases, calculating conditional frequency, and thresholding associations for recommendations.
How it connects
Connects confidence with recommender systems, hierarchical clustering, and disjoint sets.

Also explained in

Compare how 2 books explain confidence →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Advanced Algorithms and Data Structures. The section links point to the parts of the book it draws on.