Recommender system

As explained in Causal Inference for Data Science

A recommender system is a software system that suggests items or actions to users based on available information about them and the items.

A recommender system suggests items or actions using information about users and available items. Streaming services, online stores, and autocomplete systems use recommendations to help people choose among many possibilities. section 3.3.1

Prediction alone does not show that a recommendation changes behavior. A system might accurately predict a customer’s next purchase even when the customer would have made that purchase without the recommendation. section 3.3.1

The bread-purchase test

A customer buys bread every Monday, Wednesday, and Friday. Predicting the next purchase can reach 100% accuracy, yet the recommendation may have no influence. Compare users who receive recommendations with users who do not using an A/B test. section 3.3.1

Figure 3.12 Graph of a pure recommender system that does not influence the customer. This system may predict customer behavior accurately, but it does not affect their actions.

Use the causal graph to distinguish accurate prediction from an actual effect on customer behavior. section 3.3.1

Figure 3.13 Standard graph of a recommender system. The key feature here is the arrow from Historical Purchases to Recommendations, representing the recommender’s function.

Use the standard graph to trace how historical purchases inform recommendations. section 3.3.1

How Causal Inference for Data Science covers it

How deep it goes

Contrasts prediction with causation, introduces causal graphs, and motivates A/B testing without detailing recommender implementation.

Examples

Uses streaming recommendations, e-commerce suggestions, autocomplete, supermarket bread purchases, and a mobile-app A/B test.

What you can do after

Evaluate recommendation impact via an A/B test rather than relying only on predictive accuracy.

How it connects

Connects recommender systems to supervised learning, causal inference, causal graphs, A/B testing, and average treatment effect.

liveBookmanning.com

Read Causal Inference for Data Science on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Grokking Machine LearningA recommendation system is a software system that suggests items, content, or actions to users based on available information about them or their context.
• 

  liveBookmanning.com
  The Art of AI Product DevelopmentA recommendation system is a computational system that generates personalized suggestions by estimating which items or content may interest a user.
• 

  liveBookmanning.com
  AI-Powered SearchA software system that infers user interests or behaviors and delivers content or items that it predicts will match them, without requiring an explicit query for each result.
• 

  liveBookmanning.com
  Machine Learning Platform EngineeringA recommendation system is a software system that predicts or ranks items, content, or actions for users or other entities.
• 

  liveBookmanning.com
  Graph Algorithms for Data ScienceA recommendation system generates ranked or selected items for users based on user, item, or interaction information.
• 

  liveBookmanning.com
  Financial AI in PracticeA software system that generates recommendations from data or predictive models.
• 

  liveBookmanning.com
  First-Party Data ActivationA software system that predicts and presents items, content, or actions likely to be relevant to an individual.

Compare how 8 books explain recommender system →

Related concepts

• recommendationproduces · A recommendation is a suggested item, choice, or course of action, based on available information, intended to guide a decision.
• language modelused by · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.
• candidate sourcinghas part · Candidate sourcing is the process of identifying potentially suitable candidates from a larger pool for further consideration.
• similarity metricuses · A similarity measure is a rule that quantifies or characterizes how alike two objects or representations are according to selected properties.
• collaborative filteringincludes · Collaborative filtering is a recommendation technique that infers preferences or associations from interaction patterns among users, items, or other entities.
• decision treeused by · A decision tree is a branching structure that represents sequential tests or choices, with paths leading to outcomes, predictions, or actions.

This explainer was generated by Manning's knowledge graph from the text of Causal Inference for Data Science. The section links point to the parts of the book it draws on.