Recommendation Engine

Why Trust Techopedia

What Does Recommendation Engine Mean?

A recommendation engine is a machine learning (ML) system that uses explicit and implicit end user feedback to make predictions about what digital content — including ads — an end user might be interested in viewing.


In e-commerce, recommendation systems can be used to segment partner website visitors into cohorts and target them with relevant product and content suggestions. Recommendations tend to be most accurate when the engine has access to detailed information about content items as well as historical data for specific end users and cohorts.

Recommendation engines are typically categorized as being collaborative filtering systems, content-based systems or hybrid systems.

Techopedia Explains Recommendation Engine

Recommendation engines are also known as recommendation systems. As mobile apps and other advances in technology continue to change the way users choose and use information, recommendation engines are becoming an integral part of websites and software products.

How does a recommendation engine work?

Recommendation engines use machine learning algorithms and statistical modeling to calculate the probability that an end user will choose to take a suggested action.

Content-based engines base predictions around end user interest for a specific content item. When a content item has been acted upon, the engine uses metadata to identify and recommend similar content items. This type of recommender system is commonly used by news websites and can be recognized by prompts such as “You may also be interested in reading…”

Collaborative recommendation engines analyze end user behavior within a specific platform to make predictions about a specific end user or cohort. This type of recommender system can be memory-based or model-based and is commonly used by e-commerce websites. Collaborative engines and can be recognized by prompts such as “People who bought this, also purchased…”

Hybrid recommendation engines compensate for the limitations of content-based and collaborative models by using both metadata and transactional data to suggest future actions. Hybrid engines can analyze what digital content an end user has acted upon previously and recommend similar content, while also factoring in demographics and historical data generated by people with similar interests. Hybrid engines can be recognized by multiple types of prompts on the same page.

Role of implicit and explicit data

Content-based recommendation engines often rely on explicit data to make predictions. This type of data has to be entered manually by the end user. Star ratings and user profiles are a popular source of explicit data for recommendation engines.

Collaborative recommendation systems use implicit data. This type of data is based on user behavior and relies on cookies that once accepted, can be used to track an end user’s online behavior across a single website.

Hybrid recommendation engines use both implicit and explicit data to make recommendations. This type of sophisticated recommendation engine uses both content-based and collaborative recommendation prompts and can use data from multiple partner websites to make recommendations.


Related Terms

Margaret Rouse
Technology Expert
Margaret Rouse
Technology Expert

Margaret is an award-winning technical writer and teacher known for her ability to explain complex technical subjects to a non-technical business audience. Over the past twenty years, her IT definitions have been published by Que in an encyclopedia of technology terms and cited in articles by the New York Times, Time Magazine, USA Today, ZDNet, PC Magazine, and Discovery Magazine. She joined Techopedia in 2011. Margaret's idea of a fun day is helping IT and business professionals learn to speak each other’s highly specialized languages.