Recommendation Engines

Recommendation engines are specially designed systems that collect and store information (data) about a person that can be subjected to filtering and statistical analysis (analytics) to make recommendations about actions the individual could take or resources the individual could explore. For example, asking a person to name his or her three favorite musical groups provides a glimpse of the individual’s taste in music. When this information is merged with other known information such as the person’s age and address, a statistical profile can be generated to compare the individual’s musical interests with millions of other people to make suggestions about other musical groups the individual might enjoy. Recommendation engines are one component of a larger trend known as personalization technologies that seek to use large data sets to improve the user experience by making recommendations.

Recommendation engines are generally proprietary because of (a) the unique data sets (i.e., customers, subscribers, credit records) that form the base of the system, (b) specially designed algorithms that are tested and refined by mining the data for statistically significant relationships, and (c) the way in which the system is designed to present customized recommendations to users. Recommendation engines have been implemented in a variety of 21st-century applications for the general public including personalized radio, book purchasing suggestions, and movie watching recommendations. Commercial interest in recommendation engines is driven by a desire to mine customer data to reveal behavioral dispositions that can be managed in ways that stimulate increased sales. Whereas technology users may readily appreciate the value of a recommendation engine for its efficiency in discovery, it is not always clear how a specific recommendation system works or how to correct or modify misguided recommendations. Experts believe that recommendation engines will be widely implemented in education as a means of personalizing the learning experience in ways that optimizes the quality of a learner’s experience and improves educational outcomes. Before discussing the implications of recommendation engines for learning, this entry describes how these engines are used in daily life and how they work.