Rekommendation

Rekommendation is a proposal or endorsement intended to influence a choice or action. In everyday use it covers informal advice, such as tips on what to read or which product to buy, and more formal endorsements, such as letters of recommendation that attest to a person’s qualifications for study or employment.

Origin and usage: The term is used in Swedish and other Nordic languages, borrowed from Latin recommendatio through European languages. In Swedish, "rekommendation" is used for both informal advice and formal endorsements, while "rekommendationsbrev" denotes a letter of recommendation.

Contexts and types: Recommendations can be advisory (personal guidance), professional (schools or employers), or algorithmic (system-generated suggestions). In consumer contexts, recommendation messages aim to steer choices; in academia or hiring, recommendations formally vouch for a person’s abilities.

Algorithmic recommendations: Recommender systems suggest items or content based on user data and item features. Approaches include collaborative filtering, content-based filtering, and hybrid methods. They rely on historical interactions, ratings, and metadata, and raise concerns about privacy, bias, and the potential to create filter bubbles. Common evaluation metrics include accuracy measures such as precision and recall, and ranking metrics like nDCG.