Query Relaxation
Part of the Query Understanding series by Daniel Tunkelang.
Overview
Query relaxation handles the situation where a query is too specific to return meaningful results. Rather than showing an empty page, the system progressively loosens constraints until it can surface something useful. This might mean dropping the least important terms, widening category constraints, or softening specific attribute requirements. The art is in choosing the right order of relaxation — removing constraints that don’t change the core intent first — and in communicating to the user that the results shown are approximations rather than exact matches. Both extremes are bad: relaxing too aggressively produces irrelevant results, while relaxing too conservatively leaves users with empty pages.
Note: Article content behind Medium paywall — accessible at source URL with Medium account.