Concept reduction selects a subset of formal concepts for knowledge discovery without losing original information, thereby simplifying knowledge representation. However, its practical applications are limited. To address these problems while retaining its advantages, this paper proposes a novel reduction theory from the perspective of object intents, called attribute combination reduction to preserve object intents. Specifically, this paper defines attribute combination reducts and explores their properties. Moreover, this paper also examines the relationship between attribute combination reducts and concept reducts, and gives a method for obtaining partial attribute combination reducts from concept reducts. Finally, a case study demonstrates the applicability of attribute combination reduction.