Prioritized preferences and choice constraints

摘要: It is increasingly recognised that user preferences should be addressed in many advanced database applications, such as adaptive searching databases. However, the fundamental issue of how impact semantics and rankings a relation not resolved. In this paper, we model preference term involving one attribute hierarchy its underlying data values formalise notion Prioritized Preferences (PPs). We then consider multiple ranking tuples relational table. examine given set PPs on possible choices develop new Choice Constraints (CCs) relation, r. Given two PPs, X Y, CC, ≤ satisfied r, if choice according to Y no less than X. Our main results are related these notions CCs their interesting interactions with well-known Functional Dependencies (FDs). First, exhibit sound complete three inference rules for further prove each closed there exists precisely satisfies preferences. Second, establish five CCs. Finally, show soundness completeness mixed systems FD-PPs FD-CCs. All novel incorporating design modelling, since FDs together capture rich