作者: Jan Chomicki
关键词: Parameterized complexity 、 Winnow 、 Theoretical computer science 、 Relational algebra 、 Logical framework 、 SQL 、 Information system 、 Computer science 、 Data mining 、 Embedding 、 Operator (computer programming)
摘要: The handling of user preferences is becoming an increasingly important issue in present-day information systems. Among others, are used for filtering and extraction to reduce the volume data presented user. They also keep track profiles formulate policies improve automate decision making.We propose here a simple, logical framework formulating as preference formulas. does not impose any restrictions on relations, allows arbitrary operation predicate signatures It makes composition relations straightforward. We natural embedding formulas into relational algebra (and SQL) through single winnow operator parameterized by formula. possible formulation complex queries, example, involving aggregation, piggybacking existing SQL constructs. leads way definition further, preference-related concepts like ranking. Finally, we present general algebraic laws governing its interactions with other operators. preconditions applicability captured provide formal foundation optimization queries. demonstrate usefulness our approach numerous examples.