Database Reference
In-Depth Information
Chapter 5
Consistent Answers to Aggregate Queries
Abstract In this chapter, we focus our attention on the evaluation of queries of
a form different from that considered in the previous chapters. Specifically, we ad-
dress the problem of evaluating aggregate queries over data inconsistent w.r.t. aggre-
gate constraints, and introduce a technique for providing range-consistent answers
of these queries. The range-consistent answer of an aggregate query is the narrowest
interval containing all the answers of the query evaluated on every possible repaired
database. Three types of aggregate queries are investigated, namely
,
, and
SUM
MIN
queries. Similarly to the approach described in the previous chapter for com-
puting card -minimal repairs, the technique presented here is based on a translation
to the Mixed Integer Linear Programming (MILP) problem, thus enabling well-
established techniques for MILP resolution to be exploited for computing range-
consistent answers.
MAX
5.1 Aggregate Queries
In this chapter, we consider a form of queries, called aggregate queries , which is a
more expressive than that considered in the previous chapters, where the complexity
of the consistent query answer (CQA) problem is investigated for atomic queries.
Aggregate queries consist in evaluating an aggregate operator (
) over
the tuples of a relation satisfying the desired condition. In this setting, we consider a
more specific notion of consistency of answers, namely range-consistent query an-
swer (range-CQA), which was introduced and shown to be more suitable for queries
evaluating aggregates in [8]. Basically, the range-consistent answer of an aggregate
query is the narrowest interval containing all the answers of the query evaluated on
every possible (minimally) repaired database.
The following example describes a scenario where aggregate queries can be ef-
fectively used for retrieving useful information.
Example 5.1. Consider Example 1.3, where balance-sheet data of a company are
automatically acquired from paper documents. The analysis of the financial condi-
,
,
SUM
MIN
MAX
Search WWH ::




Custom Search