Querying uncertain data with aggregate constraints

Abstract

Data uncertainty arises in many situations. A common approach to query processing uncertain data is to sample many "possible worlds" from the uncertain data and to run queries against the possible worlds. However, sampling is not a trivial task, as a randomly sampled possible world may not satisfy known constraints imposed on the data. In this paper, we focus on an important category of constraints, the aggregate constraints. An aggregate constraint is placed on a set of records instead of on a single record, and a reallife system usually has a large number of aggregate constraints. It is a challenging task to find qualified possible worlds in this scenario, since tuple by tuple sampling is extremely inefficient because it rarely leads to a qualified possible world. In this paper, we introduce two approaches for querying uncertain data with aggregate constraints: constraint aware sampling and MCMC sampling. Our experiments show that the new approaches lead to high quality query results with reasonable cost.

Aggregate Query Answering on Possibilistic Data with Cardinality Constraints

“…Moreover, they do not offer solutions which do not require exponential enumeration of all possibilities. Other systems support efficient implementations of Monte Carlo methods [9], [12], but empirically we show that such sampling does not explore the full range of possibilities. Thus, these approaches do not provide correct answers for queries such as those described in Examples 1 and 2.…”

“…Recent work in parallel to this [12] discusses sampling only those possible worlds which satisfy some global aggregate constraints. While ostensibly similar, the problem differs from ours since uncertainty is only over tuple values, and techniques provide a sampling mechanism, not a data model.…”

Generating top-k packages via preference elicitation

“…Finally, in [28], the authors study the problem of constrained sampling. However, the underlying distribution they consider is discrete, whereas we use Gaussian mixtures.…”

Untitled

“…В годы начала космической эры во никла потребност в исследовании поведени идкости в услови х невесомости. К и учени данной проблематики ( адачи определени форм равновеси , условий устойчивости, описани тепловой конвекции, Таврический вестник информатики и математики , 1 (46)' 2020 проблемы малых дви ений) был привлечен отдел прикладной математики. Подобными адачами чут по е стали анимат с в Вычислител ном центре АН СССР (г. Москва) и в Институте математики НАН Украины (г. Киев).…”

“…а период почти 60-летней научной де тел ности Николай Дмитриевич c соавторами написал более 250 научных работ, более 15 учебных пособий, и дал 8 монографий (полный список трудов доступен на сайте http://nikolay-d-kopachevsky.com). Он вл етс аслу енным де телем науки и техники Украины и России, лауреатом государственной премии Украины 2013 года (в составе авторского коллектива) а Таврический вестник информатики и математики , 1 (46)' 2020 цикл научных работ по гидромеханике акономерности волно-вихревых процессов в сплошной среде , лауреатом премий имени В. И. Вернадского и кавалером Ордена а аслуги 3-й степени. Неоцениму поддер ку в и ни и работе Никола Дмитриевичу ока ывала его ена, с которой он про ил много лет в л бви и согласии.…”