Database Reference
In-Depth Information
For our purposes, we will view a
fuzzy query
as a combination of a number of imprecisely specified
(fuzzy)
conditions
on attribute values to be met. The fuzzy preferences in queries are introduced
inside
query conditions and
between
query conditions. For the former, fuzzy preferences introduced
inside
query
conditions via flexible search criteria which make possible to indicate a graded desirability of particular
values. For the latter, fuzzy preferences
between
query conditions are given via grades of importance
of particular query conditions.
The research on fuzzy querying has already a long history, starting with the seminal works of Za-
deh during his stay at the IBM Almaden Research Center in the late 1970s, and the first attempt to use
fuzzy logic in database querying by Zadeh's doctoral student Tahani (1977). The area has soon enjoyed
a great popularity, with many articles in the early period, related both to database querying per se and
a relevant area of textual information retrieval [cf. (Bookstein, 1980; Bosc & Pivert,1992a; Kacprzyk
& Ziółkowski, 1986; Kacprzyk, Zadrożny & Ziółkowski, 1989), etc.], and books, cf. (Zemankova &
Kandel, 1984). Later, the field has become an area of huge research efforts. For an early account of
main issues and perspectives, we can refer the reader to Zemankova & Kacprzyk (1993), while for re-
cent, comprehensive state of the art type presentation - Rosado, Ribeiro, Zadrożny & Kacprzyk (2006),
Zadrożny, De Tré & De Caluwe (2008), etc.
Some novel and practically relevant developments in broadly perceived data mining and data ware-
housing have greatly increased interest in fuzzy querying. A notable examples are here works on the
combination of fuzzy querying and data mining interfaces for an effective and efficient linguistic sum-
marization of data [cf. (Kacprzyk & Zadrożny, 2000a; Kacprzyk & Zadrożny, 2000b)] or fuzzy logic
and the OLAP (Online Analytical Processing) technology (Laurent, 2003).
The purpose of this paper is to present those developments of, and related to fuzzy querying in a
focused way to show their essence and applicability. We will start with a general introduction to fuzzy
querying in (numeric) relational databases, adding some remarks on the use of object oriented paradigm.
Then, we will mention some attempts to add an additional information of user specified preference
via so called
bipolar queries
which, in their particular form, make possible to include mandatory and
optional requirements. Then, we will show the usefulness of fuzzy queries as a vehicle for an effective
and efficient generation of linguistic summaries.
BACKGROUND
A
relational database
is meant as a collection of relations, characterized by sets of
attributes
and populated
with
tuples
, which are represented by tables comprising
rows
and
columns
. In what follows we will freely
use interchangeably both notions of relations and tables what should not lead to any misunderstandings.
Each relation
R
is defined via the
relation schema
:
(
)
( )
( )
( )
R A Dom A A Dom A
:
,
:
,
,
A Dom A
n
:
(1)
1
1
2
2
n
where
A
i
's are the names of
attributes
(
columns
) and
Dom
(
A
i
)'s are their associated
domains
.
To retrieve data, a user forms a
query
specifying some conditions (criteria). The retrieval process
may be meant, in our context of fuzzy querying, as the calculation of a
matching degree
for each tuple
of relevant relation(s), usually from [0,1], as opposed to {0,1} as in the traditional querying.
