Database Reference
In-Depth Information
FROM < FTS ¡ relation >
HAVING < value of α,β >
WHERE < selection=join conditions >[anyDB]/[sameDB]
Union/Intersection/Minus [anyDB]/[sameDB]
SELECT <target attributes >[anyDB]/[sameDB]
WITH SOURCE CONTEXT {optional}
FROM < FTS relation >
WHERE < selection/join conditions >[anyDB]/[sameDB]
HAVING < value of α,β >
In summary, the main features offered by FTS-SQL include:
•
FTS-SQL satisfies the closure property. Given FTS-relations, queries produce FTS-relations as
result.
•
FTS-SQL allows source options [anyDB]/[sameDB] to be specified on the SELECT and WHERE
clauses, as well as on the
union, intersection & minus
FTS operations. Source option on SELECT
clause determines if tuples from different local databases can be combined during projection.
Source option on WHERE clause determines if tuples from different local databases can be com-
bined during join.
•
Defining source predicates to operand FTS-relations, can make formulation of queries to a specific
local database. A source predicate is represented by <relation name>.source in <set of local
DB
id
>
•
Queries can be on both crisp and fuzzy attributes. By specifying different values of α,β in the
HAVING clause, the precision in fuzzy query formulation can be adjusted. α
∧
[0,1] is used for
α-resemblance of fuzzy values in fuzzy predicates, where as β
∧
[0,1] imposes a constraint while
selecting a fuzzy tuple. Default value for both of them is 1 which corresponds to crisp values.
It can be shown that some of the FTS-SQL queries involving the
anyDB
option can not be per-
formed both directly and indirectly by the normal SQL. Even when some of the FTS-SQL queries
can be computed by FSQL expressions, we believe that FTS-SQL will greatly reduce the effort of
fuzzy query formulation on fuzzy relational multidatabase of type-2.
•
Using the clause WITH SOURCE CONTEXT, tuples in the fuzzy query results can be joined with
its source related information available in
source relation
table.
Figure 2.
