T. Gaasterland, J. Minker, and A. Rajasekar. Deductive database systems and knowledge base systems. In Proceedings of VIA 90, Barcelona, Spain, October 1990.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and a cooperative response would identify the false presupposition to the user, there are no mothes in the database. USER CONSTRAINTS In an extension to the cooperative answering system, user constraints are used to account for a user s restrictions on the world when answering questions [GMR90, GRM90] A user s intentions and needs can be modelled as a set of constraints and then used to alter a user s query or the subqueries in the search tree for the query. The user constraints for a particular user can be semantically compiled with a user s query so that the search space of the ....

T. Gaasterland, J. Minker, and A. Rajasekar. Deductive database systems and knowledge base systems. In Proceedings of VIA 90, Barcelona, Spain, October 1990.

....Then A 1 : technical description of parts and how they work) System knows user is an eight year old. Then A 2 : visual description of object and what it does) The two answers use different vocabulary and describe different properties of the telephone. User constraints are introduced in [21] and formalized in [18] The notion of user constraints (UCs) is analogous to that of integrity constraints in deductive databases. A UC is of the same syntactic form as an IC, but does not have the same meaning; an IC must be logically consistent with the database, whereas a UC does not. A UC ....

....Section 3. 3) ffl EDB search ffl violations with respect to previous answers ffl explanations of derivation paths (proofs) ffl positive cooperative information ffl employs heuristics to select best cooperative information [22] ffl detect redundancy in the query ffl use UCs to filter answers [18, 21] ffl relaxation (menu driven) 20] ffl query decomposition Any violation of an IC by a query indicates a possible misconception on the user s behalf. The most general IC necessitating that the query fails is found. These IC violations are detected and reported to the user. Sometimes a query may ....

[Article contains additional citation context not shown here]

T. Gaasterland, J. Minker, and A. Rajasekar. Deductive Database Systems and Knowledge Base Systems. In Proceedings of VIA 90, Barcelona, Spain, October 1990.

....must figure out alternative queries for themselves, formulate them, and pose them to the system. Alternatively, users may have restrictions on the knowledge domain that they would like to have addressed for every query that is asked. Users can express these restrictions through user constraints [ 5, 9 ] . User constraints express states that a user wants to disallow and states that a user wants to always persist. Each time a user asks a query, user constraints are applied to the query using semantic query optimization techniques. The resulting query produces answers that satisfy the user s ....

T. Gaasterland, J. Minker, and A. Rajasekar. Deductive Database Systems and Knowledge Base Systems. In Proceedings of VIA 90, Barcelona, Spain, October 1990.

....not only is it ready for relaxation, but we also have a query which is ready for optimization and query transformation techniques, if desired. Possible techniques include the use of integrity constraints for semantic optimization [Cha85, GGL 90] and the use of user constraints to guide search [GMR90] 3.4. Representation of Type Information The notion of answers in the neighborhood of a query requires the deductive database containing knowledge akin to that of semantic networks. The database must include how the objects and properties, represented via constants and predicates, ....

....queries which we can tell have a null answer set or do not provide any answers beyond the original query. We shall see some techniques for this in the following sections. 3. Apply semantic optimization and user constraints to the relaxed query as we would to the original query [Cha85, GGL 90, GMR90] 5.2. Declarative Knowledge Taxonomy clauses yield a declarative means of control, saving us from having to posit complex control strategies to accomplish relaxation. The fundamental difference between using reflection with taxonomy and reciprocal clauses to broaden the domain of a variable ....

[Article contains additional citation context not shown here]

Terry Gaasterland, Jack Minker, and Arcot Rajasekar. Deductive database systems and knowledge base systems. In Proceedings of VIA 90, Barcelona, Spain, October 1990.

....were found that should be supported. In [16] Gaasterland, Godfrey, and Minker present a mechanism to relax a given query into a set of more general, related queries. Evaluating one of these queries results in new answers with respect to the original query, and may be of interest to the user. In [14, 19, 20], Gaasterland, Minker, and Rajasekar introduced the notion of user constraints as a means to tailor the answer set with respect to the original query to include only information in which the given user is interested. Gaasterland, Godfrey, and Minker in the survey [15] review a number of other ....

....predicate in Datalog which has only a single clause in its definition. However, rule predicates with multiple clauses may be defined in the rulebase. The rulebase may be used to store other semantic information as well. This may include knowledge that characterizes users, or classes of users. In [14, 19, 20], user constraints are introduced and developed. User constraints have the form of ICs (and, hence, are representable in Datalog in the rulebase) but have a different semantic value. They are used to shield answers to a query for which it 11 It is also called University INGRES. 12 The ....

T. Gaasterland, J. Minker, and A. Rajasekar. Deductive Database Systems and Knowledge Base Systems. In Proceedings of VIA 90, Barcelona, Spain, October 1990.

....also necessary to pronominalize when possible or when required, for example, in the second reference to X in I2. 3 Meeting the User s Restrictions Another extension to the cooperative answering system allows it to take into account a user s restrictions on the world when answering questions [GMR90, GRM90] For example, continuing on the theme of gardens, a user may say I don t want to know about lilac plants . which could be expressed as: UC: plant(X) X=lilac. so that whenever plants arise in the processing of a user s query, they cannot be instantiated as lilac plants. Using the rules ....

Terry Gaasterland, Jack Minker, and Arcot Rajasekar. Deductive database systems and knowledge base systems. In Proceedings of VIA 90, Barcelona, Spain, October 1990.

....is ready for optimization and query transformation techniques, if desired. Possible techniques include the use of integrity constraints for semantic optimization [Cha85, GGL 90] and the use of user Relaxation as a Platform for Cooperative Answering p. 6 of 21 constraints to guide search [GMR90] 3.4 Representation of Type Information The notion of answers in the neighborhood of a query requires the deductive database containing knowledge akin to that of semantic networks. The database must include how the objects and properties, represented via constants and predicates, ....

....queries which we can tell have a null answer set or do not provide any answers beyond the original query. We shall see some techniques for this in the following sections. 3. Apply semantic optimization and user constraints to the relaxed query as we would to the original query [Cha85, GGL 90, GMR90] 5.2 Declarative Knowledge Taxonomy clauses yield a declarative means of control, saving us from having to posit complex control strategies to accomplish relaxation. The fundamental difference between using reflection with taxonomy and reciprocal clauses to broaden the domain of a variable ....

[Article contains additional citation context not shown here]

T. Gaasterland, J. Minker, and A. Rajasekar. Deductive Database Systems and Knowledge Base Systems. In Proceedings of VIA 90, Barcelona, Spain, October 1990.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC