J.J. King. Query optimization by semantic reasoning. PhD thesis, Stanford University, 1981.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... data mining techniques can be used for query optimization [HHCF96, Hsu96, HK96, Hsu97, HK98, HK00, Kin81b, LS95, MC96, SHKC93, YSP97, YS89] A number of heuristics have been proposed that take advantage of association rules in order to optimize queries semantically [Bel96, CGM90, GGMR97, GGZ01, Kin81a, SSS92] We refer briefly to the heuristics presented by Siegel et al. [SSS92] If a relation R in the query has a restricted attribute A and an unrestricted cluster index attribute B, then look for a rule where the restriction on A implies a restriction on B. This heuristic implies the ....

J.J. King. Query optimization by semantic reasoning. PhD thesis, Stanford University, 1981.

....constraints like primary or foreign keys. Thus, Chakravarthy et al. Chakravarthy, Grant, Minker 1990) have defined SQO in respect of integrity constraints as to transform a query into one which is semantically equivalent to the original query, but which can be executed more efficiently. King (King 1981) argued that the costs of evaluating the transformed query plus the transformation costs should be lower than the costs of evaluating the original query. Semantic equivalence means that the transformed query has the same answer as the original query on all database states satisfying the integrity ....

....ways to use functional dependencies for SQO. But the constraints provided by a DBMS are few and often too general in the sense that they are valid in all possible database states. Another way is to provide SQO with semantic knowledge by hand which also seems no adequate technique. For example, King (King 1981) uses constraints on attribute values to optimize queries by his system QUIST. Zhang and Ozsoyoglu (Zhang Ozsoyoglu 1994) have presented also techniques for semantic query optimization which are based on implication constraints. Implication constraints are a generalization of functional ....

King, J. J. 1981. Query optimization by semantic reasoning.

....Knowledge Representation System (TKRS) Informally, semantic equivalence means that the transformed query has the same answer as the original query on all databases satisfying the IC rules. The notion of semantic query optimization for relational databases was introduced in the early 80 s by King [13, 14]; Hammer and Zdonik [10] independently developed very similar optimization methods. The key idea in [13, 14] as well as in [10] is that IC may not only be utilized to enforce consistency of a database but also to optimize user queries. During the last decade, many efforts have been made to ....

....the same answer as the original query on all databases satisfying the IC rules. The notion of semantic query optimization for relational databases was introduced in the early 80 s by King [13, 14] Hammer and Zdonik [10] independently developed very similar optimization methods. The key idea in [13, 14], as well as in [10] is that IC may not only be utilized to enforce consistency of a database but also to optimize user queries. During the last decade, many efforts have been made to improve this technique and to This research has been partially funded by the LOGIDATA project of the ....

J. J. King. Query optimization by semantic reasoning. PhD thesis, Dept. of Computer Science, Stanford University, Palo Alto, 1981.

....constraints like primary or foreign keys. However, the integrity constraints provided by the DBMS are often too general, not complete and do not reflect the special need of SQO. Another way is to provide SQO with semantic knowledge by hand which seems also no adequate technique. For example, King [King, 1981] uses such kind of constraints on attribute values in QUIST which are given by an expert. Now the arise of knowledge discovery in databases offers a new approach to solve both problems: provides SQO automatically with constraints, and extends them to constraints which are not valid in all states ....

King, J. J. (1981). Query optimization by semantic reasoning. Technical Report STAN-CS-81-857, Stanford University.

....of declarative data models and languages [Ullman 88] This is because it is often difficult to efficiently implement declarative queries. The query reformulation approach, also known as semantic query optimization approach in previous work [Chakravarthy et al. 90, Hammer and Zdonik 80, King 81, Siegel 88] addresses the problem differently from the conventional syntactical approaches [Apers et al. 83,Jarke and Koch 84] in that it brings to bear a richer set of knowledge about the contents of databases to optimize queries. The use of semantic knowledge offers more potential for cost ....

....This remaining constraint is on an indexed attribute so it is less expensive, while the retrieved data is still the same as that retrieved by the original query. In addition to adding a constraint on indexed attribute, there are many other ways to reduce the cost of the query by reformulation [King 81] For example, we can predict that the query will return NIL (empty set) The reformulation system can use the database schema to estimate the access cost of the queries and guide the search for the least expensive equivalent query. A1 A2 A3 A 1 2 B 1 2 C 0 2 Database Query Reformulation Database ....

[Article contains additional citation context not shown here]

Jonathan Jay King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Department of Computer Science, 1981.

....recursive plans [21] They noticed however that fully rigorous theorem proving might not be suited to planning, where imprecise inference is often necessary. Transformational synthesis has been used intensively in database programming, such as query processing [38] and semantic query optimization [15], where declarative query specifications are transformed into executable and efficient query plans. Freytag and Goodman applied program transformation techniques to the synthesis of iterative programs from relational query specifications [7] There has been essentially no work in the deductive ....

King, J., "Query Optimization by Semantic Reasoning"; PhD Dissertation, Technical Report STAN-CS-81-857, Department of Computer Science, Stanford University, 1981.

....Semantic query optimization is especially important in applications that require integrating multiple heterogeneous Work supported in part by BSF grant 92 00360. sources of data (e.g. CGMH 94, LSK95] The topic of semantic query optimization has been investigated in many papers (e.g. Kin81, CGM88] and is well understood for queries that can be represented as a union of conjunctive queries. For queries involving recursion, or which cannot otherwise be translated to a union of conjunctive queries (because of aggregation or duplicates) semantic query optimization has largely ....

J. J. King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Stanford, CA, 1981.

....due to the increasing complexity of queries and the heterogeneity of information sources. However, it is difficult for conventional query optimization techniques to solve all the problems for the next generation information systems. Semantic query optimization (SQO) Hammer and Zdonik, 1980, King, 1981, Siegel, 1988, Shekhar et al. 1988, Shenoy and Ozsoyoglu, 1989, Yu and Sun, 1989, Chakravarthy et al. 1990, Sun and Yu, 1994] is a promising query optimization technique that can complement conventional techniques to overcome the heterogeneity and considerably reduce query execution cost. The ....

....is less expensive than executing the original query because the system saves the time for the redundant comparisons. Though semantic query optimization is an effective and promising technique, it requires sufficient semantic knowledge to yield high cost reduction. Previous work in SQO, such as [King, 1981] , assume that the optimizer can use semantic integrity constraints given by users for the optimization. Semantic integrity constraints express rules that must be followed by the data in any state of a database. Examples of semantic integrity constraints for a hospital database are the age of a ....

[Article contains additional citation context not shown here]

Jonathan J. King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Department of Computer Science, 1981.

....for information gathering and retrieval from heterogeneous, distributed environment on the Internet. We are currently applying our approach to the problem of learning for semantic query optimization (Hsu Knoblock 1994; 1996b; Siegel 1988; Shekhar et al. 1993) Semantic query optimization (SQO) (King 1981; Hsu Knoblock 1993; Sun Yu 1994) optimizes a query by using semantic rules, such as all Maltese seaports have railroad access, to reformulate a query into a less expensive but equivalent query. For example, suppose we have a query to find all Maltese seaports with railroad access and ....

King, J. J. 1981. Query Optimization by Semantic Reasoning.

....che non possono contenere alcun oggetto della base di dati. Il prototipo presentato in questo lavoro implementa queste tecniche rendendo disponibile un software per la validazione di schemi e l ottimizzazione semantica di interrogazioni. L approccio da noi seguito, concordemente alle proposte in [6, 7, 8], e quello di applicare la conoscenza fornita dai vincoli di integrit a, espressi per imporre la consistenza di una base di dati, alle interrogazioni eseguite dall utente, trasformando un interrogazione in una equivalente che pu o essere eseguita con costi inferiori. Come interfaccia verso ....

J. J. King. Query optimization by semantic reasoning. PhD thesis, Dept. of Computer Science, Stanford University, Palo Alto, 1981.

....f( Gamma1; 1; fS1; S3g; 2; fS2g; fS3g) 3; fS1g) 1; fS2g)g: From this matrix, we can deduce for example that in the region (2; 3) the relevant information sources will be S1 and S2. 2 Related Work Our work can be viewed as a form of semantic query optimization (SQO) King, 1981; Chakravarthy, Grant, Minker, 1990; Hsu Knoblock, 1994; Levy Sagiv, 1994 ] where new subgoals are added to a query by analyzing the integrity constraints known about the data. In our context, the analogue of integrity constraints are the descriptions of the information sources. The key ....

King, J. J. 1981. Query Optimization by Semantic Reasoning. Ph.D. Dissertation, Stanford University, Department of Computer Science.

....to transform the query. Informally, semantic equivalence means that the transformed query has the same answer as the original query on all databases satisfying the integrity constraints. The notion of semantic query optimization for relational databases was introduced in the early 80 s by King [17, 18]; Hammer and Zdonik [13] independently developed very similar optimization methods. The key idea in [17, 18] as well as in [13] is that integrity constraints may not only be utilized to enforce consistency of a database but also to optimize user queries. During the last decade, many efforts have ....

....as the original query on all databases satisfying the integrity constraints. The notion of semantic query optimization for relational databases was introduced in the early 80 s by King [17, 18] Hammer and Zdonik [13] independently developed very similar optimization methods. The key idea in [17, 18], as well as in [13] is that integrity constraints may not only be utilized to enforce consistency of a database but also to optimize user queries. During the last decade, many efforts have been made to improve this technique and to generalize it to deductive databases (Shenoy and Ozsoyoglu [21, ....

J. J. King. Query optimization by semantic reasoning. PhD thesis, Dept. of Computer Science, Stanford University, Palo Alto, 1981.

....Representation System (TKRS) Informally, semantic equivalence means that the transformed query has the same answer as the original query on all databases satisfying the IC rules. The notion of semantic query optimization for relational databases was introduced in the early 80 s by King [13, 14]; Hammer and Zdonik [10] independently developed very similar optimization methods. The key idea in [13, 14] as well as in [10] is that IC may not only be utilized to enforce consistency of a database but also to optimize user queries. During the last decade, many efforts have been made to ....

....the same answer as the original query on all databases satisfying the IC rules. The notion of semantic query optimization for relational databases was introduced in the early 80 s by King [13, 14] Hammer and Zdonik [10] independently developed very similar optimization methods. The key idea in [13, 14], as well as in [10] is that IC may not only be utilized to enforce consistency of a database but also to optimize user queries. During the last decade, many efforts have been made to improve this technique and to generalize it to deductive databases [17, 18, 19, 8] More recently, some efforts ....

J. J. King. Query optimization by semantic reasoning. PhD thesis, Dept. of Computer Science, Stanford University, Palo Alto, 1981.

....518 sec. 86.5 3. Find all the languages spoken in the Pacific Rim. 3710 sec. 289 sec. 92.2 4. Find the population of the United States. 14 sec. NA NA Table 1: Experimental Results on CIA World Factbook Queries 6 Related Work Our work can be viewed as a form of semantic query optimization (SQO) [King, 1981, Chakravarthy et al. 1990] where new subqueries are added to a query by analyzing the integrity constraints known about the data. In our context, the analogue of integrity constraints are the descriptions of the information sources. One key issue in SQO is determining when the additional ....

Jonathan Jay King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Department of Computer Science, 1981.

....to reducing this cost is to search for reformulations of the query access plan that reduce the total cost. Database management systems often perform syntactic query reformulation [8] We leave that task to the respective DBMS then, and focus instead on more global semantic query reformulation [5, 9]. The idea is to transform the query resulting from the planning process into a semantically equivalent one that can be executed more efficiently. Consider the planned query illustrated in Figure 8. The final step in this query, comparing the depth and the draft, could be quite costly since the ....

Jonathan Jay King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Department of Computer Science, 1981.

....are used for predetermined cases to guide the process. A language based on lambda calculus is used to express the knowledge base and queries. The idea of semantic query optimization for relational databases in the context of Artificial Intelligence was first thoroughly investigated by King [13]. The notions of semantic equivalence between queries and merging an integrity constraint into query are formally defined. However, the constraints are on values only, and dependencies are not used. The transformation process, data directed in nature, is guided by heuristics based on knowledge of ....

J. J. King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Palo Alto, CA, 1981.

....of inductive learning are usually limited to data classifiers. In this paper, we present an approach in which inductively learned knowledge is used for semantic query optimization to speed up query answering for data knowledge based systems. The principle of semantic query optimization (King 1981) is to use semantic rules, such as all Tunisian seaports have railroad access, to reformulate a query into a less expensive but equivalent query, so as to reduce the query evaluation cost. For example, suppose we have a query to find all Tunisian seaports with railroad access and 2,000,000 ft 3 ....

....access and 2,000,000 ft 3 of storage space. From the rule given above, we can reformulate the query so that there is no need to check the railroad access of seaports, which may save some execution time. Many algorithms for semantic query optimization have been developed (Hsu Knoblock 1993; King 1981; Shekhar, Srivastava, Dutta 1988; Shenoy Ozsoyoglu 1989) Average speedup ratios from 20 to 40 percent using hand coded knowledge are reported in the literature. This approach to query optimization has gained increasing attention recently because it is applicable to almost all existing ....

[Article contains additional citation context not shown here]

King, J. J. (1981). Query Optimization by Semantic Reasoning.

.... this issue, researchers have proposed interleaving query planning and execution so that the query processor can use intermediate data to refine the part of the query plan that has not been completely executed [9, 12, 13] A relatively unexplored area is the use of semantic query optimization (SQO) [14, 15, 16, 17, 18, 19, 20, 21] for multi source query plan optimization. The advantage of SQO is that the optimizer can infer the information about intermediate data from semantic knowledge prepared prior to query execution time. Another reason is that SQO supports the extensibility of multidatabase systems because it ....

....query Q1 in Table 2 and the semantic knowledge given in Table 3. Some of the equivalent queries of Q1 inferred from the semantic rules are shown in Table 5. The optimizer deletes literal 3 on the variable status based on R1 and generates Q1.1. This is an example of constraint elimination [15] reformulation. The optimizer can add year built 1945 to Q1 from R3 and yield another equivalent query Q1.2. Adding new literals could be useful in many situations. One of them is when the added literal exploits an indexed attribute. The optimizer can also refute a query, when it infers that ....

[Article contains additional citation context not shown here]

J. J. King, Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Department of Computer Science, 1981.

....to make decisions to guide rule construction, guide rule repair, and control the size of a rule set. This paper also briefly reviews how robustness can be efficiently estimated and reports the initial experimental results. Learning for Semantic Query Optimization Semantic query optimization (SQO) (King 1981) is a promising query optimization technique for intelligent information mediators (Wiederhold 1992; Arens et al. 1993; Levy, Srivastava, Kirk 1995) that integrate heterogeneous information sources because it can complement conventional query optimization techniques to overcome the ....

King, J. J. 1981. Query Optimization by Semantic Reasoning.

....to reducing this cost is to search for reformulations of the query access plan that reduce it. Database management systems (DBMSs) often perform syntactic query reformulation [14] We leave that task to the respective DBMS then, and focus instead on more global semantic query reformulation [7, 15]. The idea is to transform the query resulting from the planning process into a semantically equivalent one that can be executed more efficiently. Consider the planned query illustrated in Figure 10. The final step in this query, comparing two geographic location codes geocode and geocode2, ....

....to search for the least expensive query from the space of semantically equivalent queries to the original one. Two queries are defined to be semantically equivalent[25] if they return identical answers given the same contents of the database. The alternative definition of semantic equivalence[15] requires that the queries return identical answers given any contents of the database, but this definition would limit us to using only semantic integrity constraints which are often not available. The use of the less restrictive definition of semantic integrity requires that the system updates ....

[Article contains additional citation context not shown here]

Jonathan Jay King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Department of Computer Science, 1981.

....Given a multidatabase query, the planner of SIMS generates a partially ordered query plan to retrieve the data. The reformulation algorithm presented here is used to reformulate this initial query plan to reduce the cost of retrieval. The query reformulation approach was initially proposed by (King, 1981) and (Hammer and Zdonik, 1980) Our approach differs from theirs and the following related work (Siegel, 1988; Chakravarthy, Grant and Minker, 1990) in that we do not rely on heuristics to guide the search in a hill climbing manner, which often results in local optima. Moreover, we consider ....

....form A B from the database. The algorithm retrieves the data that satisfy the condition A, then compiles the data for the conclusions B. 6 Related Work The semantic query optimization approach has been studied extensively in previous work (Chakravarthy, Grant, and Minker, 1990; Siegel, 1988; King, 1981; Hammer and Zdonik, 1980) These systems demonstrate the benefit of using knowledge of database contents to optimize queries. The most significant difference between our approach and theirs is that they rely on heuristics, and search for the optimal equivalent query in a hill climbing manner, ....

King, J.J., 1981 Query Optimization by Semantic Reasoning.

....addition, we have integrated the planning with the execution system [ Knoblock, 1995 ] which allows the system to dynamically replan parts of a query that fail while continuing to execute the other subqueries of the overall plan. The semantic approach to query optimization was first developed by King [ King, 1981 ] and has since been extended in a number of systems [ Adam et al. 1993; Shenoy and Ozsoyoglu, 1989; Shekhar et al. 1988; Siegel, 1988 ] Our approach to this problem differs from other related work in that we do not rely on explicit heuristics of the database implementation to guide search ....

Jonathan Jay King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Department of Computer Science, 1981.

....and repair so that the resulting rules are robust and require a minimal maintenance cost. Previously, we have applied the robustness estimation approach to rule discovery for semantic query optimization [ Hsu and Knoblock, 1994, Hsu and Knoblock, 1996b ] Semantic query optimization (SQO) King, 1981, Hsu and Knoblock, 1993, Sun and Yu, 1994 ] optimizes a query by using semantic rules, such as all Maltese seaports have railroad access, to reformulate a query into a less expensive but equivalent query. For example, suppose we have a query to find all Maltese seaports with railroad access and ....

Jonathan J. King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Department of Computer Science, 1981.

No context found.

Jonathan Jay King. Query Optimization by Semantic Reasoning. PhD thesis, Stanford University, Department of Computer Science, 1981.

No context found.

J. J. King. Query optimization by semantic reasoning. PhD thesis, Dept. of Computer Science, Stanford University, Palo Alto, 1981.

First 50 documents

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