Eric.N. Hanson, Gator: A discrimination network suitable for optimizing production rule matching, Technical Report CIS-TR-007-93, University of Florida, CIS Dept. (1993).  Home/Search   Document Not in Database   Summary   Related Articles   Check

....directly in their application environments. Furthermore, it can continuously disambiguate fuzzy information with its evolution capability. The major advantage of our framework is that it can support an inexact match without modifying the discrimination network such as rete [4] TREAT [15] or gator [6]. That is mainly due to the minimal semantic gap between our framework and the rule based languages. Such a minimal semantic gap makes it possible for the framework to be implemented on top of the languages. In other words, most of its semantics can be expressed in terms of OPS5 or ECLPS through ....

Eric.N. Hanson, Gator: A discrimination network suitable for optimizing production rule matching, Technical Report CIS-TR-007-93, University of Florida, CIS Dept. (1993).

....This example gives the flavor of how active database rule condition matching is performed by propagating information through a Gator network. A detailed description of how Gator does pattern matching by propagating tokens is not given here due to lack of space, but is presented elsewhere [5]. 3 Cost Functions So the optimizer can choose between different Gator networks, a cost estimation function is needed to evaluate the expected cost of a network. Our cost function estimates the run time cost associated with propagating tokens through the network, including the cost of: ffl ....

Eric N. Hanson. Gator: A discrimination network suitable for optimizing production rule matching. Technical Report TR-007-93, University of Florida CIS Dept., February 1993. http://www.cis.ufl.edu/cis/tech-reports/.

....way across the three nodes fi1, ff4 and ff5, then that combination would be packaged as one token and placed in the P node, triggering the rule for the new combination. The structure, pattern matching algorithms, and cost evaluation functions for Gator networks are discussed in detail elsewhere [7, 8, 17]. We have implemented Gator in Ariel and are currently testing and evaluating the new Ariel system [17] This paper presents results from a simulation conducted to test and evaluate strategies for optimizing Gator networks in an active database environment. The results of this simulation lead us ....

....and produce a lower cost network than the DP, which makes it the obvious candidate for generating the discrimination networks for larger rules. This result lead to the selection of 2PO as the Gator network optimizer for a new version of the Ariel active DBMS [17] The nature of the cost functions [7] is such that the cost of a node is a function of several parameters of its children, which include cost, update frequency and cardinality. Therefore, the minimum cost heuristic, which is taken as the sole criterion for determining the feasibility of a network, may have been too restrictive. The ....

Eric N. Hanson. Gator: A discrimination network suitable for optimizing production rule matching. Technical Report TR-007-93, University of Florida CIS Dept., February 1993. http://www.cis.ufl.edu:80/cis/tech-reports/.

....covers handling of tokens and Gamma tokens. The algorithms for processing tokens are described here in a set oriented style that is suitable for use in active database systems. A tuple at a time, recursive style of the algorithm suitable for main memory production systems is presented elsewhere [7]. In the following discussion, memory nodes of type ff and fi will be referred to together as memory nodes. Nodes that can have multiple inputs, including fi memories and P nodes, will be called multiple input nodes. The term node may be used to describe an ff memory, fi memory, or P node. 4.1 ....

....they are deleted. In turn, more Gamma tokens are generated and passed to the successor of the multiple input node. Detailed algorithms for Gamma tokens are not presented. A discussion of how Gator supports negated condition elements like those that can be specified in OPS5 [2] is given elsewhere [7]. To perform well, the Gator rule condition matching strategy needs a Gator network with a good structure. How to find an efficient network is the subject of the next section. 5 Optimization of Gator networks The key parts of an optimization strategy for building a good Gator network are ....

[Article contains additional citation context not shown here]

Eric N. Hanson. Gator: A discrimination network suitable for optimizing production rule matching. Technical Report CIS-TR-007-93, University of Florida CIS Dept., February 1993. Available by anonymous ftp from ftp.cis.ufl.edu. See README file under /cis/techreports.

....algorithms for processing tokens are described here in a set oriented style that is suitable for use in active database systems. A tuple at a time, recursive style of the algorithm that is similar to how Rete and TREAT are typically implemented for main memory production systems is presented in [5]. The following terminology will be used. Memory nodes of type ff and fi will be referred to together as memory nodes. Nodes that can have multiple inputs, including fi memories and P nodes, will be called multiple input nodes. The term node may be used to describe an ff memory, fi memory, or ....

Eric N. Hanson. Gator: A discrimination network suitable for optimizing production rule matching. Technical Report CIS-TR-007-93, University of Florida CIS Dept., February 1993.

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