Scalable semantic brokering over dynamic heterogeneous data sources in InfoSleuth

Abstract

InfoSleuth 1 is an agent-based system for information discovery and retrieval in a dynamic, open environment. This paper discusses InfoSleuth's multibroker design and implementation. InfoSleuth's brokering function combines reasoning over both the syntax and semantics of agents in the domain. The broker must reason over explicitly advertised information about agent capabilities to determine which agent can best provide the requested services. Brokering in InfoSleuth is a match-making process, recommending agents that provide services to agents requesting services. Robustness and scalability issues dictate that brokering must be distributable across collaborating processes. Our multibroker design is a peer-to-peer system that requires brokers to advertise to and receive advertisements from other brokers. Brokers collaborate during matchmaking to give a collective response to requests initiated by non-broker agents. This results in a robust, scalable brokering system.

DR-BROKERING: A semantic brokering system

“…InfoSleuth [31] is an agent-based information discovery and retrieval system that adopts "broker agents" to perform the syntactic and semantic matchmaking. The broker uses a rule-based reasoning engine, implemented in LDL, to determine which agents have advertised services that match those requested in the query.…”

A knowledge-based approach for selecting information sources

“…InfoSleuth (Bayardo et al 1997;Fowler et al 1999;Nodine et al 2003), which has its roots in Carnot, is an agent-based system for information discovery and retrieval in a dynamic, open environment, broadening the focus of database research to the challenge of the World-Wide Web. It extends the capabilities of Carnot to an environment in which the identities of the information sources need not be known at the time of generating the mapping.…”

Dynamic Service Reconfiguration and Enactment Using an Open Matching Architecture

“…Some of the earliest systems, enabled by the development of KIF (Knowledge Interchange Format) (Genesereth and Fikes, 1992) and KQML (Knowledge Query and Manipulation Language) (Finin et al, 1994), are SHADE (Kuokka and Harada, 1996) operating over logic-based and structured-text languages and the complementary COINS (Kuokka and Harada, 1996) that operates over free-text using well-known termfirst index-first information retrieval techniques. Subsequent developments such as InfoSleuth (Nodine et al, 1999) apply reasoning technology to the advertised syntax and semantics of a service description, while the RETSINA system (Sycara et al, 2003) has its own specialised language influenced by DAML-S and uses a belief-weighted associative network representation of the relationships between ontological concepts as a central element of the matching process. While technically sophisticated, a particular problem with the latter is how to make the initial assignment of weights without biasing the system inappropriately.…”