Charles K. Nicholas

Integrating Distributed Information Sources with CARROT II

We describe CARROT II (C2), an agent-based architecture for distributed information retrieval and document collection management. C2 can consist of an arbitrary number of agents, distributed across a variety of platforms and locations. C2 agents provide search services over local document collections or information sources. They advertise content-derived metadata that describes their local document store. This metadata is sent to other C2 agents which agree to act as brokers for that collection, and every agent in the system has the ability to serve as such a broker. A query can be sent to any C2 agent, which can decide to answer the query itself from its local collection, or to send the query on to other agents whose metadata indicate that they would be able to answer the query, or send the query on further. Search results from multiple agents are merged and returned to the user. C2 differs from similar systems in that metadata takes the form of an automatically generated, unstructured feature vector, and that any agent in the system can act as a broker, so there is no centralized control. We present experimental results of retrieval performance and effectiveness in a distributed environment.

Agents Making Sense of the Semantic Web

Effective use of the vast quantity of available information and services on the Internet will require multi-agent systems to be tightly integrated with existing web infrastructure. This however will be impossible unless the information on the web is presented in a semantic language, such as the DARPA Agent Markup Language (DAML), which is one aim of the “Semantic Web”. As part of our exploration of Semantic Web technology, and DAML in particular, we have constructed ITTALKS, a web-based system for automatic and intelligent notification of information technology talks. In this paper, we describe the ITTALKS system, and discuss the numerous ways in which the use of Semantic Web concepts and DAML extend its ability to provide an intelligent online service to both the human community and, more interestingly, the agents assisting them.

Sampling Methods for Building Initial Partitions

The initialization of iterative clustering algorithms is a difficult yet important problem in the practice of data mining. In this chapter, we discuss two new approaches for building such initial partitions. The first approach applies a procedure for selecting appropriate samples in the spirit of the Cross-Entropy (CE) method, and the second is based on a sequential summarizing schema. In the first approach, we use a sequential sample clustering procedure instead of the simulation step of the CE method. In this context, we state several facts related to the Projection Pursuit methodology for exploring the structure of a high-dimensional data set. In addition we review several external and internal approaches for cluster validity testing. Experimental results for cluster initializations obtained via the CE method and the first of the presented methods are reported for a real data set.