Towards Distributed Association Rule Mining Privacy

With the advancement of storage, retrieval, and network technologies today, the amount of information available to each organization is literally exploding. Although it is widely recognized that the value of data as an organizational asset often becomes a liability because of the cost to acquire and manage those data is far more than the value that is derived from it. Thus, the success of modern organizations not only relies on their capability to acquire and manage their data but their efficiency to derive useful actionable knowledge from it. To explore and analyze large data repositories and discover useful actionable knowledge from them, modern organizations have used a technique known as data mining, which analyzes voluminous digital data and discovers hidden but useful patterns from such massive digital data. However, discovery of hidden patterns has statistical meaning and may often disclose some sensitive information. As a result, privacy becomes one of the prime concerns in the data-mining research community. Since distributed data mining discovers rules by combining local models from various distributed sites, breaching data privacy happens more often than it does in centralized environments.

Challenges of the "Global Understanding Environment" Based on Agent Mobility

Among traditional users of Web resources, industry also has a growing set of smart industrial devices with embedded intelligence. Just as humans do, smart industrial devices need online services—for example, for condition monitoring, remote diagnostics, maintenance, and so on. In this chapter, we present one possible implementation framework for such Web services. Assume that such services should be Semantic-Web-enabled and form a service network based on internal and external agents’ platforms, which can host heterogeneous mobile agents and coordinate them to perform needed tasks. The concept of a “mobile-service component” assumes not only the exchange of queries and service responses but also the delivery and composition of a service provider itself. A mobile-service component carrier (agent) can move to a field device’s local environment (embedded agent platform) and perform its activities locally. Service components improve their performance through online learning and communication with other components. Heterogeneous service components’ discovery is based on semantic P2P search. In this chapter, we discuss the Global Understanding Environment as an enabling agent-driven semantic platform for implementation of the heterogeneous industrial resources, condition monitoring, and predictive maintenance. The main challenges of such an environment are presented, which are “semantic adapters” for industrial objects, diagnostic models, exchange and integration, distributed trust management, and the concept of a human as a Web service.

Mail Server Management with Intelligent Agents

Amidst the era of e-economy, one of the difficulties from the standpoint of the information-systems manger is, among others, the forecast of memory needs for the organization. In particular, the manager is often confronted with maintaining a certain threshold amount of memory for a prolonged period. However, this constraint requires more than technical and managerial resolutions, encompassing knowledge management for the group, eliciting tacit knowledge from the end users, and pattern- and time-series analyses of utilization for various applications. This chapter summarizes current methods for managing server memory by incorporating intelligent agents. In particular, a new framework for building a set of automated intelligent agents with a neural network is proposed under the client-server architecture. The emphasis is on collecting the needs of the organization and acquiring the application-usage patterns for each client involved in real time. Considerations for future work associated with technical matters comprising platform independence, portability, and modularity are discussed.

A Generic Internet Trading Framework for Online Auctions

In this chapter, we introduce a generic Internet trading framework for online auctions. We present the requirements and service of the framework. A generic OR/XOR bidding language that can express different OR/XOR combinations is adopted for Web interfaces. The framework is implemented with free open-source technologies already successfully tested in industries. Researchers can use the platform to implement different electronic-market mechanisms, simulate the market behavior of their interests, and experiment with it. We also provide future directions for the framework design.

ADAM

In recent years, the emergence of the Internet has resulted in a proliferation of data. This in turn has given rise to increasing demands of organizations to access accurate information swiftly and efficiently. Thus, the scope of functions for databases has expanded more than ever and the complexity of database systems has grown accordingly. Consequently, the burden on database administrators (DBAs) has increased significantly. The objective of this research is to address and propose a solution to overcome this problem of overburdened and expensive DBAs. This chapter focuses on relational database management systems in particular and proposes a novel and innovative multiagent system (MAS) that would autonomously and rationally administer and maintain databases. The proposed multi-agent system tool, ADAM (a MAS for autonomous database administration and maintenance), is in the form of a self-administering wrapper around database systems and it addresses, and offers a solution to, the problem of overburdened and expensive DBAs with the objective of making databases a cost-effective option for small/medium-sized organizations. An implementation of the agent-based system to proactively or reactively identify and resolve a small subset of DBA tasks is discussed and the GAIA methodology is used to outline the detailed analysis and design of the same. Role models describing the responsibilities, permissions, activities, and protocols of the candidate agents, and interaction models representing the links between the roles are explained. The coordinated intelligent rational agent model is used to describe the agent architecture and a brief description of the functionalities, responsibilities, and components of each agent type in the ADAM multiagent system is presented. Finally, a prototype system implementation using JADE 2.5 and Oracle 8.1.7 is presented as evidence of the feasibility of the proposed agent-based solution for the autonomous administration and maintenance of relational databases.

An Intelligent Multi-Robot System Using Higher-Order Mobile Agents

This chapter presents a framework for controlling intelligent robots connected by communication networks. This framework provides novel methods to control coordinated systems using higher-order mobile agents. Higher-order mobile agents are hierarchically structured agents that can contain other mobile agents. By using higher-order mobile agents, intelligent robots in action can acquire new functionalities dynamically as well as exchange their roles with other colleague robots. The higher-order property of the mobile agents enables them to be organized hierarchically and dynamically. In addition to the advantages described above, higher-order mobile agents require minimum communication. They only need connection to be established when they perform migration.

Combating Malaria

In this chapter, we investigate how complexity theory and more particularly how agent-based modeling and simulation can benefit the explanation of the impact of education on malaria health care in Haiti. Our model includes: (1) the environment, encompassing mainly cities, roads, hospitals and schools; (2) the agents, modeling the human actors, who can be safe or infected by malaria disease according to their location in the environment; and (3) a modelled agent can also be mobile or not, can reproduce, and can die. We run four kinds of experiments over a 50-year period each. Our main emerging results are growing total agent, susceptible, and immune populations in a “cyclic” fluctuation form. Furthermore, we confirm the positive impact of both education and hospitals in combating malaria disease.

Approaches and Tools to Optimize and Manage Clinical Processes

There are several continuing challenges within the health-care domain. On the one hand, there is a greater need for individualized, patient-oriented processes in diagnostics, therapy, nursing, and administration; and on the other hand, hospitals have extremely distributed decision processes and strong local (individual) autonomy with a high degree of situational dynamics. This research focuses on how to optimize processes in hospitals by using agent-based simulation and agent-based information systems that can substantially increase the efficiency of hospital process management. In particular, we describe a methodology for optimization; the ingredients of the simulation and the architecture of a management system; and the integration of existing FIPA, DICOM, and HL7 standards. We discuss example scenarios to show how simulation can improve distributed scheduling and how agent-based systems can be used to manage “clinical trials.” This research is part of the German Priority Research Program (SPP) 1083 “Intelligent Agents and their application in business scenarios.”

Information Parallax

To effectively use and exchange information among AI systems, a formal specification of the representation of their shared domain of discourse—called an ontology—is indispensable. In this chapter we introduce a special kind of knowledge representation based on a dual view on the universe of discourse and show how it can be used in human activities such as searching, in-depth exploration and browsing. After a formal definition of dualistic ontologies we exemplify this definition with three different (well known) kinds of ontologies, based on the vector model, on formal concept analysis and on fuzzy logic respectively. The vector model leads to concepts derived by latent semantic indexing using the singular value decomposition. Both the set model and the fuzzy-set model lead to formal concept analysis, in which the fuzzy-set model is equipped with a parameter that controls the fine-graining of the resulting concepts. We discuss the relation between the resulting systems of concepts. Finally, we demonstrate the use of this theory by introducing the dual search engine. We show how this search engine can be employed to support the human activities addressed above.

Building Sound Semantic Web Frameworks for Scalable and Fault-Tolerant Systems

This chapter introduces hyperservices as a unified application model for Semantic Web frameworks and proposes the WASP model as a framework for implementing them. Hyperservices are based on agent societies, provided with structured information by the Semantic Web, and using Web services as a collaboration and communication interface. The WASP model adds personalization rules to modify the agents’ perception and the HIVE architecture as Semantic Information Server infrastructure within this framework. Finally, conceptual model driven software development is proposed as a means of easy adoption to hyperservices.