Project Description

Project Objectives

The objectives of Agent Academy are:

• To develop an integrated framework for the systematic study of Intelligent Agent (IA) attributes and especially agent intelligence;
• To exploit Data Mining (DM) techniques for improving the decision making process in Intelligent Agents;
• To develop new and refine existing DM techniques for IA use and behavior;
• To develop the tools for assembling and maintaining a large repository of data on agent use and behavior;
• To create a facility that will enable the incorporation of agents into more business applications;
• To empower enterprise solutions based on Intelligent Agents applications by improving the quality of provided services;

The successful outcome of this effort is expected to propagate the use of agent related technologies into both business practices and personal use.

Functional Description of the Agent Academy

Agent Academy forms an integrated framework that receives input from its users and the Web. For demonstrative purposes of inter-system interoperability we specify a functional description of Agent Academy structure to show the communication of internal components. The main block diagram of AA is illustrated in Figure 1. AA operates as a multi agent system, which can train new agents or retrain its own agents in a recursive mode. A user issues a request for a new agent as a set of functional specifications. The Agent Factory, a module responsible for selecting the most appropriate agent type and supplying the base code for it, handles the request. A newly created untrained agent (UA) comprises a minimal degree of intelligence, defined by the software designer. This agent enters the Agent-Training Module, where its world perception increases substantially during a virtual interactive session with an agent master (AM).  Based on the encapsulated knowledge, acquired in the knowledge extraction phase, an AM can take part in long agent-to-agent (A2A) transactions with the UA. This process may include modifications in the agent’s decision path traversal and application of augmented adaptivity in real transaction environments.

The core of the agent academy is the Agent Use Repository (AUR), which is a collection of statistical data on prior agent behavior and experience. It is on the contents of AUR, where data mining techniques, such as extraction of association rules for the decision making process, are applied in order to augment the intelligence of the AM in the training module. Building AUR will be a continuous process performed by a large number of mobile agents and controlled by the Data Acquisition Module. A large part of an agent’s intelligence handles the knowledge acquired by the agent since the beginning of its social life through the interaction with the environment it acts upon.

Figure 1. The Agent Academy and its environment.

Required Technical Tasks

Our effort will be focused on the following tasks:
1. Integrating methods for agent intelligence
2. Evaluation of promising but not yet widely applied methods from AI (fuzzy systems, intelligent knowledge representation, non-symbolic machine learning, etc.)
3. Development of innovative techniques for adaptive decision making
4. Improvement of agent interoperability based on existing specifications (FIPA, OMG)
5. Adoption of well-tested methods for agent communication (FIPA ACL, KQML)
6. Application of existing standards on Data Warehousing (OMG's XMI, CWM MOF)
7. Development of behaviour tracking tools for agents trained by AA
8. Development of tools for reporting agent activities in remote web servers.