A DISTRIBUTED AGENT-BASED ARCHITECTURE FOR CUSTOMER RELATIONSHIP MANAGEMENT IN E-ENTERPRISES

Abstract

The chief objective of Customer Relationship Management (CRM) is to acquire, retain and increase the profitability and lifetime value of customers. This requires a better understanding of customer needs and expectations. Customer information acquired through transactions with the customer is stored in databases which contain valuable information about customers which could be used to business advantage. Typically, databases in organizations are distributed and as such, the data mining tool for such databases should have a distributed architecture. In this paper, we propose the design and implementation issues of an agent based architecture for data mining of such databases where individual agents communicate using KQML. We have defined certain performatives that have been implemented for this application. In this paper we present the design and implementation of the facilitator agent, the broker, the data mining agents and implementation of the performatives that we have defined for this application. Our implementation is based on Java servelets with KQML as the language used for agent communication. Sample runs on known algorithms like apriori are performed and the runs demonstrate the proof of concept for distributed agents based model.

Keywords

• Software agents
• data mining
• customer relationship management
• distributed systems
• UML