Mobile software agent enabled e-commerce: System design and profit maximizing algorithms

With the emergence of the Internet, e-commerce has become an attractive means of conducting business. Mobile agents consist of code and data that can move from one computer to another over the Internet, and offer a suitable paradigm for implementing e-commerce processes online. In this dissertation research, we describe a Java mobile agent based system that we have developed, called MAgNET (Mobile Agents for Networked Electronic Trading), for comparison shopping over the Internet. In MAgNET, an online buyer creates a shopping agent that visits different seller sites on the Internet and returns with the best offer that it can find. MAgNET can also be used by online sellers to deploy sales agents that proactively approach buyers to sell their items. For MAgNET we have developed algorithms that improve the sellers' profits by dynamically adjusting the price of the items being sold. These algorithms increase an online seller's profits by 16–20% as compared to earlier dynamic pricing algorithms. Finally, we present protocols that we have developed for the security and reliability of the agents and agent-servers in MAgNET.