An Agent-Based Approach to Web Site Maintenance

Abstract

Abstract. Web sites are public representations of corporations, businesses and governmental bodies. As such they require proper maintenance to ensure that accurate and updated information is presented adequately. Dynamically created Web sites sometimes are not an option: they add a computational overhead on the server and make the automatic indexing of pages difficult. Static Web sites may grow to an extent that manually performing maintenance operations becomes unfeasible: automatic or semi-automatic means ought to be put in place. In this paper we explain one such approach, using software agents to look after large data-intensive Web sites. Our maintenance operations are performed by a team of autonomous agents that communicate with each other as well as with Web masters or other human agents. Our approach can be naturally incorporated into existing Web sites and its use can be gradually extended to encompass larger portions of the site. Because our agents are independent, their individual malfunctioning should not stop the maintenance effort as a whole.

A comparative evaluation of maintainability: a study of engineering department's Web site maintainability

“…However, there are several areas for further study -for example, we have not compared the professionalism of web updates, the accuracy of the models, and the satisfaction of the users; we would like to extend the qualitative analysis to quantitative analysis; and we could extend this approach to evaluate the maintainability of the processes used in the industry [4]. However, we believe that the process of maintainability analysis and evaluation adopted in this paper is practical and useful to stakeholders involved in the maintenance of web-based systems both in the industry and academia.…”

Research Developments in Web Information Retrieval from Two Brazilian Universities

“…One of the learning methods for producing ranking results using GP is presented in [16]. Another interesting work developed in the UFMG group proposes a new strategy for learning ranking functions by using association rules [28].…”