Ant System in Web

The process of surfing in Web is similar to the ant colonies foraging. Ants are social insects that work more to the survival of the colony as a whole, than to a single individual component of the colony. Their physical characteristics do not allow them to have a global vision of the environment, but an important and interesting behavior of them is their foraging behavior, and in particular, how they can find shortest paths between food sources and their nest. With this insight, we see users visiting the Web as a metaphor to the ant colonies foraging process. Sometimes, they may lose their way in the immense cyberspace, without knowing where the information sources are.

The AntWeb has been modified in two generations with following three main utilities: (see our publications below in PDF files). The first one is to use it to evaluate the structure of Web sites, and the second one is to extend web sites to have an adaptive capacity. When using for evaluation, AntWeb is able to simulate the sequence followed by visitors while visiting the Web site, allowing measuring the efficiency of the web structure.

AntWeb as an adaptive Web system works as a metaphor of ant’s foraging behavior in the following way. When the user visits a Web site, the system records some information of his route as a pheromone trail left by ants. Other users with common objective may be attracted by the pheromone trail. The system is developed with a simple ant-searching algorithm, not altering the flexibility of Web pages as it only marks the links. Thus, using AntWeb the user has not to give any additional information, what makes harder the usability of many adaptive Web systems.

And the third one is to extend the Website to have an adaptive capacity, which extracts information from the user’s sequential path to change link strengths or create new links.
 

Weigang, L., Marcos Vinícius Pinheiro Dib, Wesley Martins Teles, Vlaudemir Morais de Andrade, Alba Cristina Magalhães Alves de Melo, Judas Tadeu Cariolano, "Using ants’ behavior based simulation model AntWeb to improve website organization", in Proc. SPIE's Aerospace/Defense Sensing and Controls Symposium: Data Mining and Knowledge Discovery: Theory, Tools, and Technology IV,  Editor: Belur V. Dasarathy, Vol. 4730, pp. 229– 241, Orlando, USA, April 2002.

Wesley Martins Teles, Li Weigang, and Célia Ghedini Ralha, "AntWeb - The Adaptive Web Server Based on the Ants' Behavior", in The 2003 IEEE Web Intelligence Consortium (WIC), Halifax, Canada, 2003.

Wesley Martins Teles, Li Weigang, and Célia Ghedini Ralha, "Uma heurística para guiar os usuários da Internet baseada no comportamento da formiga", IV ENIA-SBC, Campinas, Brazil, 2003 (ppt).

Weigang, L., Wu, M.
Web Search Based on Ant Behavior: Approach and Implementation in Case of Interlegis In: The 17th International Conference on Software Engineering & Knowledge Engineering (SEKE 2005), 2005, Taibei. The Proceedins of SEKE 2005. USA: Knowledge Systems Institute Graduate Schoole, 2005. v.1. p.572 - 577

Weigang, L., Wu ManQi, “AntWeb - Web search based on ant behavior: approach and implementation in case of Interlegis”,  to appear in the book of Emerging Technologies of Text Mining: Techniques and Applications, Hércules A. Prado and Edilson Ferneda (Ed.), Idea Group Inc., Hershey, USA, 2007.

Li Weigang

Wesley Martins Teles

Célia Ghedini Ralha

Judas Tadeu Cariolano

Wu Manqi

University of Brasilia - UnB
Department of Computer Science - CIC
Caixa Postal 4466
70919-970 Brasilia - DF
Brazil 
Tel: (061) 307.2702 - 215; Fax: (061) 273.3589

Initial publication : 09-15-95; Last updated: 29-12-2006.