Multi-Agent Framework for Intelligent Questionnaires on the Web

This paper presents a multi-agent framework for conducting an intelligent survey on the Internet network through Web capabilities. The advances in multi-agent technology and network computing form the basis of this framework. Performing questionnaires on the Internet has the drawback of network traffic, which, however, is more than offset by its advantages, including distribution, communication, platform independence. We have developed an agent-based framework to solve the questionnaire problem that increases reusability and understanding. Each stage of the questionnaire is assigned to an autonomous agent working alone or with other agents forming a society. The underlying knowledge of questionnaires and decision rules is transformed into objects belonging to an agent or an agent society. Agents can interact with each other through the interaction process, providing a dynamic, interactive and evolving system. All these components are located in an environment. Finally, an application using this framework, the ALBOR system, is described. The ALBOR system (Barrier Free Access to the Computer) was conceived to publish intelligent questionnaires on the Internet evaluating the physical, sight, hearing and cognitive aptitudes of computer users with disabilities and give advice about technical aids or adaptations depending on their evaluated capabilities.