Adaptive support framework for wisdom web of things

Wisdom Web of Things (W2T) is the next generation of networks, which provides ubiquitous wisdom services in a ubiquitous network in the hyper world. Adaptiveness is the key issue of realizing the harmonious unity of human-information-thing. This paper proposes a self-adaptive support framework for W2T, which has three important components: (i) An adaptive requirement description language, which is to describe the wisdom service models and self-adaptive wisdom service strategies. (ii) Forward reasoning and backward planning ability. We propose that forward reasoning can be implemented based on the Rete algorithm and backward planning can be implemented based on a Hierarchical Task Network (HTN), which enable W2T to achieve complex, rapid, and efficient reasoning and planning to provide active, transparent, safe, and reliable services. (iii) A knowledge base evolution mechanism based on a learning classifier system, which is to realize the evolution of the knowledge base, and to satisfy the dynamic requirements of wisdom services. We take a wisdom traffic system as an example to demonstrate the data conversion mechanism and the functions of the proposed self-adaptive support framework.