Web services classification for disaster management and risk reduction

Abstract

A disaster is a disruption of the society functioning that can interrupt essential services of our live. It has an important impact on human, material, economic and environment. There a several kind of disaster such as: natural, environmental emergencies and contagious disease that affects health and so on. We need serious and important resources to reduce risk that can be caused by these disasters. So it is important to establish good programs and classify the activities or services that should be launched to handle disasters. Modern technology can be effective in reducing the damage and risk caused by disasters, particularly the use of Web services in disaster management. To this end, the classification of Web services by domain can be very useful to facilitate the services invocation in the event of an emergency or disaster by the concerned authorities. In this paper, we present an approach that combines both a supervised learning method Naïve Bayes and the meta-heuristic of stochastic Local search (SLS) for services classification. SLS is used for attribute selection which reduces the space of attributes. The latter are sent to Naïve Bayes classifier to build models. To evaluate and measure the performance of our approach we used a set of 364 Web services divided into four categories (QWS Dataset). The experiment gives good results compared to other previous works.