International audienceAssociation rules mining has been largely studied by the data mining community. ARM aims to extract the interesting rules from any given transactional database. This problem is well known to be time consuming in general. This paper deals with association rules mining algorithms to cope with very large databases and especially for those existing on the web. Many polynomial exact algorithms already proposed in literature have shown their efficiency when dealing with small and medium datasets. Unfortunately, their efficiency is not enough for handling the huge amount of data in the web context requiring a real time response. Not surprisingly, some bio-inspired methods seem to be clearly more appropriate. This paper mainly proposes a new ARM algorithm based on an improved version of bees swarm optimisation with three different heuristics for exploring the search area. This approach has been implemented and experimented on different dataset benchmarks with small size, medium size and large size. These first empirical results highlighted that our approach outperforms some other existing algorithms both in terms of fitness criterion and CPU time

No abstract

This paper proposes a novel network architecture based on content routing and dissemination techniques, expressly adapted for hybrid mobile and vehicular ad-hoc networks. Hybrid vehicular ad-hoc networks (VANETs) are a subclass of mobile ad-hoc networks (MANETs) where each node is a moving vehicle equipped with one or more communication devices, potentially having access to several different physical communication channels. This results in a partly decentralized, self-organizing mobile radio network with a fast changing topology. Applications designed for VANETs still require efficient mapping techniques to associate content to the host serving it, as well as centralized databases associating vehicles and mobile nodes to a persistent and unique identifier. They also assume that the content providers are located in the infrastructure, but nodes and vehicles can be equipped with sensors and as such produce information. This means that dissemination and retrieval algorithms have to take into account the distribution of the consumers, providers and forwarder all over the network.We propose an architecture based on Van Jacobson's Content Centric Networking, extended and modified to solve some of the issues related to content access and dissemination across hybrid VANETs. The implementation and simulation on a realistic platform are presented as well as an application to a real case scenario, demonstrating the efficiency of the approach.

No abstract