A Hybrid Rule Based Fuzzy-Neural Expert System For Passive Network Monitoring (1304.7843v1)

Abstract: An enhanced approach for network monitoring is to create a network monitoring tool that has artificial intelligence characteristics. There are a number of approaches available. One such approach is by the use of a combination of rule based, fuzzy logic and neural networks to create a hybrid ANFIS system. Such system will have a dual knowledge database approach. One containing membership function values to compare to and do deductive reasoning and another database with rules deductively formulated by an expert (a network administrator). The knowledge database will be updated continuously with newly acquired patterns. In short, the system will be composed of 2 parts, learning from data sets and fine-tuning the knowledge-base using neural network and the use of fuzzy logic in making decision based on the rules and membership functions inside the knowledge base. This paper will discuss the idea, steps and issues involved in creating such a system.