A neural network contains no domain knowledge in the beginning, but it can be trained to make decisions by mapping example pairs of input data into example output vectors, and adjusting its weights so that it maps each input example vector into the corresponding output exemplar vector approximately. A knowledge base pertaining to the internal representations is automatically constructed from the data presented to train the network. Well-trained neural networks represent a knowledge base in which knowledge is spread in the form of weighted interconnections where a learning algorithm is used to alter the knowledge base from a set of given representative cases. A generic form of a neural network intrusion detector is presented in the below Figure3.2.…show more content… Support vector Machine is binary classifier, the performance of classification of support vector machine is high in comparison of another binary classifier such as decision tree, KNN and Bay,s classifier[35]. SVMs plot the training vectors in high dimensional feature space through nonlinear mapping and labeling each vector by its class. The data is then classified by deciding a set of support vectors, which are members of the set of training inputs that outline a hyper plane in the feature space. SVM is a technique for solving a variety of learning, Classification and prediction problems. The basic SVM deals with two-class problems in which the data are divided by a hyper plane defined by a number of support vectors. Support vectors are a subset of training data used to define the boundary between the two classes. In situations where SVM cannot separate two classes, it solves this problem by mapping input data into high-dimensional feature spaces using a kernel function. In high-dimensional space it is possible to create a hyper plane that allows linear separation. Compared with the ANN, the SVM have two advantages. The initial one is the global optimum can be derived. Secondly, the over fitting problem can be easily controlled by the choice of a suitable margin that separates the data. Experiential
