For a long decade clustering faced a problem of noise and outliers. Support Vector Clustering is one of the techniques in pattern recognition. Support Vector Clustering is Kernel-Based Clustering. Division of patterns, data items, and feature vectors into groups (clusters) is a complicated task since clustering does not assume any prior knowledge, which are the clusters to be searched for. Noise and outlier reduces the mapping probability of sphere in support vector clustering. Support vector clustering is inspired clustering technique form the support vector Machine. The prediction and accuracy of support vector clustering depends upon kernel function of hyper plane. Kernel function is a heart of classifier. In this paper we present review of support vector clustering technique for pattern detection and reorganisation for very large databases. The variation of performance of support vector clustering depends upon kernel of classifier. Here we discuss different method of kernel used in support vector clustering.