The primary aim of data mining is to discover patterns in the data that lead to better understanding of the data generating process and to useful predictions. Examples of applications of data mining include detecting fraudulent credit card transactions, character recognition in automated zip code reading, and predicting compound activity in drug discovery. Real-world data sets are often characterized by having large numbers of examples, being highly unbalanced, and being corrupted by noise. The relationship between predictive variables and the target concept is often highly non-linear. One recent technique that has been developed to address these issues is the support vector machine (SVM). The SVM has been developed as robust tool for classification and regression in noisy, complex domains.
The two key features of SVMs are generalisation theory, which leads to a principled way to choose an hypothesis; and, kernel functions, which introduce non-linearity in the hypothesis space without explicitly requiring a non-linear algorithm. In this presentation I outline the basic ideas of SVMs and their advanages over existing data analysis techniques, also are noted some important points for the data mining practitioner who wishes to use support vector machines.