Data Condensation in Large Databases by Incremental Learning with Support Vector Machines

An algorithm for data condensation using support vector machines (SVM's) is presented. The algorithm extracts data points lying close to the class boundaries, which form a much reduced but critical set for classification. Adopting an active incremental learning algorithm circumvents the problem of large memory requirements for training SVM's in batch mode. The learning strategy is motivated from the condensed nearest neighbor classification technique. Experimental results presented show that such active incremental learning enjoys superiority in terms of computation time and condensation ratio, over related methods.