Research on Incremental Learning Method Based on Support Vector Machine Method

An incremental learning algorithm based on support vector machine was proposed to process large-scale data or data generated in batches. Initial goal concept learnt by standard support vector machine algorithm was updated by an updating model. Compared with the existing incremental learning algorithms, this algorithm can achieve the incremental inverse process and the training time is in inverse proportion. Results show that the algorithm satisfies the three criteria of stability, improvement and recoverability. The main target of this thesis is to find a new SVM learning algorithm based on weighted incremental method above the current SVM algorithms, and this new algorithm will have a better classification performance in large-scale data situation.