A novel online incremental and decremental learning algorithm based on variable support vector machine
Yuantao Chen,Jie Xiong,Weihong Xu,Jingwen Zuo
DOI: https://doi.org/10.1007/s10586-018-1772-4
2018-01-17
Cluster Computing
Abstract:In view of the long execution time and low execution efficiency of Support Vector Machine in large-scale training samples, the paper has proposed the online incremental and decremental learning algorithm based on variable support vector machine (VSVM). In deep understanding of the operation mechanism and correlation algorithms for VSVM, each sample has increased training datasets changes and it needs to update the classifier of learning algorithm. Firstly, they are given the online growth amount of learning algorithm taken full advantage of the incremental pre-calculated information, and doesn’t require retraining for the new incremental training datasets. Secondly, the incremental matrix inverse calculation process had greatly reduced the running time of algorithm, and it is given in order to verify out the validity of the online learning algorithm. Finally, the nine groups of datasets in the standard library have been selected in the pattern classification experiment. The experimental results are shown that the online learning algorithm given in the case to ensure the correct classification rates and effective training’s speed. With the implementation of the incremental process, training meetings, the need for large-scale data storage space, result in slow training, the online learning algorithm based on VSVM can solve the problem.