Abstract
The traditional classification method is difficult to achieve good classification results when the training samples are few, and the unsupervised classification algorithms cannot use class information to improve their performance. Thus, it is necessary to apply semi-supervised classification methods. This chapter introduces semi-supervised data classification based on sparse representation and stochastic subspace. First, the dictionary in sparse representation is simplified to improve the speed and accuracy of sparse representation. To meet the needs of a complete dictionary, we combine each random subspace of sparse representation when the dimensions of the original sample are far lower than the random subspace to enhance the ability of the original data. Second, with traditional random subspace methods, all features have the same probability for selection, and a method based on attribute features is introduced, promoting the accuracy of selected key feature probabilities to enhance the final accuracy of data classification.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have