Abstract
In this paper, we applied the online semi-supervised learning and boosting method to the complex data that difficult to extract the feature, or have sequential change of samples and classes. The semi-supervised learning is a learning method using the data which include labeled samples and unlabeled samples in order to increase accuracy. The accuracy of semi-supervised learning depends on the number of labeled samples however it is necessary to large cost for gathering labeled samples from humans.
Therefore we proposed an interface application for semi-supervised learning via mobile web in order to gather the number of labeled samples. When samples are labeled by different labeling functions such as peoples or machines, the given labels are different between each other. For aggregating these labels, the ensemble learning techniques such as Boosting and Cluster ensemble are used.
