Study on Multi-label Text Classification Based on SVM

Two multi-label text classification algorithms are proposed. Firstly, one-against-rest method is used to train sub-classifiers. For the text to be classified, the sub-classifiers are used to obtain the membership vector, and then confirm the classes of the text. Secondly, hyper-sphere support vector machine is used to obtain the smallest hyper-spheres in feature space that contains most texts of the class, which can divide the class texts from others. For the text to be classified, the distances from it to the centre of every hyper-sphere are used to confirm the classes of the text. The experimental results show that the algorithms have high performance on recall, precision, and F1.