Text training method and text classifying method

Abstract

The invention provides a text training method. The method comprises the following steps: 1) computing a centre vector of each classification of training sample sets; 2) classifying samples in the training sample sets according to the centre vector of the training sample sets; and 3) for the samples which are classified incorrectly, correcting the centre vector of a classification A to which the incorrectly classified samples belong or / and the centre vector of a classification B to which the samples are classified by mistake according to a set drag weight and a set push weight. The text can be classified with high precision and fast speed according to the centre vector obtained by the training method.

Description

technical field [0001] The invention relates to the field of pattern recognition, in particular to a text training method and a classification method. Background technique [0002] With the rapid development of Internet technology, online texts have shown exponential growth. Facing the vast amount of text data, people urgently need an efficient tool that can automatically process and organize large-scale text. Automatic text classification is to automatically classify a large amount of text according to its content, so as to help people effectively process and organize text data. Therefore, the research and development of high-performance text classification methods has increasingly become a research hotspot in the field of information retrieval. Many classic machine learning methods have been introduced into text classification. The more commonly used methods are central classification method, Rocchio, nearest neighbor (KNN), Winnow, Bayesian (NB), support vector machine...

AI Technical Summary

Problems solved by technology

But their training and classification time are very long, it is difficult to meet the needs of large-scale text classification

On the contrary, the central classification method, Bayesian, Rocchio and Winnow are typical linear classifiers, and the training and classification time are linearly related to the scale of the problem, so they can meet the classification needs of large-scale text in time, but their accuracy often less than ideal

Therefore, in general, the current classification method is still in a situation where it is difficult to meet the requirements of high-performance text classification methods.

Images