A multi-level classification system and method based on news text information

Abstract

The invention discloses a multi-level classification system and method based on news text information, and relates to the technical field of file classification. The system includes a training module, a multi-level classification module and a result determining module; the training module is used for regarding all levels of categories of the news text information, training a preset training sample set through multiple machine learning algorithms and determining the number and type of classifiers corresponding to each level of category according to training results; the multi-level classification module is used for configuring corresponding multi-level classification models according to the number and type of the classifiers corresponding to each level of category, wherein the number and type of the classifiers are determined by the training module; the result determining module is used for inputting the obtained news text information to be classified into the multi-level classification models to conduct the classification, and is used for determining output results of the multi-level classification models as final classification results of the news text information to be classified. In the way, according to the multi-level classification system and method, the problem of inaccurate classification results caused by unbalanced sample data is solved in a pointed mode, the classification accuracy is effectively improved, and the classification efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of file classification, in particular to a multi-level classification system and method based on news text information. Background technique [0002] With the development of the Internet age, network resources are becoming more and more abundant, and there are more and more types. In order to effectively retrieve and utilize various resources on the network, it is particularly important to classify the above-mentioned network resources accurately and comprehensively. With the emergence and development of machine learning algorithms, more and more people have applied machine learning algorithms to news text information classification methods. [0003] However, during the process of implementing the present invention, the inventors found at least the following problems in the prior art: in many specific application scenarios, due to various reasons, the distribution of sample data may be unbalanced. When enc...

AI Technical Summary

Problems solved by technology

[0003] However, in the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art: In many specific application scenarios, due to various reasons, the distribution of sample data may be unbalanced

When encountering unbalanced data, the non-hierarchical news text information classification method implemented by the machine learning algorithm in the prior art will cause the machine learning algorithm to pay too much attention to the samples of the majority class due to the imbalance of the sample data, and make the minority class Class samples cannot be accurately identified, thus reducing the accuracy of these news text information classification methods as a whole

Images