Offline model improvement and selection method for junk short message classification

Abstract

The present invention discloses an offline model improvement and selection method for junk short message classification. The method comprises the following steps: (1) carrying out feature selection and expansion, selecting features by using a feature selection method, constructing a feature word vector, and using a feature word vector model to express an original short message text; (2) carrying out optimization training and testing on the offline classification algorithm and improvement, carrying out improvement for junk short message classification on the offline classification algorithms, according to each offline classification algorithm and improvement, carrying out data preparation on the training set and the testing set obtained in the step (1), and using the training set to carry out optimization training and testing on each offline classification algorithm and improvement; and (3) based on the offline classification algorithm selection of the evaluation criteria, putting forward an evaluation index for the junk short message classification, analyzing the test result obtained in the step (2) by using the evaluation index, and selecting an optimal offline classification algorithm.

Description

technical field [0001] The invention relates to an offline text classification algorithm, in particular to an offline model improvement and selection method for classification of junk short messages, and belongs to the technical field of text content-based identification of junk short messages. Background technique [0002] The most important thing in the text classification problem is to select and train the text classification model, and the performance of text classification depends on the text classification model to a large extent. Recently, researchers have proposed various text classification models based on machine learning, combined with multidisciplinary theories such as statistics and informatics. [0003] Naive Bayesian classification algorithm is a machine learning method based on statistics, which is widely used in text classification problems. The algorithm is based on the assumption of feature independence. Although there is often correlation between feature...

AI Technical Summary

Problems solved by technology

No training is required, but the classification error is also large. If the k value is selected too small, it is easily affected by the noise data. If the k value is selected too large, then the training samples with a large distance (dissimilarity) from the test sample will also Works on the forecast, producing wrong forecast results

Images