Information classification method based on granularity calculation

Abstract

The invention discloses an information classification method based on granularity calculation. The method comprises the following steps: acquiring information; performing word segmentation processingand obtaining a vocabulary set; performing part-of-speech tagging and word meaning tagging; obtaining category vocabularies matched with the vocabulary set from the vocabulary set according to each piece of pre-stored category information; determining the weight of each category of vocabularies in the vocabulary set based on granularity calculation; determining target category information to whichall category vocabularies contained in the vocabulary set belong; determining the weight of each piece of target category information based on granularity calculation; determining at least one category to which the information belongs according to the weight of each piece of target category information; judging whether the information and the probability of each category are greater than a presetthreshold value or not. Information classification is performed based on granularity calculation, an information classification method is optimized, operation is easy and convenient, classification efficiency is high, classification accuracy is high, and time and labor are saved.

Description

technical field [0001] The invention relates to the technical field of information classification, in particular to an information classification method based on granularity calculation. Background technique [0002] With the rapid development of information technology, especially the rapid development of mobile communication technology and computer Internet technology, the mobile communication network has undergone revolutionary changes, and information contains huge information content and value. , people put forward the need to classify information in order to process and analyze the classified information or the objects corresponding to the information. In the prior art, the method of classifying information is mainly to manually define a specific short message template , and then set up regular expression processing, structure the required data items, and classify the template data sets. Specifically, there are two main methods for classifying information. One is to jud...

AI Technical Summary

Problems solved by technology

[0002] With the rapid development of information technology, especially the rapid development of mobile communication technology and computer Internet technology, the mobile communication network has undergone revolutionary changes, and information contains huge information content and value. , people put forward the need to classify information in order to process and analyze the classified information or the objects corresponding to the information. In the prior art, the method of classifying information is mainly to manually define a specific short message template , and then set up regular expression processing, structure the required data items, and classify the template data sets. Specifically, there are two main methods for classifying information. One is to judge based on keywords, that is, first Manually classify all information, and then extract specific keywords for each type of information as an identifier for entering this data set, but in the above method, information in different data sets may match the same keyword or a group of keywords , resulting in inaccurate classification. Another method is to define a regular expression for each piece of information, and define a data set for each piece of regular expression, that is, to classify each information once. However, in the above method, each Information defines regular expressions. As the amount of information continues to increase, the workload increases sharply, and the efficiency of classification is not high;

[0003] However, the existing information classification methods are complex in operation, low in efficiency of information classification, and low in classification accuracy.

Images