Message filtering method and device

Abstract

The invention relates to a message filtering method. A Bayes classifier model is used for calculating the probability that a message is a rogue message; the property of the message is judged according to the probability of the rogue message, wherein the property of the message includes the normal message, the rogue message and the suspicious message; the rogue message is directly filtered away, and the normal message and the rogue message are remained; then, the further classification is performed by using a semanticbased deep study model; and the message is determined to be the normal message or the rogue message. Compared with the prior art, the method and the device have the advantages that the automatic study can be realized; a training set is regularly updated; the synonym is recognized; the manual annotation cost is greatly reduced; and high accuracy, robustness and stability are realized. In addition, the invention also provides a message filtering realization device.

Description

technical field [0001] The invention belongs to the fields of message processing and data mining, and in particular relates to a message filtering method and device based on Bayesian algorithm and deep learning. Background technique [0002] Message filtering technology is to filter out violent, reactionary, pornographic and other bad information by filtering information. It is often used to filter information published in the network environment, or to filter messages in barrage comments. Existing message filtering technology generally performs keyword matching on each message, artificially sets some sensitive words that need to be filtered in advance, and then judges whether the text of the message that appears contains these words, if it does, filter the message, if it does not contain , the message is considered to be a normal message, but the function of this method is very limited and inflexible, and it is necessary to manually set the words to be filtered in advance, ...

AI Technical Summary

Problems solved by technology

The existing message filtering technology generally performs keyword matching on each message, artificially sets some sensitive words that need to be filtered in advance, and then judges whether the text of the message that appears contains these words, if it does, filter the message, if it does not contain , the message is considered to be a normal message, but the effect of this method is very limited and inflexible, and it is necessary to manually set the words that need to be filtered in advance, and artificially set a number of keywords that are easily used by the message sender in other ways such as phonetic characters, forms, etc. Near words or inserting other symbols in the middle of words, etc., and the content and sentence patterns of bad news are constantly changing. Existing text filtering technologies often cannot respond to changes in a timely manner.

Images