Subject determination method and device of text specified information and computer storage medium

Abstract

A subject determination method of text specified information comprises the steps of performing word segmentation on a target text; performing part-of-speech tagging on each segmented word to obtain apart-of-speech tagging result of each segmented word; determining at least one candidate subject according to the part-of-speech tagging result of each segmented word; dividing the target text according to each determined candidate main body to obtain a sample corresponding to each candidate subject; obtaining a vector V of each sample, and inputting the vector V into a pre-trained first neural network to determine whether a sample with the specified information exists or not; and when it is determined that the sample with the specified information exists, determining that the candidate main body corresponding to the sample is the subject with the specified information. According to the subject determination method, manual annotation can be reduced and the cost is reduced.

Description

technical field [0001] This article relates to computer technology, especially a method, device and storage medium for determining the subject of text-specified information. Background technique [0002] The task of determining the subject of negative information is a common application in network public opinion monitoring. Its purpose is to determine whether the text contains negative information given the text to be analyzed, and if it contains negative information, give the name (or position in the original text) of the subject involved in the negative information. [0003] Existing statistical learning methods spend a lot of cost on artificial feature construction, which is not only time-consuming and laborious, but also leads to a lack of generalization ability of the model on new patterns other than the encoded features. [0004] The existing statistical learning method using deep neural network avoids the cumbersome process of artificial feature construction through ...

AI Technical Summary

Problems solved by technology

[0003] Existing statistical learning methods spend a lot of money on the construction of artificial features, which is not only time-consuming and laborious, but also leads to the lack of generalization ability of the model on new modes other than the encoded features.

Images