Public opinion smart supervision method

Abstract

The invention discloses a public opinion smart supervision method and a public opinion smart supervision system. The public opinion smart supervision method provided by the invention comprises the steps of S1, obtaining network public opinion texts by employing keyword lists and network public opinion website lists correlated to city emergency events; S2, analyzing the public opinion texts, thereby obtaining subjects and objects in the emergency events and emergency event characteristic identifiers; S3, carrying out semantic similarity evaluation analysis on the subjects and objects in the emergency events and the emergency event characteristic identifiers, clustering the network public opinion texts and determining hot event classes to which the network public opinion texts belong; and S4, output a clustering result and the determined hot event classes to which the network public opinion texts belong. Through application of the method, vocabularies such as geographic names, names and organization names with high event characteristic representation can be identified rapidly and accurately and the described public opinion events can be depicted deeply and carefully.

Description

technical field [0001] The invention relates to the field of network technology and language processing technology, in particular to a public opinion intelligent supervision method for analyzing urban emergency events by using public opinion texts of network sites. Background technique [0002] With the development of Web2.0 technology, digital new media such as the Internet and mobile phones are developing rapidly, and gradually replacing traditional media such as newspapers, radio, and television as the main means of public information exchange. The convenience of publishing and disseminating online texts also makes it play a more significant role in fueling urban emergencies, crises, and mass incidents. Therefore, reasonable supervision and guidance of network public opinion is of great significance for maintaining social harmony and stability. [0003] At present, the understanding and processing of Chinese texts is still mainly in the stage of keyword extraction and co...

AI Technical Summary

Problems solved by technology

In other words, the existing methods cannot further understand and process the key situations described in the text, and are not sensitive enough to the names of places, people, and organizations mentioned in the text, and the effectiveness of the method is easily affected by noise words (such as typos, typos, etc.) in the text. nonsensical text) interference and have a greater impact

For the mining of Internet public opinion texts that mainly include daily natural language communication, most of the existing methods are powerless, not to mention the accurate identification and intelligent supervision of hot events in Internet public opinion

Images