Subject-oriented customized news information extraction system

Abstract

The invention discloses a subject-oriented customized new information extraction system which comprises a news collecting subsystem, a text processing subsystem and a human-computer interaction subsystem, wherein the news collecting subsystem completes the functions of searching related news about related subjects customized by a user and extracting news texts; the text processing subsystem divides the texts into different categories, detects and tracks topics in the content of the texts on that basis, automatically generates abstracts and establishes corresponding indexes; and the human-computer interaction subsystem firstly analyzes the topics, calculates the hot degree of the topics, presents the hot topics in the groups of subjects of the topics for the user in the sequence of hot degree and simultaneously provides topic retrieval, and the user can artificially screen the obtained content and store intelligence obtained by artificially extracting the screened information into an intelligence library. With the system, news on the internet can be comprehensively collected in time and can be automatically detected, classified and tracked, and intelligence required by the user can be extracted from immense network news by exerting the cognitive ability of the user on intelligence.

Description

technical field [0001] The invention relates to a news information extraction system, in particular to a theme-oriented and customized information extraction system with news as an object. Background technique [0002] Today, with the rapid development of the Internet, using the public information release system to collect intelligence in the fields of politics, military, economy, and culture has become one of the important channels for obtaining intelligence. According to the definition of intelligence in information science, the so-called intelligence refers to real-time required information within an effective time. At present, 90% of the intelligence is obtained from the public information release system, and all kinds of news information is undoubtedly the largest amount of public information. [0003] However, Internet news is a massive information source, and it is an open and distributed information space. The following characteristics inherent in it have obviously ...

AI Technical Summary

Problems solved by technology

At present, all the found information is manually judged and collected one by one, which is very inefficient;

[0006] (2) It is difficult for a piece of online news to explain the incident clearly at one time, and there may be new situations and new problems as the incident progresses

[0007] (3) Search engines cannot guarantee the timeliness and authority of information when retrieving information, which is a very serious and even fatal weakness for intelligence extraction

Images