A Sensitive Information Monitoring Method Based on Event Search

Abstract

The invention belongs to the field of natural language processing, and particularly provides a sensitive information monitoring method based on event search, which includes the steps such as event identification, event extraction, and prewarning issuing. The method automatically and intelligently monitors a certain kind of events by utilizing the search mechanism based on the events, so as to filter a large amount of irrelevant information and obtain the higher accuracy rate of the result.

Description

technical field [0001] The invention belongs to the field of natural language processing, in particular to a sensitive information monitoring method based on event search. Background technique [0002] With the advent of the Internet information age, how to quickly obtain accurate information from the Internet, especially sensitive information, is an important indicator of social progress. Mastering information means taking the initiative, and capturing sensitive information in a timely manner lays the foundation for correct decision-making. This is of great significance to national security, government operations and commercial activities. [0003] First of all, national security is the country's top priority, and information security is also a part of national security. Timely control and processing of sensitive information is a prerequisite for maintaining national security and maintaining social stability. Secondly, as a functional department serving the people, gover...

AI Technical Summary

Problems solved by technology

[0029]First, the keyword matching of search engines cannot describe users’ search needs. Li Zizheng went to visit, but there was a report that Li Zizheng was interviewed

[0030]Second, the search engine does not consider the structure of keywords, for example, in the query "Which countries was visited by former Singapore Prime Minister Lee Kuan Yew", "Mr. Lee" and "Visit "There is a certain syntactic and semantic structure between them, but they are all ignored by search engines

[0031]Third, the search engine does not consider concept matching. For example, for the query "Mr. Li visited the country", in the results, neither "Ma Guo" nor "China" was matched above, showing that the search engine does not recognize their relationship to "Country"

Another evidence is that if only "Lee Kuan Yew" appears in a report, but not "Senior Lee", the search for "Senior Lee's interview" and "Senior Lee's interview with country" may not be searchable

[0032] Fourth, the results of search engines are still unstructured texts, which are only for human viewing, and computers cannot directly use their search results for further processing

The performance of statistics-based technology is still far from specific applications; while the technology based on artificial rules is poor in portability and requires a lot of human operation

Images