Big data-based web article forwarding recognition method

Abstract

The invention discloses a big data-based web article forwarding recognition method. The method comprises the following steps of: regularly and incessantly collecting various types of articles on the internet through a search engine, and establishing an article data warehouse according to the collected articles; and confirming to-be-recognized articles and dominance and recessiveness thereof, carrying out forwarding recognition on dominant forwarded articles through the comparison of transmission links, searching articles Pm associated with the to-be-recognized articles for the recessive forwarded article through the comparison of fuzzy Hash values, further recognizing articles Pe having forwarding relationship with the articles Pm, and carrying out rearrangement according to the sequential order of the transmission time of the articles Pe so as to find a source article. Through the method disclosed by the invention, the forwarding paths of the articles can be found, so that the problem that the manual investigation is long in time consumption and low in efficiency is solved; according to the searched articles with the forwarding relationship, data basis is provided to the related departments to forbid the transmission of harmful web articles; and moreover, the method can be used for judging the original creativity of web articles and assessing the influences of the articles.

Description

technical field [0001] The invention relates to a network article forwarding identification technology, in particular to a method for network article forwarding identification based on big data. Background technique [0002] With the rapid development of the Internet, online media is also developing strongly. As the most important form of expression in online media, online articles, including news, entertainment news, sports reports, etc., are reprinted in large numbers on Weibo, WeChat, blogs and other news media. and dissemination; on the other hand, more and more netizens are accustomed to expressing their views and opinions on various news information on the Internet. Comments form online articles, and such online articles are also widely disseminated and reprinted. [0003] A large number of false, fraudulent, and harmful articles are disseminated wantonly on the Internet. If relevant departments want to prohibit the dissemination of such articles, they must find the...

AI Technical Summary

Problems solved by technology

[0003] A large number of false, fraudulent, and harmful articles are disseminated wantonly on the Internet. If relevant departments want to prohibit the dissemination of such articles, they must find the source of dissemination of such articles. In the prior art, such articles can only be found through manual investigation. However, after an article is published on the Internet, due to the complexity of the network, it has the characteristics of multi-level forwarding, multi-path, and large forwarding volume, and finally forms a multi-level mesh forwarding path with a complex structure; The investigation mainly finds its forwarding path through netizens’ reports, gateway supervision and other means, which takes a long time and is inefficient

In particular, if this type of article is an implicit forwarding article, that is, copying or partially copying other people's articles by computer means such as copying and pasting for self-publishing, the forwarding of this type of article does not have a forwarding link or indicate the name of the article source, and it is easy to form multi-level cross-site dissemination, it is extremely difficult to find the source article through artificial ranking. Even if the source article is found, there is no effective means to effectively prohibit the dissemination of such articles

Images