Cross-domain false news detection method
A detection method and a cross-domain technology, applied in the field of fake news detection, can solve the problems of modeling rumor classification tasks, damage rumor detection performance, etc., and achieve the effect of improving detection performance and reducing performance loss
Active Publication Date: 2021-08-27
杭州中科睿鉴科技有限公司
View PDF4 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Most studies treat all fields equally and extract the common features of rumors in all fields. Rumors in different fields have both common features and characteristic features. Only considering the common features cannot model the rumor classification task well. In addition, Due to the difference in number distribution between domains, rumor features in a small domain will be submerged in a large domain, impairing the rumor detection performance in a small domain
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0033] This embodiment is a cross-domain fake news detection method, and the specific steps include:
[0034] S1. Input the news text into the trained domain common feature extraction model, extract the domain common features of the news text, and obtain the rumor classification result of the domain common features.
[0035] In this example, the domain common feature extraction model learns the domain common feature expression through the method of inter-domain confrontation training. The domain common feature extraction model includes a common feature extractor, a domain category classifier and a rumor classifier. The common feature extractor is used as The generator extracts the domain common features of the news text, uses the domain category classifier as the discriminator, and uses the rumor classifier to do the false news classification task.
[0036] In this example, textCNN is used as a common feature extractor to extract domain common features in text; a multi-layer f...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to a cross-domain false news detection method. The method is suitable for the field of false news detection. According to the technical scheme, the cross-domain false news detection method is characterized in that a news text is input into a trained domain common feature extraction model, domain common features of the news text are extracted, and a rumor classification result of the domain common features is obtained; the method also includes inputting the same news text into a trained domain characteristic feature extraction model, extracting domain characteristic features of the news text, and obtaining a rumor classification result of the domain characteristic features; and carrying out weighted summation on the rumor classification result of the domain common feature and the rumor classification result of the domain characteristic feature to obtain a judgment result of the news text through false news detection.
Description
technical field [0001] The invention relates to a cross-domain fake news detection method. Applicable to the field of fake news detection. Background technique [0002] Today, with the rapid development of the Internet, the number of netizens is gradually expanding, and online social media platforms represented by Sina Weibo and Twitter have also sprung up like mushrooms after rain. While the rise of social media has brought convenience to people, it has also made fake news spread wildly on the Internet. False information harms a wide range, ranging from small individuals to large societies. The increasingly rampant spread of rumors not only affects the economy of society, but also damages the credibility of the government and the media. [0003] Fake news is defined as: news that is deliberately fabricated and can be proven to be false. For the convenience of expression, the concept of rumors in this article is equivalent to fake news. False news on social media platfor...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/35G06F16/953G06K9/46G06N3/04G06N3/08
CPCG06F16/35G06F16/953G06N3/08G06V10/40G06N3/045
Inventor 曹娟王彦焱徐朝喜谢添李锦涛
Owner 杭州中科睿鉴科技有限公司