Algorithm for detecting fake news
Fake News what is it?
Fake news is information that is deliberately deceived or misinformed. They are distributed through electronic media or social networking sites. Fake news is published for the purpose of making a mistake or for the purpose of obtaining the political or financial gain. Fake news should be distinguished from parody or satire, the purpose of which is to entertain the recipient, not to mislead him.
Algorithm
The University of Michigan informed that it has an algorithm which exceeds people in identifying false information.
Tests have shown that the algorithm will be able to catch as much as 76% of the false information. A person, on the other hand, is able to catch 70% of the false information.
The linguistic analysis used by the algorithm is able to identify false information even when it is relatively recent so that it can be compared with other sources.
In particular, content aggregators can play an important role, where employees of such websites may not be able to catch false information downloaded from other websites or social networking sites in a timely manner.
If the effectiveness of the algorithm is confirmed in practice and works in different languages, it can protect such services as Twitter, Facebook or Google from the appearance of false messages.
Employees of large social networking sites may sometimes be unable to catch up with false information they have downloaded from social networking sites or websites at the right time.
At present, it is being verified that incorrect information is being cross-checked with a number of sources. The method is slow and many people will manage to read it before false information can be seen. The correction of incorrect information will not reach many readers.
The linguistic analysis itself consists of the algorithm intercepting grammatical structures, vocabulary, phrases or punctuation. It works much faster than people. In the future, it may be used more widely. A tool of this type can inform the reader about the reliability of a given content. It can also be used to mark content that employees of a given website need to check.
You can imagine any number of applications for this on the front or back end of a news or social media site. It could provide users with an estimate of the trustworthiness of individual stories or a whole news site. Or it could be the first line of defense on the back end of a news site, flagging suspicious stories for further review. A 76 percent success rate leaves a fairly large margin of error, but it can still provide valuable insight when it's used alongside humans.
The professor said
A team of Mihalcea-led scientists created their own set of false information. Researchers turned to the Amazon Mechanical Turk community and found people willing to convert truthful information into false information. Researchers have managed to gather a thousand pieces of information. Five hundred true and five hundred false. The information has been appropriately marked and used to learn the algorithm.
Then, scientists from Michigan downloaded real and false information from the Internet, checking it with a linguistic algorithm. It turned out that the algorithm was able to catch 76% of the false information.
The source code of the algorithm was made available online for free. Professor Rada wants social media to be able to "build their own lie detectors by themselves"'.
Sources
https://news.umich.edu/fake-news-detector-algorithm-works-better-than-a-human/
https://newsmavens.com/news/lessons-learned/1876/researcher-from-romania-develops-algorithmto-detect-fake-news
https://thenextweb.com/artificial-intelligence/2018/08/22/this-fake-news-detection-algorithm-outperforms-humans/
Photos:
https://pixabay.com/pl/fake-aktualno%C5%9Bci-media-2355686/
https://pixabay.com/pl/zapobieganie-nadu%C5%BCyciom-oszustwo-3188092/
https://pixabay.com/pl/kod-programowanie-komputer-danych-1486361/
The article is dedicated to @katerinaramm
Good article - helpful algorithms are always..., well - helpful :)
The one which detects false info is especially useful.
Thank you for the good job @krasnalek
The one algorithm I am looking for, is the one which will exchange internet and electronic world for the real one.
Have a nice day :)
PS: Have you seen this movie "Ready Player One"?
I liked it.
Thank you. Unfortunately, I did not see the film :D
Worth of seeing (in my opinion) - it is on DVD now.
Check out a trailer on YouTube :)
Congratulations @krasnalek! You have completed the following achievement on Steemit and have been rewarded with new badge(s) :
Click on the badge to view your Board of Honor.
If you no longer want to receive notifications, reply to this comment with the word
STOP