Lacus VS Lakshmanan, Conversation
Published 1 hour ago: June 19, 2023 at 9:00 PM
Fake news is a complex issue and can span text, images and video.
There are several ways to generate fake news, especially for written articles. Fake news stories can be created by selectively editing facts such as people’s names, dates and statistics. Articles can also be completely fabricated using fabricated events and people.
Fake news stories can also be machine-generated, especially since advances in artificial intelligence have made it particularly easy to generate false information.
harmful effects
Questions such as “Was there voter fraud during the 2020 US election?” or “Is climate change a hoax?” You can answer these questions right or wrong, but questions like these can contain incorrect information.
Misinformation and disinformation, or fake news, can have a negative impact on many people in a short period of time. The concept of fake news existed long before technology advanced, but social media has exacerbated the problem.
A 2018 Twitter study showed that fake news stories are more likely to be retweeted by humans than by bots, and are 70 percent more likely to be retweeted than truthful stories. The same study found that true stories take about six times as long to reach a group of 1,500 people, and while true stories rarely reach more than 1,000 people, popular fake news can reach up to 100,000 people. It turns out that there is
The 2020 US presidential election, the COVID-19 vaccine, and climate change have all been the subject of misinformation campaigns with grave consequences. Losses due to misinformation about COVID-19 are estimated at US$50-300 million per day. The cost of political misinformation can lead to civil unrest, violence, and even loss of public trust in democratic institutions.
Misinformation detection
Misinformation detection can be done with a combination of algorithms, machine learning models, artificial intelligence, and humans. The key question is, if misinformation is discovered, who is responsible for controlling it, if not stopping its spread? Only social media companies are in a position to actually control the spread of information through their networks.
A particularly simple but effective means of generating misinformation is selective editing of news articles. For example, consider a Ukrainian director and playwright arrested and charged with ‘justifying terrorism’. This was accomplished by replacing “Russia” in the original sentence of the actual news article with “Ukraine”.
Detecting online misinformation and controlling its growth and spread requires a multifaceted approach.
Communication in social media can be modeled as a network, with users forming the points in the network model and communication forming the links between them. Retweeting or liking a post reflects the connection between the two points. In this network model, disinformation disseminators tend to form a much more tightly connected core-perimeter structure than truth-dissemination users.
My research group has developed an efficient algorithm for detecting dense structures from communication networks. This information can be further analyzed to detect instances of misinformation campaigns.
Since these algorithms rely only on the communication structure, content analysis performed by algorithms and humans is required to confirm instances of misinformation.
Detecting manipulated articles requires careful analysis. In our study, we used a neural network-based approach to detect such tampering by combining textual information with an external knowledge base.
stop the spread
Detecting misinformation is only half the battle. Decisive action is needed to stop its spread. Strategies to combat the spread of misinformation on social networks include both intervention by internet platforms and the launch of counter-campaigns to neutralize fake news campaigns.
Intervention can be hard, such as terminating a user’s account, or soft, such as labeling a post as suspicious.
Networks powered by algorithms and AI are not 100% reliable. Just like not intervening with counterfeit items, intervening with genuine items by mistake is costly.
To that end, we designed a smart intervention policy that automatically decides whether to intervene on an item based on its predicted veracity and predicted popularity.
Fighting fake news
In launching a counter-campaign to minimize, if not neutralize, the impact of a misinformation campaign, consider the stark difference between truth and fake news in terms of the speed and pervasiveness of each. There is a need.
In addition to these differences, responses to stories also vary by user, topic, and post length. Our approach considers all these factors and devise effective counter-campaign strategies that effectively mitigate the dissemination of misinformation.
Recent advances in generative AI, especially AI powered by large-scale language models such as ChatGPT, have made it easier than ever to create high-speed, high-volume articles, detect misinformation, and generate large-scale, real-time The challenge is to combat its proliferation in Our current research continues to address this ongoing challenge with enormous implications for society.
Want to know more about the future of AI, chatbots and machine learning? Check out the full content artificial intelligenceor see the guide Best Free AI Art Generator and Everything we know about OpenAI’s ChatGPT.
Lacus VS LakshmananProfessor of Computer Science, University of British Columbia
This article is reprinted from conversation Under Creative Commons License.read Original work.
