You are reading the article How To Identify Fake News From Real News Online updated in December 2023 on the website Hatcungthantuong.com. We hope that the information we have shared is helpful to you. If you find the content interesting and meaningful, please share it with your friends and continue to follow and support us for the latest updates. Suggested January 2024 How To Identify Fake News From Real News Online
Thanks to modern technology, we live in an era of unprecedented connectivity, real-time updates, and incredible amounts of data at our fingertips.
The irony is that despite access to reliable information, disinformation runs rampant.
And it’s not just flat-earthers, moon landing hoax believers, and tinfoil JFK conspiracy theorists.
There is a lot of seemingly credible information out there that, upon further inspection, turns out to be inaccurate, out of context, or flat-out false information.
The amount of fake news has grown at an incredible rate.
And as the purveyors of these fabrications and half-truths have grown increasingly media-savvy, it’s becoming harder to determine what’s real and what’s “fake news.”
This is a significant problem. Globally, 76% of people acknowledged worries about weaponizing false information or fake news – an all-time high.
Social media platforms have contributed to this, often serving as echo chambers in which these stories get much more exposure than they would ever have received in the pre-internet world.
And this, in turn, offers them a false sense of credibility.
While even the most intelligent people can be mistaken occasionally, maintaining a healthy degree of skepticism and evaluating the source of information can combat your vulnerability to misinformation.
What Is Fake News?In its research guide, the University of Michigan Library describes fake news as “those news stories that are false: the story itself is fabricated, with no verifiable facts, sources or quotes.
This separates fake news from misinformation, which is inaccurate information created or shared by mistake without the intention to deceive.
An example is the famous “Dewey Defeats Truman” headline from the Chicago Tribune.
On the other side is fake news or disinformation, which is intentionally false or misleading.
First Draft News, a nonprofit coalition created to fight false information online, identifies seven different types of mis- and dis-information:
Satire or parody – no harm is intended, but it has the potential to fool.
False connection – headlines, visuals, or captions that do not support the content and thus mislead the reader.
Misleading content – content is purposefully framed to portray a person or issue in a particular light.
False context – stories are presented with false or missing contextual information.
Imposter content – uses fabricated quotes or information from genuine sources.
Manipulated content – manipulated information or images meant to deceive.
Fabricated content – 100% false content intended to deceive and do harm.
These stories are increasingly dangerous in the digital age. Many people consume stories on social media without fact-checking or confirming that such headlines exist for a “shock” factor.
Once these stories are shared and popularized, people begin to believe them and accept them as the truth.
Jokes And Satire Are Not Fake NewsThey are intelligent satire pieces intended to be humorous – not real – and their entire sites are based around their readers being knowledgeable about this strategy and theme.
A Word On Mainstream NewsAccording to the 2023 Trust Barometer, an annual survey of global trust and credibility by Edelman Data & Intelligence, 50% of people distrust the media, and 46% view it as a divisive force in society.
A 2023 Gallup poll found Americans’ trust is even lower than the international average, with only 36% expressing a fair amount of confidence in mainstream media.
This varies by political affiliation, with 68% of Democrats trusting major news sources, as compared to 31% of independents and just 11% of Republicans.
However, most mainstream news sources have to undergo editorial reviews. Readers recognize them as being prestigious forms of journalism.
The Weather Channel, PBS, the BBC, CBS, ABC, NBC, and the Wall Street Journal were among the most trusted across ideological groups.
However, of these, only the Weather Channel was viewed favorably by conservatives, who trust Fox News and Newsmax as reliable sources of information.
How To Differentiate Between Fake News And Real NewsWhile it’s not easy, there are some things you can do to help determine whether a story is real or fake.
Here are some tips to help you differentiate between fake news and real news stories:
What Is The Site?Most of the fake news sources that go for “shock” value and produce fake stories are not as recognized. Look into the source itself and see whether it is a website that you can trust.
Check The DomainMany fake news stories use URLs and domain names similar to reputable news sources to mimic them.
Just like you should never enter personal information into your bank’s website without first verifying it’s not a phishing site, you should double-check the URL of the news sites in social media posts.
What Are The Authors’ Sources?Good news stories contain links to other reputable reporting by respected organizations.
They contain interviews with individuals who can confirm or deny they made a claim. They are supported by evidence, dates, and other data that can be fact-checked.
Be wary of sources that cannot substantiate their claims.
Sometimes a source may sound reputable in a piece, only for you to later find out no such person exists.
For example, the satire site Snickersee published an article about the return of bottlenose dolphins to Lake Michigan.
If you were not familiar with the site, at first glance seemed legitimate, even going so far as to quote a scientist with the Great Lakes Fish and Water Quality Society.
However, no such scientist or organization exists, and dolphins have never lived in the Great Lakes.
Fact Check!When in doubt, fact-check the information that you read. You can start with a simple search using the keywords or the event that is being reported.
You can also use sites like PolitiFact, FactCheck, and Snopes – which are incredibly reputable fact-checking sites for various issues and topics (not just politics).
Examine The Website CloselyIt is essential not to look at one story alone but the full spectrum of details on the site.
Is there another fake-looking or shocking headline? What does the overall website look like? How is the user experience?
Sometimes doing a little further digging will make it evident if a news story is fake.
However, this can be tricky as many fake news sites intentionally mislead visitors. They may look like the real deal, all the way down to the logo in the masthead.
Just remember sites like chúng tôi or chúng tôi have teams of web professionals on staff, including UX and UI specialists.
A reputable site is unlikely to be slow-loading, clunky, and cumbersome to use.
Likewise, glaring spelling or grammar mistakes, while not impossible, do not usually make it through the proofing and editing process.
Check WikipediaThis may be surprising, but Wikipedia is an excellent source for identifying fake news sites.
It maintains an ever-growing list of fake news sites. This list includes sites like American News, which have created disinformation shared via social media, as well as satire sites and those used to spread malware.
Fact Check With GoogleTo combat false information and aid fact-checkers, journalists, and researchers in their work, Google added Fact Check Tools to its toolbox in 2023.
This collection of more than 150,000 fact checks allows users to access information from “reputable publishers” around the world to ensure integrity.
To use this tool, enter a topic or person into the search bar.
Google will then pull information from its network of trusted publishers to give you at-a-glance fact-checking.
For example, despite rumors that Tesla CEO Elon Musk has invented a new digital currency, Google Fact Check is easily able to squash this.
Put Technology To WorkThere are several digital tools you can employ to combat fake news.
A list compiled by the RAND Corporation, an American global policy think tank, includes several digital tools for fighting online disinformation.
These include everything from browser extensions like BotSlayer, which helps track and detect information manipulated on Twitter, to verification and bias checking tools like NewsCheck Trust Index, to collections of resources that promote media literacy skills like NewseumED.
Read Past The HeadlineThis practice is particularly problematic on social media platforms, where studies have shown up to 59% of people will share articles without reading past the headline.
In response, both Facebook and Twitter have implemented features that push users to read articles before sharing them in their feeds.
Be Aware Of Your Own BiasesEveryone likes to feel smart. And everyone wants to have their opinions confirmed. That’s why some people will doggedly stick to their version of reality even in the face of overwhelming evidence that they’re wrong.
While you may not be that stubborn, everyone has implicit biases. And that’s why it’s imperative not to let what you want to believe override the facts.
Confirmation bias occurs when you put more stock into something that confirms what you already believe while minimizing information it doesn’t. Conservative-minded people trust right-leaning news sources, and liberals rely on liberal-minded platforms.
Help Combat The Spread Of False NewsAs long as there has been news, there have been bogus stories.
But whereas a malicious actor spreading false information about Alexander the Great during the Siege of Tyre would only be able to influence a few thousand people, social media has created an environment in which anyone can spread disinformation globally, potentially to millions of people.
According to Avaaz’s global activism nonprofit study, the top 100 most popular false or misleading stories about the 2023 U.S. presidential election had 162 million views.
By contrast, there are only 168 million registered voters in the country.
The same Avaaz report stated that an October 2023 poll found roughly 91 million registered voters saw misinformation about mail-in voter fraud on Facebook. And 35% of registered voters believed this.
Fake news is a significant problem in the cyber age. And at the end of the day, it doesn’t boil down to beliefs or opinions. Facts don’t change, no matter what your views are. Water is still wet, and ice cream is still cold, even if it that inconvenient for some people.
And while social media platforms are attempting to uncover solutions to this issue, much of the burden will ultimately fall on you, the user.
But this makes it essential that each person does their part to combat the spread of fake news.
You should approach every news article, video, or social media post skeptically. Even if the originator is a reliable source of information, at the end of the day, they’re only human.
And that means, just like everyone else, they can fall victim to incorrect information, inaccurate statistics, or quotes taken out of context.
Do not share fake news on your social channels. Even if you’re only sharing it for its sheer outlandishness, you’re increasing its visibility, so don’t make this mistake.
If your friends or family members are sharing disinformation, message them and let them know. And be sure to mention how you identified it as fake.
Hopefully, this will help others identify it as false news and avoid sharing it further.
Everyone has to do their part, but if we all do our part, we can at least stem the tide of this unfortunate phenomenon.
More Resources:
You're reading How To Identify Fake News From Real News Online
Are Facebook And Twitter Winning Against Fake News?
Facebook recently announced it is making strides against fake news. It cited several studies that independently concluded Facebook has dramatically decreased the reach of fake news by high percentages. But percentages don’t tell the whole story.
Facebook has indeed made great strides, as you will see below. Yet when you look at the numbers instead of percentages, a study reveals that fake news engagement on Facebook as of July 2023 was 70 million per month.
What’s going on at Twitter is another story.
Since the 2023 US Election, Facebook has announced twelve spam crackdowns while Twitter only announced five. The research shows that Twitter’s fight is not going so well.
According to Facebook:
As these studies attest, we’ve invested heavily in our strategy to fight misinformation since the 2023 US elections. We continue to roll out updates that address some of the outstanding gaps mentioned in these studies.
Methodology Lists of Fake News SitesThe study authors gathered lists of fake news sites as published by various sites and researchers, including the following:
Politifact
FactCheck
Snopes
BuzzFeed
European Research Council paper
Share and Engagement MetricsFacebook engagement and Twitter share metrics were collected from BuzzSumo.
Comparison with Real News SitesThe researchers also compiled a list of legitimate news websites, as well as sites from the arts, business, sports and other niches.
Results of Research StudyNews sites displayed an upward trend on both Twitter and Facebook. The trend on Facebook is clearly because of the changes in the news feed to emphasize news, especially local news.
Fake News TrendsFake news sites increased on Facebook and Twitter leading up to he 2023 election. After the election, engagement rates plummeted by 50% on Facebook while Twitter shares kept rising.
Facebook Engagement versus Twitter Shares“Facebook engagements fell from a peak of roughly 200 million per month at the end of 2023 to roughly 70 million per month at the end of our sample period.
As a point of comparison, the 38 major news sites… including the New York Times, Wall Street Journal, CNN, Fox News, etc.—typically garner about 200-250 million Facebook engagements per month.”
As you can see, Facebook’s metrics for fighting fake news are superior to Twitter’s. However, as of the end of the study in July 2023, Facebook was still serving up 70 million engagements with fake news per month.
The reach of fake news on Facebook is equal to 35% of all legitimate news websites combined. That’s a staggering amount of fake news that is still circulating on Facebook.
Facebook has announced several more spam crackdowns that affected fake news sites since the release of this study. So the amount of fake news reaching Facebook members today is quite likely less than the 35% reported by this study.
Is Making Strides Against Fake News Enough?The results are clear. Facebook is making strides in the fight against fake news. Twitter is achieving less success and taking less actions taken against fake news.
Yet with 70 million people engaging with fake news as of July 2023, Facebook still has more battles to fight.
Read Facebook’s announcement here.
News Roundup 31St May 2023
This week, Facebook has updated us on how it is enforcing its Community Standards with its third report. The report includes new data in a bid to make sure its enforcement is more transparent.
New research has revealed that brands who form emotional relationships with customers are likely to see better brand loyalty, showing that the human element and storytelling really do benefit companies.
Finally, Twitter has announced a new feature. Those who live stream will now have the opportunity to add audio-only guests to their video, which could be used in a number of ways.
Learn more about all of this with our news roundup.
Facebook releases third Community Standards Enforcement ReportFacebook has published its third Community Standards Enforcement Report, which covers Q4 2023 and Q1 2023. The report includes new data points to provide an improved understanding of the process that Facebook goes through when removing content.
The social media giant has now added data relating to content appeals and how much appealed content was restored. This is the first time that Facebook has included this information.
In addition to this data, it is also including data on regulated goods, detailing how the platform is performing when it comes to removing attempts to sell illicit goods, with a focus on drugs and firearms.
These additions to the report mean that the Facebook report now includes metrics across nine policies within its Community Standards. These include/ adult nudity and sexual activity, child nudity and sexual exploitation, bullying and harassment, fake accounts, regulated goods, hate speech, spam, global terrorist propaganda and violence and graphic content.
Appeals and correcting mistakesFacebook has admitted that its enforcement of Community Standards isn’t perfect and so it is making its process for dealing with any mistakes a bit more transparent with its report. The data will show how much content was restored following user appeals and how much it restored on its own.
The platform can restore content without an appeal for a number of reasons, including:
When multiple posts of the same content have been removed, Facebook can use one person’s appeal of this decision to restore all affected posts.
An error is identified during the review of removed content and so it is restored before an appeal takes place.
If a post is removed because it could feature a malicious link, it can be restored if Facebook learns that the link isn’t harmful.
Regulated goodsWhile Facebook has had policies about drug and firearm sales in place for years, this is the first time this information is being included in the report. While it has used a range of methods when dealing with this type of post previously, it introduced AI in the summer of 2023 to better identify content that violates its regulated goods policies.
According to Facebook, this has enabled it to take more action, usually before people have had to report content. The platform reports that it took action on around 900,000 pieces of content relating to illicit drug sales in Q1 of this alone. Of these, 83.3% of the content was detected proactively.
Facebook hopes that this extra information will help it fulfil its promise to become more transparent when it comes to how it develops its policies and how it measures its enforcement of them.
Emotional bonds with brands result in customer loyaltyBrands that are able to form emotional bonds with customers are more likely to have a high customer retention rate, according to new research. A study from Deloitte Digital has found that appealing to customer emotions directly relates to trust and repeat custom.
According to the findings, companies who appeal to customer’s emotions on four levels – data boundaries, connection, loyalty and consistency – have a better chance of developing meaningful and beneficial relationships.
Some 62% of the consumers surveyed said they felt a relationship with a brand. Of these, 76% said they have stuck with that brand for four years or more due to the fact they trust it. In fact, trustworthiness is the most important factor when it comes to brand favourability, with 83% of respondents selecting it.
After trustworthiness was integrity (79%) and honesty (77%), showing that transparency is a vital aspect of cultivating customer relationships. This means that building trust is the first step in cultivating lasting relationships with consumers.
Building this trust can then aid in the development of future relationships, as customers who trust a brand are more likely to recommend it. In fact, just under half (44%) of those surveyed said that they have endorsed a product based on emotional criteria. Of these, 60% said they use words like “happy”, “love” and “adore” when recommending brands they feel a connection to.
Timothy Greulich, experience management practice leader at Deloitte, said: “Customers and consumers are longing for human connection with the organizations and people they tend to do business with. When pressed on what those connections look like, they described their relationship in the context of a friendship, more importantly, a two-way relationship, where both parties are responsible and engaged in the health of the friendship/relationship.”
Another factor that influences customer trust is how brands use consumer data. The research found that three-quarters (75%) of customers want a brand to know why a purchase was made and over half (52%) want that brand to know they are satisfied with their purchase. However, 35% don’t want brands to search their browser history and 59% don’t want brands responding to social posts, whether they are negative or positive.
While issues like privacy breaches can cause an end to a consumer’s relationship with a brand, the way that a brand reacts to the situation can make a huge difference. Some 77% said they would continue buying from a brand if they are offered an apology that seems genuine when an issue arises.
This goes to show that perfection is not the key to creating and maintaining customer relationships. If a brand is able to appear human, trustworthy and honest while also holding their hands up when something goes wrong, it is likely to lead to lasting relationships that offer high ROI.
Facebook is bringing a new add type to its social media network that could provide a boost to mobile games. Facebook users will now be able to trial a mini version of a mobile game through its new ad format, allowing them to see if the game is for them before downloading it.
According to Facebook, people were 60% more likely to open a mobile game they had downloaded if they had the chance to test a playable ad first compared to those who downloaded without first testing the game.
Facebook will also be offering vertical video and zip file support, which will be available in the next few weeks.
Head of insights at Differentology, Dan Brilot said: “The mix of low-rolling monthly subscriptions, in addition to episodic content, means SVOD has unusually high levels of loyalty as a category, with one in six claiming to always go to an SVOD service first when deciding what to watch.
“The figure rises significantly for 16-to19-year-olds, with over half (53%) never going to live TV first, while 17% of the age groups claim the number of paid SVOD subscriptions in their household will increase over the next three years. New entrants, such as Britbox, seem set for success, provided the price point is in line with current SVOD providers.”
The research findings follow on from the recent announcement that Coca-Cola is integrating with Netflix’s Stranger Things season three, which will see the launch of New Coke, suggesting that this could be the direction that the streaming service is going.
Twitter launches new live streaming with guestsTwitter has announced its new live-streaming option that will mean users can add audio-only guests to their streams. Tests for the featured were spotted by some users earlier in May but the new Live Guests option is now being rolled out across the platform.
The feature means that viewers can join in a conversation if the streamer grants permission. People can join a stream with their audio only, with their avatar appearing on the screen. This feature could be used to host question and answer sessions, or interviews live on Twitter.
The platform said of the new streaming feature: “Conversations on Twitter can take place in many ways, and our latest update to live video brings another dimension to how you can discuss what’s happening. Starting today, your Tweets are going £IRL and now you can hot a live video and invite up to three people as guests. Those you allow to join can be heard by everyone and can drop off at any time.”
While the option is a good step forward for Twitter, it seems to be slightly behind other social media networks. Instagram and Facebook Live already allow split-screen live streams with a guest, which is something Twitter doesn’t allow. However, the ability to invite up to three guests could be enough of a difference to bring people to this feature instead.
Inform – News Aggregation Supersite Launches
Inform – News Aggregation Supersite Launches
Today next-generation news aggregation site Inform unveiled itself. I spoke to CEO Neal Goldman a couple of weeks ago and got a demo of the site, which was very impressive. I don’t have a great deal of time this morning to run through the site’s features, but here’s what makes it different:
Personalization: There’s a lot now and more coming. You have to register for the personalization features, which include the ability to save articles and create a personalized directory of content or sources/feeds. There’s no Furl/MyWeb capability, however, which would be a very nice feature.
The blog content aggregation is much more limited than the news content right now. It’s not a blog search engine per se. However, it does contain considerable blog content from top news-oriented blogs. And users can add any site/feed they want so it becomes, effectively, a substitute for a newsreader.
Stories that span multiple pages are presented in their entirety in a single window (the site doesn’t send readers back to the original source sites). Ads appearing on those sites also appear with the news content. (Subscription/registration required content is gated and presented as it would appear if one navigated directly to the content site.)
How does this affect newspaper sites?
(There’s a long digression here that I won’t go into about the value of newspaper brands and how to maintain them over time.)
This is a very powerful tool and, in a way, the real fulfillment of the “personal newspaper” that was discussed hypothetically years and years ago, as the way people would be reading news in the future.
There are critics who think it’s too complex and confusing (paidContent.org, for example). I might agree that’s there’s a lot here and it’s not fully intuitive for mainstream consumers. But I think these guys wil get that—I had a discussion with them about duplicative personalization elements—and simplify over time.
The newspapers (or other news media) are going to ultimately be compelled to buy this site. It’s just a question of how long do they wait.
Seo For News Publishers: Your Next Must
This post was sponsored by News & Editorial SEO Summit (NESS). The opinions expressed in this article are the sponsor’s own.
Search engine optimization has varying tactics across industries, especially for news publishers.
Our SEO efforts differ from those of, say, ecommerce or local.
We know it’s not a one-size-fits-all approach.
That’s why the News & Editorial SEO Summit (NESS) is perfect for you!
On October 4 & 5, you can get access to the best minds in the SEO and publishing industries who will be sharing their insights and expertise.
You’ll discover:
The most effective strategies for news content and evergreen SEO.
What drives Google’s quality-based algorithms in News and Discover.
The latest on AMP and technical SEO for publishers.
How to use analytics data to empower your digital growth.
Brought to you by chúng tôi one of the most engaged news SEO communities on the web, this event provides you with crucial information about organic search strategies that will help grow your traffic.
Tap Into A Wealth Of News SEO ExperienceNESS 2023 will feature speakers from established publications as well as some of the top experts in search today.
NESS 2023 SpeakersGet new, current SEO information, content, tips, and tricks from the greatest SEO minds in the news industry:
Claudio E. Cabrera, The Athletic: Live Journalism, SEO Strategies & Tactics.
Leonie Roderick & Ben Dilks, The Times: What do Paywalls mean for SEO?
Mark Burka, Angi: Commerce & Affiliate SEO for Publishers.
Carolyn Shelby, Dawn Patrol LLC: Using Automated Articles Successfully.
Alli Berry, TheStreet: Content Workflows for SERP Domination.
Koray Tuğberk GÜBÜR, Holistic SEO & Digital: How Search Engines Leverage Opinion-based Articles for Ranking.
John Shehata, Condé Nast, NESS & NewzDash Founder: Top Stories State of the Union.
Barry Adams, Polemic Digital & NESS Founder: The Latest in Technical SEO for News Websites.
NESS 2023 PanelsIn addition to our 10 individual summit events, you’ll have a chance to ask your more specific news SEO questions at our two growth-oriented panel discussions:
Ask The SEOs: Where you can ask this panel of experts, Claudio E. Cabrera, Carolyn Shelby, Koray GÜBÜR, Barry Adams, and John Shehata, any questions about their website’s SEO.
Career Growth: Hear from Lily Ray, Chris Moran, Paul Shapiro, Louisa Frahm, Barry Adams, and John Shehata, as they talk about their career paths in SEO and news publishing.
After attending this event, you will be able to apply what you’ve learned to your websites to boost your organic traffic.
Use “SEJ22” To Save 25% On Your Ticket →
Make sure your on-page and technical SEO knowledge is up to date. With all the layoffs and surprise algorithm updates, upskilling and networking is now more critical than ever.
Ask Your News SEO Questions: Entirely Online & LiveAll sessions are going to be live, and recordings will be available only to ticket holders after the event.
We will be mimicking an actual live event where you get great opportunities to hang out with the experts, talk to them one-on-one, and power up your connections and knowledge with networking in the booths!
With a single ticket, you will get full access to all talks over both days and have the opportunity to ask your questions directly to our speaker panel at the closing of the event.
What Is The News & Editorial SEO Summit?Even though there are many SEO communities and summits, there hasn’t been an event dedicated to news publishers or SEO professionals who work with news sites.
Visibility in Google’s ecosystem is a crucial source of readers for all online publishers, and information about how to maximize this can be hard to find.
The News and Editorial SEO Summit (NESS) is here to address the unique challenges that the news industry faces regarding SEO.
This second annual, live online event will provide expert insights, direction, and priceless networking opportunities with the best minds in publishing and SEO.
From Google News to Discover, from Top Stories to news apps, you will learn what it takes to grow your presence in all organic locations where news is shown.
A Summit Made For News PublishersNESS is great for:
Journalists and editors involved with the day-to-day writing and publishing of news content who want to make sure their stories get the best chance of ranking in Top Stories and Google News.
Web developers who want to make sure their websites adhere to Google’s latest technical requirements and follow SEO best practices for crawling and indexing.
SEO professionals working with publishers who want to upgrade their knowledge and learn from experts in the field.
Audience growth strategists looking for ways to maximize traffic and find new avenues for organic search visits.
Claim your space and buy a ticket to the second annual NESS summit today!
How Neural Network Can Identify Hidden Vibrations From Earthquake Data?
In the past many years, scientists have invented and discovered various techniques to understand and map the structure of the Earth. Through this, they can identify resources such as oil reserves, geothermal sources, and, more recently, reservoirs where excess carbon dioxide could potentially be sequestered. While tracking the seismic waves that are produced naturally by earthquakes or artificially via explosives or underwater air guns, scientists can get an idea of the type of structures that lie beneath the surface. There is a narrow range of seismic waves — those that occur at low frequencies of around 1 hertz — that could give scientists the clearest picture of underground structures spanning wide distances. But these waves are often drowned out by Earth’s noisy seismic hum, and are therefore difficult to pick up with current detectors. Specifically generating low-frequency waves would require pumping in enormous amounts of energy. For these reasons, low-frequency seismic waves have largely gone missing in human-generated seismic data.
In the past many years, scientists have invented and discovered various techniques to understand and map the structure of the Earth. Through this, they can identify resources such as oil reserves, geothermal sources, and, more recently, reservoirs where excess carbon dioxide could potentially be sequestered. While tracking the seismic waves that are produced naturally by earthquakes or artificially via explosives or underwater air guns, scientists can get an idea of the type of structures that lie beneath the surface. There is a narrow range of seismic waves — those that occur at low frequencies of around 1 hertz — that could give scientists the clearest picture of underground structures spanning wide distances. But these waves are often drowned out by Earth’s noisy seismic hum, and are therefore difficult to pick up with current detectors. Specifically generating low-frequency waves would require pumping in enormous amounts of energy. For these reasons, low-frequency seismic waves have largely gone missing in human-generated seismic data. In order to bridge this gap, MIT researchers have come up with a machine learning approach. According to a paper published in journal Geophysics, the researchers have described a method in which they trained a neural network on hundreds of different simulated earthquakes. According to MIT News, when the researchers presented the trained network with only the high-frequency seismic waves produced from a new simulated earthquake, the neural network was able to imitate the physics of wave propagation and accurately estimate the quake’s missing low-frequency waves. The innovation could allow researchers to artificially synthesize the low-frequency waves that are hidden in seismic data, which can then be used to more accurately map the Earth’s internal structures. Paper co-author Laurent Demanet, professor of applied mathematics at MIT said, “The ultimate dream is to be able to map the whole subsurface and be able to say, for instance, ‘this is exactly what it looks like underneath Iceland, so now you know where to explore for geothermal sources. Now we’ve shown that deep learning offers a solution to be able to fill in these missing frequencies.” Demanet’s co-author is lead author Hongyu Sun, a graduate student in MIT’s Department of Earth, Atmospheric and Planetary Sciences. Sun and Demanet incorporated a neural network for signal processing, specifically, to recognize patterns in seismic data. According to them, if a neural network was fed enough examples of earthquakes, and the ways in which the resulting high- and low-frequency seismic waves travel through a particular composition of the Earth, the network should be able to, as they write in their paper, “mine the hidden correlations among different frequency components” and extrapolate any missing frequencies if the network was only given an earthquake’s partial seismic profile. Demanet says, “If a network has seen enough examples of how to take a male voice and transform it into a female voice or vice versa, you can create a sophisticated box to do that. Whereas here we make the Earth speak another frequency — one that didn’t originally go through it.” However, the neural network has its own limitations where it is only as good as the data that are fed into it. If a new input is wildly different from the bulk of a network’s training data, there’s no guarantee that the output will be accurate. Moreover, to address this challenge, the researchers say they plan to introduce a wider variety of data to the neural network, such as earthquakes of different strengths, as well as subsurfaces of more varied composition. “Using this neural network will help us find the missing frequencies to ultimately improve the subsurface image and find the composition of the Earth,” Demanet says.
Update the detailed information about How To Identify Fake News From Real News Online on the Hatcungthantuong.com website. We hope the article's content will meet your needs, and we will regularly update the information to provide you with the fastest and most accurate information. Have a great day!