Predictive Analytics: Revolutionizing Fraud Detection and Prevention

Predictive analytics is revolutionizing the way businesses detect and prevent fraud. By leveraging the power of advanced algorithms, machine learning, and big data, organizations can now predict and identify potential fraudulent activities before they even occur. This innovative approach to fraud detection and prevention is transforming industries and helping businesses protect their assets, customers, and reputation.

Fraud is a pervasive problem that affects businesses of all sizes and industries. In fact, according to the Association of Certified Fraud Examiners (ACFE), organizations lose an estimated 5% of their annual revenues to fraud. This amounts to a staggering $3.7 trillion in global losses. Traditional methods of fraud detection, such as manual reviews and rule-based systems, are no longer sufficient to combat the ever-evolving tactics of fraudsters. As a result, businesses are turning to predictive analytics to stay one step ahead of the criminals.

Predictive analytics involves the use of statistical techniques, machine learning algorithms, and data mining to analyze historical data and make predictions about future events. In the context of fraud detection, this means analyzing large volumes of transactional data to identify patterns and anomalies that may indicate fraudulent activity. By doing so, businesses can quickly detect and prevent fraud, minimizing financial losses and reputational damage.

One of the key advantages of predictive analytics is its ability to adapt to new and emerging fraud schemes. Traditional rule-based systems rely on pre-defined rules to identify potential fraud, which can be easily circumvented by savvy criminals. In contrast, predictive analytics models can automatically learn from new data and adjust their predictions accordingly. This makes them far more effective at detecting and preventing fraud, even as fraudsters continue to develop new tactics.

Another benefit of predictive analytics is its ability to reduce false positives. False positives occur when legitimate transactions are flagged as potentially fraudulent, leading to unnecessary investigations and customer dissatisfaction. Predictive analytics models can be fine-tuned to minimize false positives, ensuring that businesses can focus their resources on investigating genuine cases of fraud.

The financial services industry is one of the sectors that has been most significantly impacted by the adoption of predictive analytics for fraud detection and prevention. Banks, credit card companies, and other financial institutions are using predictive analytics to monitor transactions in real-time, flagging suspicious activity for further investigation. This has led to a significant reduction in losses due to fraud, as well as improved customer satisfaction and trust.

The insurance industry is another area where predictive analytics is making a significant impact. Insurers are using predictive models to identify potential cases of fraud, such as staged accidents or exaggerated claims. By detecting these fraudulent activities early, insurers can reduce their losses and maintain the integrity of their risk pools.

In addition to financial services and insurance, predictive analytics is being used to combat fraud in a wide range of industries, including healthcare, retail, and telecommunications. As businesses continue to generate and store vast amounts of data, the potential applications of predictive analytics for fraud detection and prevention will only continue to grow.

In conclusion, predictive analytics is revolutionizing the way businesses detect and prevent fraud. By leveraging advanced algorithms, machine learning, and big data, organizations can now predict and identify potential fraudulent activities before they even occur. This innovative approach is transforming industries and helping businesses protect their assets, customers, and reputation. As fraudsters continue to develop new tactics, predictive analytics will remain a critical tool in the ongoing battle against fraud.

View original post here:

Predictive Analytics: Revolutionizing Fraud Detection and Prevention - CityLife

from the this-is-embarrassing dept

Weve written a lot about AB 2273, Californias Age Appropriate Design Code (AADC) that requires websites with users in California to try to determine the ages of all their visitors, write up dozens of reports on potential harms, and then seek to mitigate those harms. Ive written about why its literally impossible to comply with the law. Weve had posts on how it conflicts with privacy laws and how its a radical experimentation on children (ironically, the drafters of the bill insist that theyre trying to stop experimentation on children).

Weve also written about how NetChoice, an internet company trade group, has sued to block the law as unconstitutional, and how I filed a declaration explaining how the law would violate the rights of both us at Techdirt and our users.

That lawsuit has continued to move forward, with California filing a pretty laughable reply saying that it doesnt regulate speech at all. NetChoice has filed its own reply as well, highlighting how ridiculous that is:

The State claims that AB 2273 regulates data managementnonexpressive conduct, Opp. 11not speech. Nonsense. AB 2273s text expressly requires services to mitigate or eliminate risks that a child could encounter potentially harmful content online. Content was the through-line in the legislative process: Defendant Attorney General Bonta praised the Act precisely because it would protect children from harmful material and dangerous online contentin other words, speechand Governor Newsom lauded the law for protect[ing] kids from harmful content. The States own expert, who mentions content in her declaration 71 times, derides preexisting laws specifically because they only cover data management, not content. Radesky Decl. 98. The State cannot evade the Constitution by pretending the Act regulates only business practices related to the collection and use of childrens personal information, Opp. 11, when the laws text, purpose, and effect are to regulate and shape online content. Like Californias last attempt to restrict the ideas to which children may be exposed, Brown v. Ent. Merchs. Assn, 564 U.S. 786, 792, 794 (2011), AB 2273 violates the First Amendment

It appears that Governor Newsom may have realized how badly this case is going to go for him. Days after NetChoice filed that reply, Newsom sent NetChoice an angry letter demanding that it drop the case.

The text is quite remarkable and bizarre. Newsom sounds angry. Perhaps because he realizes (per the above) that his own words in support of the bill and how it should be used to block content are going to make him lose this case.

Enough is enough. In light of new action and findings released by the U.S. Surgeon General, I urge you to drop your lawsuit challenging Californias childrens online safety law.

Except, as we just detailed, the Surgeon Generals report does not find that the internet harms kids, and actually makes it clear that most kids benefit from social media. Straight from the report that it appears Newsom did not read:

A majority of adolescents report that social media helps them feel more accepted (58%), like they have people who can support them through tough times (67%), like they have a place to show their creative side (71%), and more connected to whats going on in their friends lives (80%). In addition, research suggests that social media-based and other digitally-based mental health interventions may also be helpful for some children and adolescents by promoting help-seeking behaviors and serving as a gateway to initiating mental health care.

But, Newsom appears to have only read the headlines that misconstrue whats in the actual report. His letter then goes into full on moral panic mode:

Every day as our children browse the internet to connect with one another, build community, and learn, they are also pushed to horrific content and exposed to data mining and location tracking. This reality is dangerous to their safety, mental health, and well-being. Thats why, last September, I was proud to sign the California Age-Appropriate Design Code Act a bipartisan, first-in-the-nation law that protects the health and privacy of children using online platforms and prohibits online services from encouraging children to provide personal information.

Except, nearly everything in that paragraph is wrong. Embarrassingly so. There is no evidence that children are pushed to horrific content. It is true that there may be horrific content online, but the idea that companies are pushing kids to that content is not supported by the evidence. Furthermore, its rich that hes complaining about data mining and location tracking while saying that this bill prohibits companies from seeking personal information from kids when the laws age assurance requirements suggest the exact opposite. To comply with the law, websites will be effectively required to demand information from users to determine a likely age.

As I explained in my own declaration in the lawsuit, at Techdirt we have bent over backwards to learn as little about the folks who read our site as possible. But under the law, we will likely be compelled to institute a program in which we are required to determine the age of everyone who visits. In other words, the law requires more data mining, not less, and explicitly requires it for children.

Newsom continues the nonsense:

Rather than join California in protecting our children, your association, which represents major tech companies including Google, Meta, TikTok, and Twitter, chose to sue over this commonsense law. In your lawsuit, you have gone so far as to make light of the real harms our children face on the internet, trivializing this law as just being about teenagers who say unkind things, insufficiently like anothers posts, or are unhappy about the omission of a trigger warning.

Again, nothing in this law actually protects children. Instead, it puts them at much greater risk of having information exposed, as weve noted. It will also make it next to impossible for children to research important information regarding mental health, or to find out the information they need to help them deal with things like eating disorders, since it will drive basically all of that content offline (at least where kids can reach it).

As for the claim that NetChoice is trivializing this law, thats obviously bullshit to anyone who has read the filings in context (which apparently does not include this angry Governor Newsom). The references in that paragraph are in NetChoices motion for a preliminary injunction, but taken completely out of context. Theyre not trivializing the issues children face: theyre pointing out that the way the law is drafted (i.e., very, very badly), it also applies to those more trivial situations. From the preliminary injunction filing:

AB 2273 also adopts a boundless conception of what speech must be restricted, including speech that cannot constitutionally be restricted even for minors. The requirement that services enforce their own policies, id. 1798.99.31(a)(9), will lead them to suppress swaths of protected speech that the State could not restrict directly. See supra IV.A.1.b. The bar on using algorithms and user information to recommend or promote content will restrict a providers ability to transmit protected speech based on the users expressed interests. And the laws restrictions on content that might be detrimental or harmful to a childs well-being, id. 1798.99.31(a)(1)(b), (b)(1), (3)-(4), (7), could restrict expression on any topic that happens to distress any child or teen. This would include a range of important information children are constitutionally entitled to receive, such as commentary or news about the war in Ukraine, the January 6, 2021 insurrection at the United States Capitol, the 2017 Unite the Right rally in Charlottesville, school shootings, and countless other controversial, significant events.

More fundamentally, the harm the law seeks to addressthat content might damage someones well-beingis a function of human communication itself. AB 2273 applies to, among other things, communications by teenagers on social media, who may say unkind things, insufficiently like anothers posts, or complain harshly about events at school; the use of language acceptable to some but not others; the omission of a trigger warning; and any other manner of discourse online. See, e.g., Mahanoy Area Sch. Dist. v. B. L., 141 S. Ct. 2038 (2021) (Snapchat post fuck cheer made high school students visibly upset)

So, no. The lawsuit is not trivializing harms children face by saying that its nothing more than kids saying unkind things, NetChoice is (accurately) pointing out that the broad language of the law means that it could be applied to those situations, rather than ones dealing with actual harm.

Its pathetic and embarrassing that Newsom would imply that this paragraph was trivializing harms. His complete and total misread of whats in the lawsuit is trivializing the seriousness of his states own law that is violating 1st Amendment rights.

Anyway, Newsom goes on:

Yet at the same time you are in court callously mocking this law, experts are confirming the known dangers of online platforms for kids and teens: Just days ago, the U.S. Surgeon General issued an advisory on the profound toll that social media takes on kids and teens mental health without adequate safety and privacy standards. Your association and its members may be interested to learn of the Surgeon Generals urgent findings about the sexual extortion of our children, and the alarming links between youth social media and cyberbullying, depression, suicide, and unhealthy and dangerous outcomes and behaviors.

Honestly, this is making me wonder if Newsom ever reads anything. Because, as we discussed that is not what the Surgeon Generals report says at all. It literally says that there are widespread benefits to social media and then says we do not have enough evidence regarding whether or not its harmful. It notes there are concerns, and some correlational studies, but nothing proving a causal link. It notes that we need more research on that point.

So how the hell is Newsom claiming that it is claiming there is a profound toll from social media? The report does not say that.

As for the Surgeon Generals urgent findings about the sexual extortion of our children, again Newsom is blatantly misstating what the report says. It notes that the internet has been used for sexual extortion, which is a fact, but nothing in the AADC will stop bad people from being terrible. The report does not say anything about this fact being urgent or requiring social media companies to magically make people stop being bad. It just mentions such things as the kind of problematic content that exists online.

As for the alarming links between youth social media and cyberbullying, depression, suicide, and unhealthy and dangerous outcomes and behaviors thats AGAIN misreading the Surgeon Generals report. Again, it does mention those things, but does not discuss alarming links. It highlights correlational concerns again, and suggests further research and caution. But does not mention any sort of causal link, alarming or not.

In fact, with regards to cyberbullying, the Surgeon Generals recommendations talk about better educating teachers, parents, and children on how to deal with such things. And, its one policy recommendation around cyberbullying is not to force websites to censor content, as the AADC does, but rather to support the development, implementation, and evaluation of digital and media literacy curricular in schools and within academic standards.

In other words, what the Surgeon General is kinda saying is that our policy makers are the ones who have failed our kids by not teaching them how to be good digital citizens.

Governor Newsom, that ones on you.

So, so far we have Newsom lying about the law, lying about the filings from NetChoice, and now lying about the Surgeon Generals report. I know its a post-truth political world we live in, but I expect better from Californias governor.

But hes not done yet:

The harms of unregulated social media are established and clear.

The Surgeon Generals report not to mention the even more thorough report from the American Psychological Association literally say the opposite. They say it is not clear, and much more research needs to be done.

Governor Newsom, you should stop lying.

It is time for the tech industry to stop standing in the way of important protections for our kids and teens, and to start working with us to keep our kids safe.

Stomping on 1st Amendment rights and lying about everything is not keeping our kids safe Governor.

Filed Under: aadc, ab 2273, age appropriate, data, gavin newsom, moral panic, social media, studies, surgeon general, teenagersCompanies: netchoice

See the original post:

Governor Newsom Desperately Begs NetChoice To Drop Its Lawsuit ... - Techdirt

Researchers have discovered a treasure trove of arthropods, such as these, on the sea floor in the ClarionClipperton Zone, located in the central and eastern Pacific Ocean.Credit: SMARTEX Project, Natural Environment Research Council, UK (smartexccz.org)

A company is expected to request authorization in July, for the first time ever, to mine the ocean floor for metals such as cobalt and nickel. At the same time, researchers warn that a crucial database that maps deep-sea biodiversity and that could factor into the decision to approve such a licence contains errors and data gaps.

Seabed mining is coming bringing mineral riches and fears of epic extinctions

The International Seabed Authority (ISA), a body associated with the United Nations that oversees deep-sea mining in international waters, currently allows only mining exploration. According to its website, it has approved 17 companies and government entities to study the mining potential of the ClarionClipperton Zone (CCZ), a region of the sea floor that spans up to 6 million square kilometres of the central and eastern Pacific Ocean and that holds metal-rich clumps of sediment. Nauru Ocean Resources, a subsidiary of The Metals Company, based in Vancouver, Canada, has been exploring the sea bed, with an eye towards gathering metals needed for electric-vehicle batteries and other electronics. It plans to apply for a commercial mining licence in a month or so. If approved, operations could begin in 2024.

Scientists worry about allowing companies to start mining the sea bed because little is known about deep-sea habitats and biodiversity, so its environmental effects are unpredictable.

The ISA runs a database called DeepData, which is meant to tackle some of these concerns, as well as to enable research projects. The database contains information that the ISA requires contractors to collect during their deep-sea exploration missions. These biological, geochemical and physical data include, for example, the species that they encounter and the chemicals present in the water.

But the analysis of DeepData, published in the journal Database on 30 March1, revealed flaws that worry the researchers who conducted the study.

Contractors would like to mine the sea bed for metal-rich clumps of sediment called polymetallic nodules.Credit: Courtesy of the NOAA Office of Ocean Exploration and Research, 2019 Southeastern U.S. Deep-sea Exploration.

It strikes me as irresponsible to be relying on the database in its current form to assess the impact of mining on the sea-floor environment, says Muriel Rabone, a data scientist at the Natural History Museum in London, who led the analysis. Rabone told Nature that the analysis was performed independently of the ISA, but that the agency cooperated to enable data access. It was also consulted on the scope of the study and an early draft of the manuscript.

The ISA protests some of the findings, however, saying that the report is out of date. On 12 July 2021, the researchers downloaded data collected in the CCZ to run their analysis. Since then, the ISA has made significant improvements to address quality assurance and control issues with DeepData, it says.

Responding to this criticism, Rabone maintains that the database still contains flaws. Even with its faults, its helping to point to thousands of species on the sea floor that had never been seen before results published just this week. There is work to do yet, she says.

Of the 40,518 records that the researchers analysed for the Database study, about one-quarter were duplicates, which could lead to an underestimation of species richness in the deep sea, they say. The scientists think duplicates can arise partially because the database lacks unique codes to identify individual records.

The ISA says that, like any database, DeepDatas features and the quality of its data are improving with the years due to technological advances. It adds that it has identified and corrected duplicate records. Also, it is collaborating with the World Register of Marine Species, which catalogues and classifies marine organisms, and is sharing data with the Ocean Biodiversity Information System a data hub that has helped to clean up the data and make them more widely available.

Brisingid sea stars, like this one, live on the sea floor in areas rich with metals.Credit: Courtesy of the NOAA Office of Ocean Exploration and Research, 2019 Southeastern U.S. Deep-sea Exploration

Looking at the database today, however, Rabone says that some duplicate data still exist, and that many records still do not have a unique identifier.

The team also found that DeepData contained inconsistent information for instance, records that catalogued two species under the same name. And a lot of environmental data were missing. When contractors submit their data, they use a form with fields such as species name and fauna class size. The researchers found that 90% of the total data in various fields were missing.

The ISA says it has already updated its forms to address some of these issues and is designing workshops and training for contractors to ensure that data quality and control are improved.

Scientists track damage from controversial deep-sea mining method

Rabone would like the workshops to be open to the scientific community, which she says can provide feedback on the database. Stefanie Kaiser, a deep-sea ecologist at Senckenberg Research Institute in Frankfurt, Germany, who was not involved with the study, agrees, and says that, if the database were improved, it could be useful for researchers, giving them access to all the information collected by the contractors.

But the ISA says workshops are for only contractors, because they provide the data, although it acknowledges that the academic community has assisted contractors with presentations and preparing annual reports.

Despite the disagreements over DeepData, researchers are already learning from the database. Rabone formed an official partnership with the ISA to lead the first census of metazoan biodiversity on the CCZs sea floor. The endeavour found more than 5,500 species in the region, of which 92% are new to science, including many worms and arthropods. The findings were published on 25 May in the journal Current Biology2.

Read the original post:

Major ocean database that will guide deep-sea mining has flaws ... - Nature.com

Artificial Intelligence (AI) has exploded in the public mind over the last year, with many examples of AI-generated text, artwork and even full video. The talk of AI replacing entire industries is back at the front of the conversation with lawyers and accountants often mentioned as vulnerable.

Artificial intelligence is not going to replace experienced professionals right now, but people who can use AI effectively will quickly outpace those who cant. AI is changing how many people work and its here to stay. Your clients will need your AI insights, and your organization may need your AI knowledge.

So, how do you make sure it doesnt leave you behind?

You may find that you actually know more about AI than you thought you did that youve been using AI, every day, for many years. This article will explain the most common AI terminology you need to know to engage in the AI conversation in your workplace.

Artificial intelligence, or AI for short, is broadly defined as a branch of computer science that aims to develop intelligent machines, which can perform tasks that typically require human intelligence. This includes learning, problem-solving, decision making, etc. Examples of AI from everyday life range from maps and navigation to text editors and autocorrect to chatbots and digital assistants, like Siri.

An algorithm is a set of rules to be followed in calculations or other problem-solving operations, especially by a computer. In a math equation, an algorithm is the method we use to solve long division problems. A search engine, like Google, uses algorithms to find the most relevant information for the searcher.

Machine learning (ML) is a subset of AI that can learn without following explicit instructions by inferring patterns in data using statistical models and algorithms. Examples of ML include social media feeds, product recommendations, and image recognition.

Natural language processing (NLP) focuses on generating human language both spoken and written not robotic speech or restrictive text. Natural language processing applies algorithms to extract and analyze language data in a way that computers can process.It is imperative for machines to be able to process enormous amounts of data to be able to mine it and organize it and ultimately, to translate it and output human-seeming content.

Natural language search (NLS) is a type of search method that allows users to interact with a computer system or search engine using everyday language instead of formalized search queries or specific commands.Natural language search means that when youre searching for a new gym by searching for gym, your results will include most places that are focused on fitness, regardless of whether the name of the business actually includes the word gym. From a traditional gym, a CrossFit or yoga studio, Google understands that gym and studio in this instance have a similar meaning.

Your search inquiry doesnt have to be all inclusive (gym and studio and fitness and yoga and CrossFit and health and club) to get all-inclusive results. You get to type like a human, not a robot. And with a little help from machine learning, its going to keep your results local.

As a professional, it also means that when youre searching through research documents and briefs, you dont always have to use exact-match language to find precisely what youre looking for.

Data mining is the process of looking for relationships, correlations, and patterns within large data sets. Technology systems scour data and recognize anomalies within the data at a scale that would be impossible for humans. This analysis helps predict outcomes, finds potential wrongdoings, and notices questionable trends, and that information derived can be useful in a variety of ways.

Your recommendations (hopefully) keep getting better.By analyzing the patterns of people who also buy or are interested in the same products as you, a store can make relevant suggestions based on that data. This same concept plays out in Netflix recommendations or targeted advertisements online.

To put it simply,structureddata isorganizeddata, defined within a particular structure. It may be referred to as quantitative data. It is objective and easy to export to and store in Microsoft Excel or a larger database. The way it is organized is consistent and easily identifiable, which makes data mining better. Structured data is also less complicated to analyze and distill.

On the other hand, unstructured data isnt organized. It has no externally defined structure, and cannot be easily exported, stored, or organized. And its the bulk of what most organizations deal with daily. It includes most text-heavy data, such as reports, Microsoft Word documents, emails, and webpages.

Structured data has made it easy for you to complete searches and inquiries for decades.And because the data is organized and objective, you can be sure that the results you are shown are the most accurate.

For example, transactional data from a sales report e.g. Rep X sold Y units of product Z, for a total revenue this year of $$ is structured data, and easily analyzed. But that same reps detailed listing of feedback from new users of the product during implementation is unstructured, and has traditionally been difficult to examine, analyze, and quantify.

Recent advances in AI, such as Large Language Models and Foundation Models, have largely been about using vast repositories of unstructured data in new ways. See the companion article on machine learning for more details.

Big data refers to the data sets that are too large or complex to be handled by traditional data-processing software. Big data is a combination of structured, semi-structured, and unstructured data. Examples of big data include customer databases, all the information posted on a social media site, or trade data from the New York Stock Exchange.

Everything discussed above is generally well-understood and has been in use in many professional products for years decades in some cases. Remember the first time you typed a plain-language search to find what you needed? That was a type of AI. You should be relatively comfortable with these AI concepts, since youve been using them for years in your daily life, probably without even realizing.

But AI is not standing still. The consumer versions of Large Language Models have been all over the news, specifically ChatGPT and other chatbots. These AI technologies have the potential to disrupt the work of all professionals and you should be in the know. To stay on top of recent AI developments, visit our hub on artificial intelligence and read our companion article on generative AI and chat bots to learn more.

Read more:

Artificial intelligence terms professionals need to know - Thomson Reuters