predict
Latest
Researchers say Oura rings can predict COVID-19 symptoms three days early
Researchers believe the Oura Ring and AI-guided models can predict COVID-19 symptoms three days before onset with over 90 percent accuracy.
New wristband could predict aggressive outbursts in people with autism
Researchers are hoping a new wearable wristband will help predict aggressive outbursts in people with autism. The device monitors heart rate, sweat production, skin surface temperature and arm movements. It can predict outbursts 60 seconds ahead of time with 84 percent accuracy. While that might not seem like much time, it could give caretakers an opportunity to relax the individual wearing the wristband and make sure everyone is safe.
DeepMind AI can predict kidney illness 48 hours before it occurs
Acute Kidney Injury (AKI) kills 500,000 people in the US and 100,000 in the UK annually, often because it's not detected soon enough. Researchers want to use AI to change that. DeepMind, the Alphabet-owned AI company, partnered with the US Department of Veterans Affairs (VA) to develop an AKI prediction algorithm. In a paper published in Nature today, the partners share their findings that the algorithm can predict the presence of AKI up to 48 hours before it happens. The model correctly identified 9 out of 10 patients whose condition worsened to the point that they needed dialysis.
London police raise privacy hackles with gang violence software
London's Metropolitan Police Service (MPS) has completed a 20-week study in a bid to more accurately predict whether specific gang members are likely to commit violence. The software, developed by Accenture, pulls data together from systems already used by the MPS and runs it through an analytics engine. It's looking at geography, past offenses, associations with other criminal and even social media postings. An MPS spokesman told the BBC that the system would record and analyze threats or negative comments made by gang members on social media, for example.
MIT-created algorithm predicts likelihood of running red lights, places blame where appropriate
The bad news: Math will always judge you. The good news: It'll still be there to judge everyone else. In a recent article published in the IEEE Transactions on Intelligent Transportation Systems, MIT's Professor Jonathan How and three colleagues announced they've created an algorithm that can predict whether an oncoming car is about to run a red light one or two seconds before a possible collision. The algorithm can compute the likelihood of a vehicle running a red light based on its rate of deceleration as it is approaching the intersection with a level of precision down to mere milliseconds. The team, which applied the algorithm to more than 15,000 vehicles during the study, used instruments that monitored vehicle speeds and locations as well as when the lights turned red. When the results were tallied, they found that they were able to correctly predict who would run a red light 85 percent of the time. In other news, MIT is working on a much simpler algorithm capable of predicting when your significant other will break up with you, the formula factoring in at least four behavioral elements from the last season of "Jersey Shore."
Cambridge researchers tout new location-based method to predict friends on social networks
Friend suggestions on social networks may already be a little too eerily accurate for some, but a team of researchers from Cambridge University now say they can do one better. They've devised a method that doesn't simply rely the usual friends-of-friends approach, but on where those people tend to hang out. According to researcher Salvatore Scellato, "it turns out that the properties of the places we interact can determine how likely we are to develop social ties," and that places like offices and gyms are better indications of potential friends than football stadiums or airports. That notion was borne out in their research (conducted over a period of four months using Gowalla), which found that "about 30 percent of all new social links appear among users that check-in to the same places." With the two prediction methods combined, the researchers say they're able to account for 66 percent of all new social ties. No word if they've moved onto predicting crimes next. [Image credit: Gowalla]
How accurate are your favorite video game analysts?
Man, being a video game analyst must be a sweet job. You can put out all sorts of crazy predictions and no one will ever call you on it if you end up being wrong, right? Well, yes, right, until today, because today Kotaku has completed a massive project to rank the truthiness of predictions from some of the most prominent video game industry analysts out there.The results show that even the best prognosticators out there aren't all that infallible -- Michael Pachter, Jeetil Patel and Mitshuhiro Osawa all top out the scale at 60% correct. That might not sound so good, but it looks absolutely glowing when compared to Billy Pidgeon's dismal 10% truthfulness rating (with 20% still to be determined). Want to avoid his ignominious fate? Just be like Piers Harding-Rolls and predict stuff so far into the future no one can tell how accurate your calls are until you've moved on to used car sales or something.The full piece takes an interesting look behind the scenes in analyst-land and provides some good reasons to be cautious of all those stories with headlines that start with "Analyst predicts ..." Do yourself a favor, take a few minutes and give it a read.
Breakfast Topic: Things they'll never announce at BlizzCon
We've all heard rumors and speculations about what's due to be announced at BlizzCon today: a new expansion, a new class, and who knows what else. Now let's turn aside from what we think will happen and consider the infinite possibilities of what will not happen. Yes, indeedy -- today's the day we'll look into our crystal balls and talk to our Ouija boards, predicting with 100% accuracy all the things Blizzard will not announce today at BlizzCon! To get your inspiration going, I'll give you a few examples. New Expansion: The Search for Mankirk's Wife (in which the new area is just an expanded Barrens -- ten times larger than the original) New Class: Teletubby. Available races: Gnome, Blood Elf. Signature spell: "Infuriating Giggle" New Profession: Retail Salesmanship. Lets you set up shops and stand around all day waiting for customers. Raid bosses will drop epic brand name contracts.