Who doesn't want an AI dino?
First it was Jeopardy!, then it was cancer, e-commerce and cooking. Now, IBM’s Watson artificial intelligence system is powering a line of…
Just the ones that need ATTN
A text-analysis startup called Parakweet (whose initial product focused on book recommendations) has launched a new application, called InboxVudu, that’s designed to…
Move over, Hello Kitty
IBM has struck a deal SoftBank Telecom Corporation to bring the IBM Watson artificial intelligence (or, as IBM calls it, cognitive computing)…
Hilary Mason breaks it down
When Hilary Mason talks about data, it’s a good idea to listen. She was chief data scientist at Bit.ly, data scientist in…
Happy ZIP codes, longer lives
University of Pennsylvania researchers have found that the words people use on Twitter can help predict the rate of heart disease deaths in…
Who needs chefs anyhow?
IBM has recruited a couple of new partners in its quest to mainstream its Watson cognitive computing system: financial investment specialist Vantage…
Smart devices, by Facebook
Facebook has acquired Wit.AI, a San Francisco-based startup building a speech-recognition platform for the internet of things. The company launched early in…
Coming soon to an app near you
There’s more to speech recognition apps than Siri, Cortana or Google voice search, and a San Francisco startup called Expect Labs aims…
Smart and budget-friendly, too
The Department of Veterans’ Affairs is working with IBM to analyze hundreds of thousands of VA hospital medical records using the Watson…
Meanwhile, under the surface
The artificial intelligence technique known as deep learning is white hot right now, as we have noted numerous times before. It’s powering…
A Palo Alto startup called MetaMind launched on Friday promising to help enterprises use deep learning to analyze their images, text and…
Nuance Communications is sponsoring a contest called the Winograd Schema challenge that aims to replace the usual conversation-based attempts to pass the Turing test. According to its backers, the Winograd challenge places more emphasis on intelligence and less on trickery.
Targeted machine learning applications continue to raise money and big-time users, possibly at the expense of general-purpose platforms targeting potentially much broader use cases.
Researchers at both Stanford University and Google have fused state-of-the-art neural network techniques for vision and language in order to create hybrid systems that can analyze images and produce natural-language explanations of what’s happening in them.
IBM’s Watson group has invested an undisclosed amount of money in Pathway Genomics to help the company deliver an app that gives…
Amazon just showed off a new connected device that combines the skills of Siri with a speaker. It’s a cool idea and looking at other products like it can tell us what Amazon needs to get right.
“Insteon, I’m home!” With those words, an Insteon home automation system and a Microsoft Windows Phone 8.1 handset, you could unlock your doors and turn on your lights.
A text-analysis startup called Aylien has released an add-on for Googles spreadsheet application, which is actually pretty handy. It’s far from perfect at sentiment analysis as most services are, but it’s easy to use and does a good job extracting the stuff that matters.
IBM is getting into the freemium space, targeting individual business users with a new data analysis service called Watson Analytics. It helps users analyze data using natural language queries, and could help IBM fend off the myriad products threatening its analytics business from the bottom up.
Expect Labs is expanding the scope of its MindMeld API with a new offering focused specifically on enabling voice-powered mobile recommendation apps…
A handful of new research projects from Google, IBM and the Allen Institute for AI highlight the ongoing quest to build computer systems capable of analyzing written language based on understanding concepts rather than just keywords.
Jeremy Howard, the former president and chief scientist of predictive-modeling platform Kaggle, is back with a new startup he thinks can revolutionize medical diagnostics using deep learning. There’s a lot of work to be done, but a lot of reason to be optimistic.
Yahoo has been rebuilding its research focus, and now a 250-person strong Yahoo Labs is trying to build the future of content. Full of experts in everything from personalization to deep learning, the group wants to make finding and consuming content as easy as possible.
The startup founded by Google and Yahoo engineers uses Apache Spark to power an easy to understand user interface that resembles Google Docs.
Netflix explained how it’s using data analysis to do more than recommend movies in a blog post this week. From optimizing bitrate to churning through user feedback, advanced algorithms are helping ensure that minimal issues affect the streaming experience.
Deep learning is all the rage among the tech scene right now, and that’s more a result of its utility than because it sounds cool. Some questioned the feasibility of the Secret Service’s requested “sarcasm detector,” but deep learning could help there, too.
Gigaom has written a lot about artificial intelligence over the years. Here are three timelines tracking the rise of deep learning and other learning systems, IBM Watson and AI discussions at Gigaom conferences.
Apixio has raised a $13.5 million series C round of venture capital from Bain Capital Ventures and several angel investors. The company’s technology extracts…
Researchers from Allen Institute for AI have built a computer system capable of teaching itself many facets of broad concepts by scouring and analyzing search engines using natural language processing and computer vision techniques.
Expect Labs has revamped its MindMeld artificial intelligence API to support new languages, more devices and better speech recognition. The company is trying to power a new generation of apps that build on what Siri started.
A Palo Alto startup called Wit.AI wants to enable a world of intelligent dialog with the devices around us, but conquering voice commands is the first step on its mission.
Facebook artificial intelligence director and NYU researcher Yann LeCun delved into the future of AI last week in an Ask Me Anything session on Reddit. Here are his thoughts on why Her is a long way off, but is on the right path.
Artificial intelligence techniques such as natural-language processing and computer vision will someday revolutionize our world. First, they’ll probably help retailers sell us more stuff.
The Defense Advanced Research Projects Agency, or DARPA, is building a set of technologies to help it better understand human language so it can analyze speech and text sources and alert analysts of potentially useful information.
IBM has made another investment out of the $100 million it has set aside to fund companies using the Watson cognitive computing…
A startup called Gridspace is trying to reinvent the meeting process by taking the note-taking out of it. Its Memo system combines hardware, software and machine learning to let a computer take notes while the people focus on ideas.
Microsoft crammed numerous announcements into a three hour keynote event on Wednesday at its Build event. Windows, Windows Phone and Xbox all get new features and some of them are shared, bringing more consistency to the Windows user experience.
AlchemyAPI has released a new deep-learning-based API it says can automatically categorize content into inventory suitable for targeted advertising. It’s among a handful of improvements to AlchemyAPI’s service and in the deep learning space, in general.
Anticipatory search is already here, but now apps are evolving into anticipatory analysis engines as well, using data from our actions to serve up information before we ask for it.
DataRPM is one of a handful of companies trying to move business intelligence into its next generation by incorporating natural language processing and a search-like experience. InterWest Partners led its series A round.
Though these technologies are complex, it’s crucial that we understand them and how they work since they are powering the next wave of applications, performing tasks so complicated that traditional programming techniques can’t solve them.
An Irish startup called Aylien is getting into the natural-language processing space with a set of APIs for text analysis. It’s not the first company to do this, but it might be the most unique.
IBM announced its Watson Mobile Developers Challenge on Wednesday. The company is pushing Watson as a cloud service hard because it knows it has its work cut out to win developers away from startups and large companies like Google also pushing AI via API.
Expect Labs has unveiled the MindMeld API, a set of artificial intelligence capabilities delivered as a service. Developers can use it to create smart applications that know what types of content and search results to recommend, and when.
Netflix is the latest company to acknowledge that it’s researching new approaches to artificial intelligence that could help improve its products. Although it hasn’t said where it might apply deep learning models, the company has plenty of image and text data to learn from.
Everyone’s more interested in artificial intelligence since news broke that Google acquired a secretive startup called DeepMind. The technology has big promise, but make no mistake: It’s not sentient yet, and Google is far from alone in its quest.
The operational cost of normalizing and mining human-generated data is significant and requires a sound strategic understanding of technologies and business goals.
The legal profession has undergone a lot of unpleasant changes since the Great Recession struck in 2008. New data-analysis technologies and a new approach to thinking about data could help firms operate leaner, meaner and better.
While much of the talk around data the past few years has been about how much we can store and what we can learn from it, the future — and this year’s Structure Data conference — is about what we can do with data.
A group of Stanford machine learning students has created a new service for analyzing and classifying passages of text. But the highlight is an easy-to-use feature for classifying whether tweets are positive, negative or neutral in tone.
Yahoo has acquired SkyPhrase and will incorporate the team into Yahoo Labs. SkyPhrase had built a natural-language processing platform that returned relevant statistics in response to search queries entered using everyday language.
A Japanese project aimed at creating a computer system smart enough to pass the University of Tokyo entrance exam scored above average on a recent test run of sample math questions, highlighting some its progress as well as some problems.
The relevancy-defining, edge-weighting algorithms of Google’s Knowledge Graph, Facebook’s Open Graph and Gravity’s Interest Ontology are closely guarded company secrets. Imagine if that data was available to everyone — it would be as disruptive as Amazon Web Services. The internet would be a better place.
You didn’t think all the research Microsoft has done around deep learning was just for show, did you? The company’s deep learning models are now powering voice commands on the Xbox One platform, thanks to a direct connection to Bing.
IBM has upped the ante in the API game by making its Watson question-answering system available as a service. That’s right, Watson could soon power your smartphone app.
Deep learning is one of the hottest trends in big data right now and is currently underpinning the cutting edge in areas such as natural language processing and image recognition. Here’s a brief guide about what it is about who’s doing it.
One way to get more things done is to combine tasks with events in a single place. That’s exactly what Fantastical 2 for iOS does and it makes it easy to do so thanks to improvements in its natural language processing.
Online health community HealthTap has launched a new paid mobile app, called TalktoDocs, that gives patients a voice-controlled way to get answers to their medical questions.
IBM has shared some details about a new project called WatsonPaths that lets doctors actually interact with the system to understand how…
A group of researchers from Stanford has been working on deep learning models that can make sense of whole sentences at a time, and has recently trained its models on a large collection of online movie reviews.
Yelp has announced the winners of its inaugural Yelp Dataset Challenge, and the four entries it chose actually seem pretty useful. They…
Fantasy football is a big business that thrives on data, making it a great way to prove out a new technology and possibly earn a few bucks. A startup called SkyPhrase, for example, is putting its natural-language processing technology to use on NFL statistics.
SwiftKey, a London-based startup that sells a popular “smart” keyboard for Android devices, has closed a $17.5 million series B led by…
A group of Spanish researchers have developed a chatbot that poses as a teenager in order to catch pedophiles online. It’s an interesting idea and mix of technologies, even if it’s not fully baked.
Ginger Software, a Tel Aviv-based startup, is using algorithms to help non-native English speakers improve their writing.
While visualizations have gotten plenty of attention as options for getting good stuff out of data, In-Q-Tel’s investment in Narrative Science suggests information in paragraphs could work too.
Ancestry.com wants to help users of the family-history site share their discoveries. Now it’s devised methods for turning isolated facts into full-on stories.
Looking for a book suggestion? Culling information from your Twitter feed and turning that into accurate recommendations is harder than it looks, but Parakweet is looking to use natural language procesing to do just that.
When it comes to using big data technology effectively, there’s a lot to like about SaaS. When companies like BloomReach create and analyze massive web-wide data sets, they automate insights that almost no individual company could discover on its own.
With its acquisition of Lucky Sort, Twitter seems to be acknowledging that it’s a data company after all. The plan appears to be building a services that would do for Twitter equivalent to services such as Google Trends and Google Analytics.
MailChimp wasn’t always a big data company, but 12 years into its existence the company is using its mountains of email data to do everything from modeling spam to connecting subscribers.
In a post on Facebook’s engineering blog, engineers discuss the ways in which natural-language processing helps interpret what users plug in to suggest the best possible queries.
Thiel Foundation subsidiary Breakout Labs has funded two new startups called SkyPhrase and Stealth Biosciences that, respectively, are trying to reinvent natural language processing and improve our ability to interact with individual cells.
A San Francisco company has raised $1.4 million in seed funding to bring to market a tool for processing text in any language in the world.
A team of Stanford researchers has developed a method for mining the text of doctors’ notes to identify adverse reactions from prescription drugs. The technique could spot problems years before the current FDA-reporting process can.
Natural-language processing powers the Instant Answers feature from business intelligence startup DataRPM, which could help more people easily get insights from their big sets of data.
Natural language user interfaces show great promise for a variety of situations, but Nuance Communications CTO Vlad Sejnoha sees a need for standard formats if new entrants aren’t to be blocked as data sources.
Just when you thought spam was under control, a new breed of spammers is taking up new methods to infiltrate our inboxes, search results and social media feeds. Data science could make them very effective.
IBM is turning Watson loose on lung cancer, offering up a cloud-based service designed to let doctors from around the country find the best-possible treatments for their patients.
First, it was semantic search and knowledge graphs surfacing information related to our keyword searches. But there’s a handful of companies working to make relevant content come to us, whatever we’re doing.
IBM is giving Rensselaer Polytechnic Institute in New York its own Watson system similar to the one that crushed its human competitors on Jeopardy!. The goal is to give Watson new skills and push it into new industries.
A group of British researchers recently analyzed 2.5 million newspaper articles in order to prove that new data analysis techniques, such as machine learning and natural-language processing, can accurately classify media content. They hope their approach can save academicians untold hours of manual labor.
A new research paper from Google highlights the importance of big data in creating consumer-friendly services such as voice search on smartphones. More data helps train smarter models, which can then better predict what someone say next — letting you keep your eyes on the road.
Social data pulled from online health forums and the comments section of blogs is helping patients learn about side effects to various drugs and could ultimately help them figure out the medications that suit them best.
The Todai Robot project has received attention because of its goal to build a system capable of passing a college entrance exam — a tough test for a supercomputer that’s all the more difficult when confined to a single laptop computer.
Now that touch input has revolutionized the mobile device market, what’s next? Voice interaction and natural language processing (NLP) surely have to be high on the list. But it’s not always easy for app developers to enable voice features. Enter OneTok, a cloud-based NLP solution.
In yet another move to simplify the process of patent search, Google is now tackling the particularly tough problem of prior with a new feature. But all this innovation begs the question of whether Google, not the USPTO, should become the de facto patent search engine.
Solariat Founder and CEO Jeffrey Davitz has a message for anyone trying to leverage social network data to make money: “The fundamental problem with social is yes, it’s big data, but it’s mostly big, sucky data.” Targeting users means deciphering what they really want.
Even if you haven’t heard of LivePerson, chances are you’ve encountered one of its products while browsing online. It’s the company behind those pop-up windows offering real-time chat with a representative, and it uses big data to decide which visitors are worth what type of attention.
The moods of men can be captured by a web app, which can then recommend the appropriate spiciness of chicken wings they should eat, and even suggest appropriate mood music thanks to Spotify. Welcome to the data-driven future. Adjust your personal privacy setting accordingly.
DataPop, a startup using big data to deliver custom online ads, has raised a $7 million Series B round. The company’s technology uses big data techniques such as natural-language processing and semantic association to automatically generate online ads based on what a web user has searched for.
Companies are hot on social media for a number of reasons, but perhaps chief among them should be that social platforms can create focus groups at a scale never before possible. Given the right big data tools and techniques, the insights can be fantastic.
We’re entering the age of the smart personal assistant, as computers increasingly listen and understand what we’re saying and fulfill our requests in real time. Siri and Watson have gotten a lot of attention but we’ll see this type of service in a variety of areas.
Beepl opened its social Q&A site to the public, with the hopes of challenging Quora and others by finding specialists to answer user questions. It does that with technology that matches users with the queries that they are most interested in and most qualified to answer.
Speech technology is poised to be a game-changer for smartphones, especially as they get embedded into operating systems and hardware. Nuance CTO Vlad Sejnoha said speech is transforming from an alternative to text input into a powerful tool that can connect users more quickly to information.
Thanks to a slew of recent technology advances, the smartphone is turning into a truly personalized information assistant that can handle everything from your appointments to traffic reports. A new Android app called reQall Rover wants to turn your smartphone into an intelligent information assistant.