 Your new post is loading...
Your new post is loading...
Viktor Mayer-Schonberger and Kenneth Cukier earlier this month published a book together called "Big Data: A Revolution That Will Transform How We Live, Work, and Think" that examines the drivers, trajectory and impact of big data analytics.
Mayer-Schonberger is a professor at the Oxford Internet Institute at Oxford University, and is on the advisory boards of Microsoft (NASDAQ: MSFT) and the World Economic Forum. Cukier is the data editor of The Economistand a prominent commentator on developments in big data. They shared an excerpt of their book this week in Quartz, about how the algorithms used in big data are creating artificial intelligences that no human can understand, making it quite the challenge to retrace the steps that led to a determination.
Software code is decipherable. Programmers can trace their way back step by step to find an error, and then fix it. And logs generated by network elements can help network operators determine why something may have routed incorrectly. But data scientists can easily paint themselves into a corner.
It's an incredible image that falls in the annals of both technology and aviation history as it's a 5MB hard drive being loaded onto a Pan Am plane in 1956. it weighs one ton. Although there's no details with the photo on how much this hulking monster of early computing costs or where it was heading, it's pretty safe to say it's a huge novelty for the crew, who look down in awe at the forklift and its load. Happily we have also advanced in our airline cargo technology.
How do you improve the mobile experience when all you have is a visibility over a carrier’s network? That is the question that started carrier IQ (CiQ) in 2005. They set out to open up a window on the devices. To show what was happening inside what was until then a closed ‘black box’ after it left the factory. Seven years later, the company now has a tremendous technology foundation, from a unique agent software which gathers system data on smartphones (180 million rolled out so far), to an incredible analytics big data platform that’s able to deliver real-time drill-down information on technical parameters of a phone as well as trends and statistics. Last ime I checked CiQ was processing over 7 terabytes of smartphone system data each month… They have a great business team, too, and managed to develop solid relationships with major mobile carriers in the US and abroad. And weathered an incredible storm in late 2011 — more on this at the bottom of this post.
Having more data does trump a better algorithm, but it's not that simple.
Researchers in Spain used a huge archive known as the Million Song Dataset, which breaks down audio and lyrical content into data that can be crunched, to study pop songs from 1955 to 2010. A team led by artificial intelligence specialist Joan Serra at the Spanish National Research Council ran music from the last 50 years through some complex algorithms and found that pop songs have become intrinsically louder and more bland in terms of the chords, melodies and types of sound used. .../... (Abstract from Reuters available at http://www.reuters.com/article/2012/07/26/us-science-music-idUSBRE86P0R820120726)
The Pathway explores how systems responsive to behaviour patterns and sensor data enable new experiences better suited to individual user needs. It aims to solve specific design challenges around the self learning process, balancing dynamic and consistent interface elements, approaches to privacy management and making it easy for users to correct contextual mistakes. • In which scenarios do real-time sensor data and behavioural patterns combine most effectively to reveal user context?
• What is the right balance between dynamic, contextually responsive elements and consistency in the user interface?
• How can contextually responsive experiences respect privacy yet learn user behaviour quickly enough to deliver benefit?
• How do graceful failure techniques allow users to correct inaccurate contextual guesses and reassure the system is learning from mistakes?
Search over 6.5 million UK mobile phone contracts instantly by network, brand, cashback, free gifts, minutes, texts, data and monthly cost.
Twitter data is potentially lucrative for the global communications business. It can inform when to serve up advertisements on the site and for what kinds of products. It also produces cool graphs and charts.
Face à une profusion d'informations de plus en plus hétérogènes et de moins en moins structurées, un des enjeux majeurs pour l'ensemble des acteurs économiques dans les dix ans à venir sera leur capacité à exploiter, analyser et transformer en valeur ces avalanches de données produites, et ce, si possible, en temps réel.
A video interview with internet guru Doc Searls, fellow of Harvard University, in which he describes the concept of ‘VRM’ (Vendor Relationship Management) and how it will revolutionise customer relationships.
The presidential campaigns have the technology to know more about voters than any other bids in history... "In recent primary states, Romney aired two very different ads on local news websites: one for supporters, and another for those who may not support him. This sort of thing could pop up on television as well very soon. Within five years, Will Feltus of the National Media predicts, advertisers and politicians will only target households where the messages can have maximum effect. And just like that, the days of shared commonality over a played-out commercial will be over. Women won't have to endure Cialis marketing, and men will be skipped when it comes time to air the Tampax ads."
|
import•io, the cloud-based Big Data platform, is announcing it has raised £600,000 ($900,000) from Wellington Partners, Alta Vista founder Louis Monier, and long-time French technology investor Emmanuel Javal to help the award-winning start-up expand and develop its platform. import•io turns the web into a database, releasing the vast potential of the data trapped within the web. It lets companies, developers and coders extract, connect and fuse disparate sources of data from the web to let them build and create powerful new data sets and applications. “We believe that the Web is full of useful data, but the reality stands: the Web was designed for documents, not data. At import•io, we want to allow anyone to harness this trapped data. We democratize access to it by making it simple to turn Web sites into data sources, and to fuse them to create incredibly rich new data sets in a single real-time, virtual database that can be easily and consistently accessed,” says David White, CEO of import•io.
"a tiny San Francisco startup, Adamant Technologies, is trying to give your iPhone the senses of smell and taste, too. The company has created a computer chip that works with a bunch of tiny sensors that "can take the sense of smell and taste and digitize them," explains Sam Khamis, Adamant's founder and CEO. This is not about turning your smartphone into some kind of scratch-and-sniff thing that emits scent. It's about letting your phone or computer or other medical devices smell for themselves. This was a pretty tricky problem to solve. A computer can easily identify a chemical in the air, but put a bunch of them together and it's stumped. For instance, humans can tell when there's pizza and chocolate chip cookies in the same room. Computers have a harder time with that."
Numenta is a startup with a cloud-based prediction engine for streaming data. The company launched in 2005 and went into beta earlier this year. It is ready enough now to start making a difference that the New York Times profiled its founder, Jeff Hawkins, this week and called his company a brainy big data company--a play on Hawkins' theories on neuroscience.
In fact, Numenta's core technology, the Grok prediction engine, was developed based on a theory of the neocortex. Hawkins' approach may upset the big data applecart, according to the Times, because it focuses on real-time streaming sensor data and goes against the grain of mass storage that is driving much of the development on databases such as Hadoop.
Hawkins told the Times that "much of this will be a relic within a few years." He said Hadoop won't go away, but it will manage a lot less stuff and that querying databases won't matter as much. "It only makes sense to look at old data if you think the world doesn't change," Hawkins said.
The history of Customer Relationship Management (CRM) systems holds valuable lessons. The first wave of CRM systems got a boost with success stories in an HBR Article ("Automation to Boost Sales and Marketing") in 1989 by Moriarty and Swartz. Billions of dollars and millions of hours were spent to build and operate these systems. But as early as 1990, the Conference Board cautioned that half the companies that had purchased such systems regretted going down the path they did. In the second wave, led by Siebel Systems, tens of billions of dollars were spent. Many surveys around the year 2000 reported that 50% to 70% of these implementations failed. Now in its third wave, recent CRM efforts are more successful, having matured with the lessons of the early failures and the emergence of industry-tailored solutions that are quicker to deploy and lighter to sustain. But it's been a long time coming.
So how then should you think about big data for your sales organization? Here are three recommendations:
Consumer adoption of cloud services will continue to increase in the coming years, with one-third of personal data being stored in the cloud by 2016, Gartner predicts. Today, only about 7% of personal data is stored in the cloud. But, Gartner researchers say the ability for consumers to capture data on their smartphones and tablets, using cameras and video recording devices, will drive the need to store data outside of local traditional options, such as on a personal computer hard drive or an external hard drive. Plus, the advent of technology allowing data to be automatically uploaded to the cloud, will mean that 36% of consumer data will be stored in the cloud within four years.
Que se passe-t-il sur Internet en une minute ? Une infographie compile toutes les données échangées et générées sur le Net en 60 secondes.
Via francoisloviton
Google just unveiled numbers showing an alarming jump in the number of government demands for private user data. We took the technology giant to task for its report's shortcomings. But at least Google is moving toward transparency.
Via Pierre Paperon
Two key members of the Facebook team that created the Hadoop query language Hive are launching their own big data startup called Qubole on Thursday. Co-founders Ashish Thusoo and Joydeep Sen Sarma are taking that experience to provide a managed version of Hive that’s hosted on the Amazon Web Services cloud computing infrastructure. The founders know from whence they speak when it comes to Hadoop and Hive, which is a framework and associated query language that sits above Hadoop and turns into something resembling a traditional SQL-based data warehouse. Both worked at Facebook from 2007 through 2011 and held senior positions on the data infrastructure team, where they played major roles in creating Hive and scaling Facebook’s Hadoop cluster to 25 petabytes of compressed data (it’s now up past 30 petabytes). Thusoo also spent a year as the project lead for Hive with the Apache Software Foundation.
La révolution numérique est derrière nous. Des milliards d’êtres humains sont aujourd’hui instruits et informés, équipés et connectés. Leur désir de créer, de communiquer et de partager n’a jamais rencontré autant de possibilités de passer à l’acte. Ces milliards d’individus composent une « multitude » puissante, mouvante et active, qui bouleverse l’ancien ordre économique et social et ouvre la voie au troisième âge du capitalisme. La multitude est désormais la clef de la création de valeur dans l’économie. Qui sait susciter, capter et redistribuer la créativité de la multitude peut devenir un géant de l’économie numérique. Qui accepte et nourrit la multitude peut gouverner avec une efficacité jamais atteinte. A l’inverse, qui ne voit pas que l’essentiel de l’intelligence et de la puissance est en dehors de son organisation risque de se faire balayer par les champions de l’économie numérique : ceux qui, par des innovations radicales, font alliance avec la multitude. La multitude est devenue la clef du succès des organisations. Elle est la richesse des nations d’après la révolution numérique.
The ability to understand user context is increasing continuously. New sensors and data points colour ever more detail into the picture of who a user is, what they’re doing and where they are interacting with digital services. The way in which design responds to context, however, lags this new availability of contextual data points. This is due partly to the relative complexity of creating contextually responsive digital experiences and partly to a misconception among designers that users are willing to change their context in order to gain access to particular products. In reality, such life altering products are few and far between. Most experiences, and particularly those distributed across multiple digital touchpoints, can be made better by embracing contextually responsive principles.
|
According to a new book, "data scientists can easily paint themselves into a corner" as no human can possibly understand the kinds of artificial intelligence created by such algorithms and answer the "how" or the "why" questions...