To install click the Add extension button. That's it.

The source code for the WIKI 2 extension is being checked by specialists of the Mozilla Foundation, Google, and Apple. You could also do it yourself at any point in time.

4,5
Kelly Slayton
Congratulations on this excellent venture… what a great idea!
Alexander Grigorievskiy
I use WIKI 2 every day and almost forgot how the original Wikipedia looks like.
Live Statistics
English Articles
Improved in 24 Hours
Added in 24 Hours
What we do. Every page goes through several hundred of perfecting techniques; in live mode. Quite the same Wikipedia. Just better.
.
Leo
Newton
Brights
Milds

Media intelligence

From Wikipedia, the free encyclopedia

Media Intelligence uses data mining and data science to analyze public social and editorial media content. It refers to marketing systems that synthesize billions of online conversations into relevant information that allow organizations to measure and manage content performance, understand trends, and drive communications and business strategy.

Media intelligence can include software as a service using big data terminology.[1] This includes questions about messaging efficiency, share of voice, audience geographical distribution, message amplification, influencer strategy, journalist outreach, creative resonance, and competitor performance in all these areas.

Media intelligence differs from business intelligence in that it uses and analyzes data outside company firewalls. Examples of that data are user-generated content on social media sites, blogs, comment fields, and wikis etc. It may also include other public data sources like press releases, news, blogs, legal filings, reviews and job postings.

Media Intelligence may also include competitive intelligence, wherein information that is gathered from publicly available sources such as social media, press releases, and news announcements are used to better understand the strategies and tactics being deployed by competing businesses.

Media Intelligence is enhanced by means of emerging technologies like semantic tagging, Natural Language Processing, sentiment analysis and machine translation.

YouTube Encyclopedic

  • 1/3
    Views:
    6 458
    49 047
    23 279
  • MIT Intelligence Quest Launch: The Future of Intelligence Science
  • Astronauts High on Science and Low on Intelligence!
  • The Science of Disinformation: Counter Intelligence Programs, Psy-Ops, Astroturfing

Transcription

Good morning. I'm Jim DiCaralo. I'm privileged to lead a science community here at MIT, the brain and cognitive sciences community whose mission is to understand the brain and how that gives rise to the mind. Now, the science quest to understand human intelligence is one of the most exciting frontiers of our field. It's really a quest to understand ourselves. And it's aligned with a technology quest of developing intelligent systems. This is a unique time in human history. It's a unique chapter in MIT'S legacy as we see tremendous new opportunity for synergy. And today, some of my colleagues and I would like to tell you about that synergy in our vision of the future. Now, all the fields of artificial intelligence and machine learning have led to many technological advances that continue to transform and improve our lives. None of these systems are truly intelligent, as you'll hear about in a moment. We're still very far from real AI. However, as brain and cognitive scientists, we know that real AI is possible because we study a real intelligence machine every day. The machine behind your eyes can navigate new situations, infer what others believe, use language to communicate, write poetry and music to express how it feels, create math to build bridges, devices, and life saving medicines. In short, it's built civilization from scratch. The human brain can do what AI systems cannot yet do. If harnessed, the science of human intelligence provides an incredible opportunity to replenish the well of AI algorithms. Now, the human brain, shown here, is a 20 watt machine. To some, the idea that the brain is a machine is itself an astonishing hypothesis as illustrated by this quote from the late Francis Crick, you, your joys, your sorrows, your memories, your ambitions, your sense of personal identity and free will, are in fact, no more than the behavior of a vast assembly of nerve cells and their associated molecules. As scientists, we have the opportunity and, indeed, the obligation to reverse engineer this brain machine. But this is not a simple task, as can be illustrated by thinking about this far less advanced, less intelligent machine but that one that we know more about. In a similar analogous quote, your remarkable favorite app, all its performance an amazing user interface is in fact, no more than the behavior of a vast assembly of transistors. So many of you in the audience know that this is true, that that device and its amazing interface is due to those transistors, it's building blocks shown on the left. But you can't understand how that amazing performance on that app comes from without understand the complexity that lives between. Similarly, as brain cognitive scientists, we study the amazing intelligence and cognition of humans. And we know the building blocks of that system. They're called neurons. And there are connections called synapses. You have billions of neurons in each of your brain. We know something about what intervenes between those neurons as they're put together into circuits that give rise to algorithms that are underlying our cognition. But the challenge of our science is to discover which arrangements of neural building blocks can explain and underlie our intelligence behavior. So how do we propose to do this? Well, as we make measurements and discoveries as scientists, shown here on the left, we must, and we are building algorithms that offer the potential of showing how those neurons are connected to give rise to the intelligent behavior, shown here in the center. You could refer to those as models, or our intelligent algorithms, are hypotheses about brain function. And here's where we see great synergy with similar efforts from our machine colleagues engineers that work in software hardware robotics, as they are great at synthesizing, and creating, and are also interested in building things that can explain intelligence. So we're united around the mission of intelligence algorithms. And we aim to leverage, in the core, the complementary strengths of both sides of this equation and to build algorithms at human relevant scales. These algorithms will, in turn, inform our science through these reverse green arrows that you see here, about hypotheses of a brain function, where intelligence comes from, and inform our engineering as possible new technologies, shown by the blue arrows. We believe this approach, over time, will again, replenish the will of AI algorithms that can impact the greater MIT and, indeed, the greater world, through the work of the MIT IQ bridge that you'll hear later today. Why do we think this is a good bet? We think this is a good bet based on past success. Let me give you just one recent breakthrough example. You probably all recognize this person. He's our leader, Rafael Reif. But how did you do that so quickly and effortlessly? This object recognition problem was a longstanding problem in computer science and brain and cognitive sciences. For decades, neuroscientists like myself and others accumulated knowledge about how the brain solves this problem. And that knowledge, this schematized at the top of this slide, in its deep neural network that was discovered within the brain and the arrangement of neurons and processing the image locally. And we learned a lot about the responses of neurons along that system as well as the behavior of that system. Yet, all of that work, decades of work, was not enough to tell us the visual intelligence algorithms working in the brain. Similarly, engineers working in a separate silo were trying to build computer vision algorithms that could do vision cast like this one. And all of those algorithms fell far short of human capabilities. Now, true progress came when researchers used a combination of science and engineering, specifically some researchers began to build networks out of brain like components, so-called deep neural networks with brain like neural elements, like this example schematized at the bottom here. They also use mathematical models of how to teach those networks proposed by cognitive scientists. My lab and others showed there was a remarkable similarity from the artificial neurons at the bottom network to the actual neurons in the brain shown at the top. It was a very tight match. As well as a very good similarity between the behavior, the visual behavior of the human visual system, and the behavior of these artificial systems. And in 2012, these brain like deep neural networks took over computer vision, and the deep learning revolution and that transformative AI that resulted was launched. We're not finished even with this vision problem. There are feedback loops you can see on the slide at the top that aren't even present in those feedforward networks. We're working on those now, as we think they're important to scene understanding. But the big lesson I want you to understand is that when engineers guide their algorithm building efforts with discoveries and measurements from science, then you get transformative progress. This is why we believe in, this is why we're confident in, this is, indeed, why we need the MIT corps. But here, I should also remind you, we're just scratching the surface. I'm talking about core visual object recognition shown at the top. That's just one aspect of visual intelligence. Imagine what we can do if we can understand all the other aspects of human visual intelligence. Things like intuitive physics, intuitive psychology, language acquisition, and many others that you'll hear some later today. Now, this slide shows moonshot research goals that our faculty have brainstormed thus far. Bottom line, we have very exciting and ambitious goals, and we think with a combination of science and engineering that they are doable. Why now? Behavioral measurements are rapidly increasing in our ability to obtain them both in adults, but also in developing and learning children, as you hear about from my colleague, Laura Schultz, in a moment. Neural measurements have been preceding a pace in animal models for the last decade. Our ability to measure more and more neurons simultaneously, learn more and more about the brain, as well as make new measurements from human brains that you'll hear about from my colleague Rebecca Saks in a moment. New algorithms are being hypothesized based on both theoretical considerations and observations of human behavior from colleagues all across MIT, not just in brain cognitive sciences, but places like robotics, and CSAIL, and others. You'll hear about a couple of examples of this from my colleagues Tommy Poggio and Josh Tenenbaum, who will speak in a minute. And rapidly advancing computational power is allowing us to build these algorithms at relevant scales along with new ways to implement that hardware in more efficient ways. And you'll hear a bit about that in session two. But I want to stress that because of that kind of recent success with deep networks, showing that science and engineering works together, many engineers are now motivated to build algorithms using brain like components, artificial neural networks, and to interface those networks those rooms with the brain. And you'll hear about some of those ideas from Antonio Torralba and Daniela Rus in a moment. In short, we're poised for an incredible decade of progress in intelligence research. Today, you'll hear a lot about AI. And I hope I've convinced you that work in brain and cognitive sciences around intelligence is a guiding path towards real AI. But this quest is likely to lead to many other world changing impacts. For example, a deep understanding of human learning in both its strengths and its weaknesses could have transformative impact on education. An engineering description of the brain will help us pinpoint when, where, and how something can go wrong. And armed with this, we'll likely be able to see new ways to ameliorate things like Alzheimer's, autism, hearing and vision loss, and many other conditions. Let me end by reminding you that over time, the MIT IQ corps could answer one of the biggest basic science questions of all time. How does the arrangement of billions of neurons make you, you? Thank you.

Technologies used

Different media intelligence platforms use different technologies for monitoring, curating content, engaging with content, data analysis and measurement of communications and marketing campaign success. These technology providers, such as BuzzCovery Meltwater, Synoptos, Radian 6, or Sysomos may obtain content by scraping content directly from websites or by connecting to the API provided by social media or other content platforms that are created for 3rd party developers to develop their own applications and services that access data. Facebook's Graph API is one such API that social media monitoring solution products would connect to pull data from.[2] Technology companies may also get data from a data reseller, such as DataSift (acquired by Meltwater), Gnip (acquired by Twitter), LexisNexis, or Dow Jones/Factiva.

Some social media monitoring and analytics companies use calls to data providers each time an end-user develops a query. Others archive and index social media posts to provide end users with on-demand access to historical data and enable methodologies and technologies leveraging network and relational data. Additional monitoring companies use crawlers and spidering technology to find keyword references, known as semantic analysis or natural language processing. Basic implementation involves curating data from social media on a large scale and analyzing the results to make sense out of it.[3]

See also

References

  1. ^ Leslie Nuccio (January 19, 2015). "Digital Breadcrumbs and the New Media Intelligence". Social Media Today. Retrieved March 23, 2017.
  2. ^ "Graph API". Retrieved 2015-05-14.
  3. ^ De, Shaunak; Maity, Abhishek; Goel, Vritti; Shitole, Sanjay; Bhattacharya, Avik (2017). "Predicting the popularity of instagram posts for a lifestyle magazine using deep learning". 2nd IEEE International Conference on Communication Systems, Computing and IT Applications (CSCITA): 174-177. doi:10.1109/CSCITA.2017.8066548.
This page was last edited on 1 November 2018, at 14:30
Basis of this page is in Wikipedia. Text is available under the CC BY-SA 3.0 Unported License. Non-text media are available under their specified licenses. Wikipedia® is a registered trademark of the Wikimedia Foundation, Inc. WIKI 2 is an independent company and has no affiliation with Wikimedia Foundation.