/* ---- Google Analytics Code Below */

Wednesday, November 30, 2016

Amazon Delivers Serious AI Services

   More tools being added to the box.  Previously some from IBM and Google.  Can remarkable AI be far behind?  In Computerworld.

AWS comes out swinging with A.I. services
Trying to entice enterprises and catch up with cloud rivals, AWS releases three artificial intelligence services   By Sharon Gaudin  

 .... At its re:Invent conference today, Amazon Web Services CEO Andy Jassy announced three artificial intelligence (A.I.) services that will be available to enterprise users this year, with more expected in 2017.

The three services being rolled out are Amazon Rekognition for image recognition; Amazon Polly for text-to-speech services; and Amazon Lex, the technology inside its smart device Alexa, offering speech recognition services.

Now enterprise AWS customers can use A.I. to, for example, search for images that show a mountain next to a lake, or a city with specific architecture. ... " 

via Walter Riker. 

Deep Learning and Optimization

Technical point about deep learning, diagrams at the link:

The hard thing about deep learning
Deeper neural nets often yield harder optimization problems.  By Reza Zadeh

At the heart of deep learning lies a hard optimization problem. So hard that for several decades after the introduction of neural networks, the difficulty of optimization on deep neural networks was a barrier to their mainstream usage and contributed to their decline in the 1990s and 2000s. Since then, we have overcome this issue. In this post, I explore the “hardness” in optimizing neural networks and see what the theory has to say. In a nutshell: the deeper the network becomes, the harder the optimization problem becomes. ... " 

Key New AI Deals

In Inc:  Some key acquisitions in AI that are worth examining:

Artificial Intelligence Deals Top $1 Billion
Apple, Google, Intel, Salesforce and Samsung are among the most aggressive acquirers   By Lisa Calhoun  .... " 

Service Innovation Talk: Emotional Intelligence

Intriguing ...

" ... Eric J. McNulty’s presentation for the ISSIP Service Innovation Weekly Speaker Series.
Title of his presentation is “Integrating Human Factors into Incident Command: Emotional Intelligence and Beyond”.
http://www.ericmcnulty.com
https://www.linkedin.com/in/emcnulty  
http://www.strategy-business.com/author?author=Eric%20J.%20McNulty

Eric is a Harvard-affiliated writer, speaker and educator with a passion for purpose-driven leadership. He uses his deep experience and expertise to help build adaptive capacity, trust-based cohesion, and resilience in leaders and their organizations. Eric works to increase the leadership capacity and capability in individuals, organizations, and communities to advance positive change. He is an author, speaker, and educator drawing inspiration from nature, the complex environments of cities, and the seemingly endless creativity of his fellow humans. His goal — and satisfaction — is tangible impact. http://ericmcnulty.com/about/who-is-eric/

ISSIP Service Innovation Weekly Speaker Series
Wed 7:30 am pacific time (San Francisco time zone)/ 10:30am EST
Co-hosts: Haluk Demirkan, Jim Spohrer, Yassi Moghaddam, DJ Christman, Heather Yurko

Join WebEx meeting
https://issip.webex.com/issip/j.php?MTID=m980f2a9da470d264408be87884cdcb2a
Meeting number: 927 413 167
Meeting password: innovation

Join by phone +1-415-655-0002 US Toll
Access code: 927 413 167   ...  "

Tuesday, November 29, 2016

Genes, Games, Randomness and the Power of Algorithms

In the CACM:  Sex as an algorithm.   Fascinating piece.  Not really technical, but a conceptually dense description.  Reminds me of some of our own experiments with genetic algorithms to solve complex corporate problems.  More in the tag below.  These were not really successful,  but why?

Probably because we were not successful in creating the right representation of the problem to work with.  Never obvious.    Even less so for this kind of solution method.   Also such genetically found algorithms are by their nature 'black boxes' whose operation cannot be directly explained.    Worth the scan or read for broad insight into the idea of genetically finding useful solutions.

IOT and Marketing

A fairly broad look at what IOT is, but interesting statistics on adoption.

Among Marketers, IoT Is Seen as Critical Emerging Technology
More businesses are in mature deployment stages this year  ...    "

Pitfalls in the Internet of Things

Video and Transcipt from Bain.

Ann Bosche: Avoiding Common Pitfalls in the Internet of Things

Enthusiasm for the Internet of Things continues to grow—already, leading vendors have made $75 billion in M&A investments. While many customers are excited by the opportunities the IoT presents, they still struggle with early solutions from vendors. Ann Bosche, a partner with Bain's Technology practice, discusses the most common mistakes made by vendors and the critical opportunities that will position them to win.   ... " 

Rain and Echo for Voice Driven Marketing

Interesting view of how skills are being built for voice channels.  I note that it repositioned some of the expertise for Tide stain detective, which is now available as an Echo Skill

How This Agency Built Its Own App Using Amazon Echo Voice skills have become big business for Rain By Lauren Johnson

Rain has picked up a few new clients with its voice expertise. Amazon
Last October, New York-based digital agency Rain worked with Campbell's to find a new way to push out the CPG giant's library of recipes, creating one of the first branded "skills" for Amazon's voice-controlled Echo device, which helped consumers find dinner recipes on demand.

While the agency expected a bit of a PR boost from the Campbell's work, voice and the so-called Internet of Things has actually become a big growth area for Rain over the past year, as more agencies look for ways to tap into emerging devices for brands. Shortly after creating the Campbell's skill, Rain became one of a handful of Amazon's go-to developers tasked with building "skills"—voice apps that let consumers ask questions to Amazon's AI platform named Alexa.   .... 

In the past year, Rain has signed 1-800-Flowers, P&G's Tide, Yahoo Fantasy Football and Liberty Mutual-owned Safeco as clients, helping to build branded skills that dole out information and entertainment to consumers. For Safeco, the agency built a glossary of insurance terms so that consumers can ask Alexa questions like, "What is umbrella insurance?" or "What does liability mean?" For 1-800-Flowers, users can automatically place flower orders to their friends or family members.  .. " 

Getting More Insights from Data

Good piece from Tableau:  Find hidden insights in your data: Ask why and why again by Andy Cotgreave.     I will add, always check with your business process context.  Some why's should always come from that context.

Monday, November 28, 2016

Bots as Filters and Beyond

Been examining the current uses and extent of bots,  Since our own experiments with bots we always looked at bots as a means to augment.

This Fortune article reminded me that they can also filter under dynamic contexts as well,  and driven by pattern recognition.   And the results could be much more subtle, and less noticeable than just blocking.   Good to remember.

Microsoft Confirms Its Chinese-Language Chatbot Filters Certain Topics   by Kevin Lui   

Xiaoice appeared to evade users’ questions deemed sensitive by China.
Microsoft’s  MSFT 0.13%  Chinese-language AI chat bot filters certain topics, the company confirmed Monday, although it did not clarify whether that included interactions deemed politically sensitive.

Last week, CNNMoney and China Digital Times reported that Xiaoice would not directly respond to questions surrounding topics deemed sensitive by the Chinese state. References to the Tiananmen Square massacre of 1989 or “Steamed Bun Xi,” a nickname of Chinese President Xi Jinping, would draw evasive answers or non sequiturs from the chat bot, according to the report. ... "  

Big Data and Decision Making

Always has been my point to directly link data analytics and business decisions.  This post suggests some links and tools ...

Big Data Make Better Business Decisions
Big data refers to a data set that can not be crawled, managed, and processed, parsed within a certain period of time by conventional software tools. Big data technology is used to deal with big data - crawl, manage, process and parse huge volumes of raw data and quickly access valuable information from a wide variety of data types. The technologies for big data include massively parallel processing (MPP), data mining for power grids, distributed file systems, distributed databases, cloud computing platforms, the Internet, and scalable storage systems. ... " 

Inducing Feelings about Faces

At Brown University. Interesting, borderline scary results.  We continue to learn more about brain function, but there is still much more to know.  'Reading' the brain still problematic. Writing into it is more so.   Still follow neuromarketing capabilities.

With MRI technique, brain scientists induce feelings about faces
In a new study, researchers report they were able to train unknowing volunteers to develop a mild but significant preference or dislike for faces that they had previously regarded neutrally. ... " 

Economics of Machine Learning

Thoughtful piece.   But as in all economic models,  If you accept the economics without the un intended and unknown consequences.

The Simple Economics of Machine Intelligence
by Ajay Agrawal, Joshua Gans, Avi Goldfarb

The year 1995 was heralded as the beginning of the “New Economy.” Digital communication was set to upend markets and change everything. But economists by and large didn’t buy into the hype. It wasn’t that we didn’t recognize that something changed. It was that we recognized that the old economics lens remained useful for looking at the changes taking place. The economics of the “New Economy” could be described at a high level: Digital technology would cause a reduction in the cost of search and communication. This would lead to more search, more communication, and more activities that go together with search and communication. That’s essentially what happened.

Today we are seeing similar hype about machine intelligence. But once again, as economists, we believe some simple rules apply. Technological revolutions tend to involve some important activity becoming cheap, like the cost of communication or finding information. Machine intelligence is, in its essence, a prediction technology, so the economic shift will center around a drop in the cost of prediction.

The first effect of machine intelligence will be to lower the cost of goods and services that rely on prediction. This matters because prediction is an input to a host of activities including transportation, agriculture, healthcare, energy manufacturing, and retail.  .... " 

Sunday, November 27, 2016

Smart Skin Mirror

Links to some of our smart mirror and virtual beauty innovation efforts, but emphasizes skin.  Useful calibration?   In Engadget: 

The first smart mirror you can actually buy focuses on your flaws   
By Cherlynn Low  @cherlynnlow 

Daily photos of your face are meant to flag problems with your skin.

Staring lovingly at your reflection in the mirror gets old, especially if it's not a magical one that tells you you're the fairest of them all. But what if all that time you already spend in front of a mirror could actually make you the most beautiful person in the land? Many smart mirrors already make this promise, but HiMirror is the first one you can actually buy, and it costs $189. The secret sauce comes in the form of a camera that captures your au naturel visage to analyze your complexion, before offering you tips on how to get better skin. I tried it out recently and was impressed by how many things it could do.

The 5.5-pound mirror is more cutesy than the futuristic thanks to its pink accents. It houses a 14-inch LCD screen and has a camera sitting in the middle of a ring flash, which helps light your face for better images. HiMirror snaps a picture of your face every day to analyze your wrinkles, dark spots, dark circles, red spots and pores. The idea is to track how your skin changes over time, and flag your so-called problems so you can take "early measures to improve upon your skin and achieve your beauty goals."     ... ' 

Stick on Microscope Lenses

Kind of thing that I could have used in a previous role.   I do like the extension of smartphones with new kinds of sensors.   And the simplicity of this.    In the Verge:  This is the best way to put a microscope in your pocket ... Blips’ stick-on lenses super-charge your phone’s camera, but good results take effort   by James Vincent   

New Kinds of Translation with Google

This is technical, but is now production on all Google translations services, even in Google Home, where I have been casually testing it.   Phrases and words.  Getting closer to a Babel fish?   Quite a peek into something deep.

Zero-Shot Translation with Google’s Multilingual Neural Machine Translation System  Posted by Mike Schuster (Google Brain Team), Melvin Johnson (Google Translate) and Nikhil Thorat (Google Brain Team)  Google Research Blog.  

In the last 10 years, Google Translate has grown from supporting just a few languages to 103, translating over 140 billion words every day. To make this possible, we needed to build and maintain many different systems in order to translate between any two languages, incurring significant computational cost. With neural networks reforming many fields, we were convinced we could raise the translation quality further, but doing so would mean rethinking the technology behind Google Translate.

In September, we announced that Google Translate is switching to a new system called Google Neural Machine Translation (GNMT), an end-to-end learning framework that learns from millions of examples, and provided significant improvements in translation quality. However, while switching to GNMT improved the quality for the languages we tested it on, scaling up to all the 103 supported languages presented a significant challenge ........ 

Finally, the described Multilingual Google Neural Machine Translation system is running in production today for all Google Translate users. Multilingual systems are currently used to serve 10 of the recently launched 16 language pairs, resulting in improved quality and a simplified production architecture.   ... 

Arup and 3D Printing

We did some work with the Engineering and  design firm Arup in the innovation space.   Was interesting to see they are doing work with 3D printing.

This Intelligent 3D Printer Is Building Big, Beautiful Structures  by Alison E. Berman

Ai Build, a London-based startup, aims to pave the way to 3D printing on large scales.

The company is equipping industrial-grade Kuka robotic arms with artificial intelligence and "3D printing guns" to 3D print large structures that focus on maximizing efficiency with labor and materials.

Founder and CEO Daghan Cam dreamed up the technology while considering traditional commercial construction and wondering what a more efficient and automated process might look like.

In October, the company partnered with engineering consulting firm Arup Engineers to debut the 3D printed “Daedalus Pavilion” at the GPU Technology Conference in Amsterdam. The structure is roughly 16 feet wide and 14 feet tall. Its 48 parts were printed in 15 days and assembled in less than one.

Ai Build's system uses video cameras outfitted with machine learning algorithms to allow robots to learn from their mistakes—meaning they can operate more quickly, correcting for errors on the fly instead of moving slowly to prevent them. According to Cam, Ai Build's arms can print in half the time it would take using standard techniques.   .... " 

Saturday, November 26, 2016

Thinking about Business Process as Recipes

Finally got a chance to look in detail at the Allrecipe skill in Amazon's Echo Alexa.  This lets you voice search 60K+ recipes and then have them recited step by step or sent to your device. Works well, but has limitations in ease of use, depending on your context.    We used this-rough cut model of process to frame a method:  What do you need?, what do you do?   A recipe of resources and actions. Useful beyond a to-do list.

Such a method could also be used to store and access steps in a business processes.   Perhaps leading to better standardization of process.  This probably works best in hands-free situations.   Useful for a salesperson in a car, or a repair person stepping through a checklist.   The search for the process step description could add other context sensitive and bot-like information to the interaction,  based on supporting data.

Today we expect to have ready access to online text resources on lap tops or smartphones,  wherever we are.    So the method of using voice interaction has its own limitations.  But can also be seen as another supportive channel to provide knowledge about process.  Leading to better results.

Check out Allrecipes.   How would you modify it to do process?  Someone working on this?  Also on other platforms like phones and Google Home too?   Be glad to talk, contribute.   - Franz Dill

Technological Change: Harvey Nash Tech Report

To register for the entire report.    Fascinating piece by a participant.  Links to my recent Columbia course on managing change, this is part of that change.

Almost half of tech professionals expect their job to be automated within ten years – Harvey Nash

Technology Survey 2017
Technology is 'eating itself' / Continual skills development key to career success
Forty five per cent of technology professionals believe a significant part of their job will be automated within ten years, rendering their current skills redundant.

The change in technology is so rapid that 94% believe their career would be severely limited if they didn’t teach themselves new technical skills.

This is according to the Harvey Nash Technology Survey 2017, representing the views of more than 3,200 technology professionals from 84 countries.

The chance of automation varies greatly with job role, with Testers and IT Operations professionals most likely to expect their job role to be significantly affected in the next decade (67 per cent and 63 per cent respectively), and CIO/VP IT and Programme Management least affected (31 per cent and 30 per cent respectively).


Friday, November 25, 2016

Thoughts on Simplifying Innovation

In Innovation Excellence, some interesting and simple thoughts about getting to real innovation.   The Tesla of Bikes Got a Brilliant Idea   by Gijs van Wulfen    

Questions to Ask Your Data Scientists

(And questions they should ask you)  Crucially this is not about recruiting, but to make sure you are managing them to get value from proposed solutions.   Management and data scientists should know the process, data, and value expectations.   Too often they don't

Better Questions to Ask Your Data Scientists
by Michael Li, Madina Kassengaliyeva and Raymond Perkins

The intersection of big data and business is growing daily. Although enterprises have been studying analytics for decades, data science is a relatively new capability. And interacting in a new data-driven culture can be difficult, particularly for those who aren’t data experts.

One particular challenge that many of these individuals face is how to request new data or analytics from data scientists. They don’t know the right questions to ask, the correct terms to use, or the range of factors to consider to get the information they need. In the end, analysts are left uncertain about how to proceed, and managers are frustrated when the information they get isn’t what they intended.

At The Data Incubator, we work with hundreds of companies looking to hire data scientists and data engineers or enroll their employees in our corporate training programs. We often field questions from our hiring and training clients about how to interact with their data experts. While it’s impossible to give an exhaustive account, here are some important factors to think about when communicating with data scientists, particularly as you begin a data search. .... " 

Red 212 Content Strategy Agency

Brought to my attention:  Red212:  A Content Strategy Agency

" ... Red212 was founded on the idea that business can be a powerful force for change. We are a full service, integrated marketing agency offering digital, traditional and cultural solutions that take brands to their boiling point.

The more successful your business is, the more you can do for our world. That’s why our mission, our passion, is creating measurable marketing campaigns so that your business can thrive.

At 211 degrees Fahrenheit water is merely hot. But by raising the temperature just one degree, to 212, change happens. Water boils. Steam is created with a force so powerful it can hurtle a locomotive across a continent.

That one degree that makes all the difference. And so we are …

RED212. An independent content strategy agency.

Red212 creates loyalty, market share and social impact by connecting like-minded consumers & fueling the conversation around your brand.

To truly engage today’s consumer, it takes an understanding of what motivates and speaks to them on a human level. Today’s media environment requires knowledge that informs the right message, for the right audience, at precisely the right time. It is at the intersection of data, consumer understanding, and creativity where engagement occurs and relationships are cultivated to inspire action.  .... " 

Making Wikipedia Visible

Intriguing challenge is how to make any data usefully visual.  Quant data has well established ways. Database structures have semantic networks.   Here is linking defined by hyperlinks within the Wikipedia.  How can we get value from these visualizations?   Some visual examples of Wikipedia at the link.

In the Wolfram Blog: by John Moore
Making Wikipedia Knowledge Visible

Over the past few months, Wolfram Community members have been exploring ways of visualizing the known universe of Wikipedia knowledge. From Bob Dylan’s networks to the persistence of “philosophy” as a category, Wolfram Community has been asking: “What does knowledge actually look like in the digital age?”

Mathematician Marco Thiel explored this question by modeling the “Getting to Philosophy” phenomenon on Wikipedia. “If you start at a random Wikipedia page, click on the first link in the main body of the article and then iterate, you will (with a probability of over 95%) end up at the Wikipedia article on philosophy,” Thiel explains. Using WikipediaData, he demonstrates how you can generate networks that describe this phenomenon.

He is able to document that about 94% of all Wikipedia articles lead to the “Philosophy” page if one follows the links as instructed, generating in the process some mesmerizing and elegant visualizations of the way that we categorize information.  .... " 

Thursday, November 24, 2016

China Has an Echo-like Device

The basic idea continues to spread in the Smart Home.  I am continuing to test the Google Home in a formerly all Echo environment.   See my curated list of assistants.  Will there be a race to more useful home intelligence?

China now has an Amazon Echo-like device called the ‘LingLong DingDong’ by Bryan Clark  in TheNextweb: 

" ... The DingDong, made by LingLong (this is all very fun to say) costs just $118 and provides answers, plays music and audiobooks, and manages schedules much like Amazon’s Echo. It’s operated largely the same way, to boot. To wake the device, one of three commands will work: DingDong DingDong, Xiaowei Xiaowei, or BaiLing BaiLing. Once awake, it’s capable of doing much of what the Echo does, only it’s currently clearly lacking in developer support. So far, LingLong reports, customers are primarily using it to play music from a library of about three million songs. .. .. " 

Bletchley Park to Become Cyber College

Been following the restoration of British code breaking and early computing site Bletchley Park for years.   Now it is slated to become a Cyber Education College.  In Mashable.  Good idea.

Wednesday, November 23, 2016

SAP HANA and Amazon Echo

Have always been interested in how there might be practical business uses of voice driven assistant systems.  I am sure there are examinations occuring in industry to test the idea.  I happened on this post in the SAP blog, regarding how to link Alexa to the  SAP HANA Cloud Platform, and securing its use.   Technical.   Just an exploration, it seems, but a starting point.  Voice interaction can be an efficient channel for assistance.  Note though that a voice driven system is fundamentally insecure.

" ... In this blog I would like to discuss some aspects concerning securing an Amazon Echo device using the Alexa voice service and SAP HCP.

Let me set the scene – you are creating a voice interface using an Amazon Echo device.  For now you want the sales team to use the Echo device to retrieve customer information from Hybris C4C, maybe list customers and then get some internal order (budget/expense) information from SAP S/4Hana. ... " 

Disney Canvas for Storytelling

We long wanted to see something like this.  For the non artist to express themselves in stories.  Is it easy enough?   Something closer to storyboarding might be best.  Makes sense this would come out of Disney.

More Than Animation: Software Supports Animated Storytelling
EurekAlert (11/17/16) Jennifer Liu

Disney Research has developed CANVAS, a computer-assisted tool for creating narratives, and Story World Builder, a graphical platform in which people can create "story worlds" populated with characters and props. The new tools are designed to eliminate the distracting details encountered when telling animated stories, which can hamper creativity. With CANVAS, users can synthesize a three-dimensional animation to get feedback on the narrative at any point, and it also can work in the background to fill holes in the plot to ensure the story makes sense. Researchers presented CANVAS this summer at the ACM SIGGRAPH/Eurographics Symposium on Computer Animation in Zurich, Switzerland. The researchers say Story World Builder simplifies the task of creating story worlds by leading the user through the process of defining characters and objects and of describing how events within those worlds unfold. Professionals as well as casual users could take advantage of the tools. "Our aim is to produce technologies to enhance the creative process," says Disney researcher Steven Poulakos. "We embrace several principles, including iterative design, reuse to minimize the cost of content production, and computer assistance to reduce complexity for creating story worlds and authoring stories."    .... 

More on this in Disney Research.


Slack Handling Your Mail

Still some time before we get to this but it would be nice. Fine tuning the details, and handling exceptions well will be key.  Current Slack is no where close.  In Inc: 

Slack Co-Founder on How Artificial Intelligence Could Eliminate the Need to Check Your Email Ever Again     ...  As Slack gets smarter, you might have to spend far less time sorting through your messages.  ..... By Kevin J. Ryan

Using IBM Watson in Business

This oversimplifies the task of using a capability like IBM/Watson, but is still a good introduction. Non technical, suggestive as to the possibilities.   Also an excellent list of companies using Watson.  Pointers to more detail.  By William Vorhies in DSC:

" ... In our recent reviews of historical Watson and the modern Watson of today we concluded that IBM’s Watson Group may have the first or at least the current strongest comprehensive AI platform. 

 This is the first time that we know of that all three elements of AI have been brought together in a single user friendly platform: image processing, text and speech processing, and knowledge retrieval. 

This is not so much a platform for data scientist to use to expand the capabilities of AI as it is a platform for business users (with the aid of data scientists) to exploit the capabilities of modern AI by building new products and services.

To wrap up this review of Watson, we wanted to provide some thought-starters on what new services or even new businesses you might build on Watson.

Oh, and regarding new businesses, did we mention that developers who join the Watson Ecosystem are eligible to become a Watson “partner” with a shot at the $100 Million funding IBM is making available to startups plus support and access from IBM business and technology advisors. ... " 

Blockchain for Tracking the Supply Chain

More depth on this effort.   Reliable tracking and trransparency creates potential data for many kinds of analytics.   And simply knowing where things are to recall them.

Walmart and IBM test blockchain for custom supply chain tracking
by Duncan Riley  In SiliconAngle: 

Distributed ledger blockchain technology may be be known for its original use for powering bitcoin and more recently for its financial technology implementations. Now, it’s being applied to supply chain management and tracking.

IBM Corp. and retailing giant Wal-Mart Stores Inc. have teamed up to utilize the blockchain to track Walmart’s pork supply chain in China. Using technology designed by the open source Hyperledger Project, software that builds blockchain tools, and run in conjunction with a private database co-developed by IBM, Walmart’s blockchain implementation is able to track pork as it moves from the farm to the shelves. It’s also being tested as a way to track packaged goods in the U.S. as well.

The use of the blockchain to track goods all comes down to product safety and the ability to identify and remove items that have been recalled. Currently, if a consumer becomes ill as a result of tainted food, a retailer or manufacturer can take days to track down the source of the product. By contrast, a blockchain based platform data can be easily tracked from the point of origin to the consumer themselves, allowing Walmart to identify items that need to be recalled and more importantly where the product issue may have occurred.   .... "

Tuesday, November 22, 2016

Machine Learning Performance Improvement

Nicely done, a number of simple improvement ideas, which are usually the best ...

Machine Learning Performance Improvement Cheat Sheet
by Jason Brownlee on November 23, 2016 in Machine Learning Process

32 Tips, Tricks and Hacks That You Can Use To Make Better Predictions. .... 

The most valuable part of machine learning is predictive modeling.
This is the development of models that are trained on historical data and make predictions on new data.

And the number one question when it comes to predictive modeling is:

How can I get better results?
This cheat sheet contains my best advice distilled from years of my own application and studying top machine learning practitioners and competition winners.

With this guide, you will not only get unstuck and lift performance, you might even achieve world-class results on your prediction problems.   .... " 

Predicting Demographics with Twitter

Always an interesting challenge.  Note the mention of Mechanical Turk, a method we examined.

In Techrepublic:
"  ... Big data can reveal inaccurate stereotypes on Twitter, according to UPenn study ... What can you guess about a person, based on their tweets? A new study examined how 140 characters can shape assumptions about a person's gender, age, education, and politics. Here's what it found.  ...  " 

By Hope Reese  | November 16, 2016

When Computers Become Smarter than Humans

Knowledge@Wharton looks at the often posed question.  The answer is, they already are.  First they did math better, then searched better, connected to people better, played chess better, stored data better  ...     But these are all relatively narrow areas. Becoming less narrow every day.  At what time do all of these areas combine into everything we do?    Even in these still relatively narrow areas. And is that then assumed to be 'smart'?  Smart,  even for humans,  is always in some context.

What if Computers Become Smarter Than Humans?  Nov 22, 2016

You’re ordering tickets to a play or a big sports event online. You’re almost done when that annoying Captcha screen comes up and makes you type some blurry letters and numbers into a box. This step, as most people know, is to ensure that you’re just a person buying tickets and not a computer program deployed to illicitly to grab up a bunch of seats.

But why can’t a computer that can perform calculations astronomically faster than humans identify the letter B just because it’s in a fancy font with a strikethrough, or the number 5 in a fuzzy photo of a front door? Why is it so easily baffled by something the average second-grader can handle?

The answer lies in understanding the current state of artificial intelligence (AI) — what it’s capable of, what is still beyond its grasp, and how we may be rocketing toward an increasingly intelligent technology without enough thought about the implications for ourselves and our planet. This is according to Tim Urban, author of the quirky, stick-figure-illustrated, popular blog Wait But Why, which counts Tesla CEO Elon Musk and Facebook CEO Mark Zuckerberg among its fans. He spoke recently at the McNulty Leadership Program’s Authors@Wharton speaker series.  .... "

Business Traffic in Real Time

We aimed to use public real-time webcam images to gauge traffic.   Google maps can now provide some of that same type of data, perhaps to estimate demand of products and resource use. Can see this being very useful in combination with other kinds of data.  Smart city applications as well.

Google Maps now tells you how busy places are in real time
by Eric David

Google Maps has offered its “Popular Times” feature for about a year now, allowing users to gauge how busy a business will be at certain times of the day based on past history.
Now, Google is making the feature a bit more useful — and creepy — by showing real-time stats using location data taken from people’s mobile devices.

“Just in time for the Black Friday swarms, we’re adding a real-time look at how crowded a place is right now, to help you decide where and when to go,” Jamie Aspinall, a product manager at Google, wrote in a blog post. “Whether you’re rushing to pick up a last minute gift or seeking a lively bar for some festive spirit, check Popular Times for a sneak preview of what to expect when you arrive.”

The new feature uses anonymized information from the Google Maps app that tracks where a device is located. More specifically, Google’s location data code tracks when users enter a predefined geographical boundary, for example a specific address or GPS location, and when they leave. Since Google Maps is one of the most popular map tools around, this effectively means that Google knows roughly where you are at all times. .... " 

Chatbots in Business

Not unlike digital assistants, but usually more primitive to date.  Usually act in a conversational mode in an existing stream of interaction.  Use AI, natural language anlysis and pattern analysis to link to knowledge in a given context.

AI chatbot apps to infiltrate businesses sooner than you think  by  Bridget Botelho  ... Editorial Director

Artificial intelligence chatbots aren't the norm yet, but within the next five years, there's a good chance the sales person emailing you won't be a person at all. .... " 

Monday, November 21, 2016

Process Innovation with Data in a Digital World

Have always thought in terms of process.  By doing that you can think not only what a process produces in terms of value, but also in terms of data that can be used for further examination and improvement of how things get done.    Even just mapping the process generally can provide  actionable value ....

This Bain article By Chris Brahm and Peter Guarraiais is insightful about an overview of the approach, and makes my  point.  Good complete read:

" ... Remember when the typical reengineering approach boiled down to eliminating, simplifying and automating processes? Seems quaint at a time when digital technology allows companies to transform how they operate, from one end of the value chain to the other. Breakthroughs in areas such as communications technology, analytics, Big Data and the Internet of Things have ushered in an entirely new set of tools to generate more value for customers. The goal of reengineering hasn’t changed, but the means have become significantly more powerful.   ... " 

AllRecipes Launches 60K Recipes for Amazon Echo

Allrecipes launches a 60K recipe file skill for the Amazon Echo, and other Alexa driven systems and devices.    Great idea.  Certainly useful for potential hands-free activity of cooking.  I have yet to look at it.

But it brings up some general questions.  Just being hands-free is not enough for a voice driven cognitive system.   When you use recipes its not just about choosing and hearing a recipe.  Its about pausing while you perform a task.   Its also about scanning an ingredients list, considering adaptations, looking at a picture of what its supposed to look like.  None of these necessarily lend themselves to a sound and voice based system, unless the user interface supports typical cooking interactions.   We saw similar issues when planning maintenance type systems using augmented reality.

Will be checking out the Allrecipes Skill system.    And have more to say about the usefulness of the approach.   More on the announcement

( This was at first not available, but now as of 11/26/2016 is launched and working )

Table Delivery from Kiosk Orders at McD's

The test is now over, and rollout is starting to all 14K US stores.   Benefits seen were lower perceived wait times,  higher average order value,  about $1 per order.   More and better interaction with employees.   No mention of decreased labor required, or fewer payment options at kiosk.   Is robotic cooking/delivery far away?   Given the ubiquity of McDs, this could also end up 'training' the public to use and expect kiosk interaction more broadly in retail.   We experimented with kiosk interaction in retail.   Good expert comments at the Retailwire post:

Will table service be a difference-maker for McDonald’s?   by Tom Ryan
McDonald’s last week announced plans to roll out table service to all of its 14,000 U.S. restaurants.

After customers place their orders at a touch-screen kiosk and pick up a digital location device, orders will be delivered to their tables. Orders can still be made from the counter or picked up at the counter after using the kiosk.  ... " 

Strong Networking

In Inc:

5 Simple Habits of Strong Networkers
Implement these habits, and you'll go from sweaty-palmed to self-assured. By Kat Boogaard  ... 

Fairly obvious, but worth repeating, am always surprised when I discover people that do not network well.   Don't follow up well.  Today it's so easy and fast.  

Walmart Tests Blockchain to take Unsafe Food From Shelves

An unexpected application.  Blockchain in retail compliance. In Engadget:

Walmart tries using blockchain to take unsafe food off shelves
It could pull bacteria-laden produce before you have a chance to buy it.    by Jon Fingas 

It's scary when stores have to pull unsafe food. It can take days to pinpoint the batch or supplier that caused the problem, which could lead to people getting sick... or at best, force the store to yank more products than necessary. Walmart believes technology could offer a better way. It's trying out blockchain's distributed ledger as a means of identifying the sources of poisonous food. Digital receipts for the food identify everyone in the supply chain, ranging from the original location through to the inspectors and shipping companies. If everything passing through a specific warehouse is making people ill, Walmart can identify that troublesome food and yank it within minutes -- and there's no easy way for unscrupulous suppliers to alter the info. ... " 

Sunday, November 20, 2016

Webcams and Maps

My writing about the famous coffee pot Webcam, and its internet of things implications, reminded me of work we did looking at the output of public webcams as data sources for shopping traffic and activity.  Different than the gathering of data in our lab, linked to work with IBM,  but still a source of public data to combine with private.  In 2009 Google announced a layer of public webcam to be added to maps.   Regrettably later this effort was removed and added to a separate map effort.  There are still many directories of Webcams still online.

Augmented Beauty

Another look at an old idea.  Been connected with a half dozen attempts.  Will augmented reality tech finally make this work?  See my tags below for more examples.

Augmented Beauty: MakeupPlus Allows Users To Try New Looks In Advance  Instantly try on a new look with this beauty product application.   By Kevin Eta, 

We’ve already had one use of augmented reality (AR) today with Vuzix’s Blade 3000 Smart Sunglasses but what about an AR assistance not in the workplace but for your personal look. That’s the aim of a new AR application from developers Meitu called MakeupPlus that will allow people to try on new looks without even having to reach for the concealer. ... " 

Faces of Customer Education

So what is the best way?   Advertisements?  Blogs?  Twitter Streams?  Have been operating with all of these for some time,  but what is engaging and efficient?  Using which components?  Is it really just education?    Good breakdown.  In CustomerThink.

The Faces Of Customer Education: 3 Proven Strategies     by Larry Alton   

Customer education is a powerful tool. It can increase clients’ engagement with your company and its products, build relationships, and provide an added value component to your service.

It can also increase customers’ trust in your firm. But how can you offer education to your customers most effectively without coming across as pushy or making it seem as if they don’t know how to engage with your products?

Presenting valuable knowledge isn’t necessarily an easy task. Within the larger framework of customer education, however, there are a few strategies that seem to do well consistently, including on-site reference content, social media posts that directly engage customers, and podcasts.

Ultimately, it’s about finding ways to be friendly and helpful to onboard customers and track their training in a way they’ll hardly notice. Here are some ways to do that.  ... " 

Overcome Resistance to Change by Enlisting the Right People

Just completed teaching a course at Columbia University on Change Management in Analytics.  Used some thoughts from this article.   Good thoughts.

Overcome Resistance to Change by Enlisting the Right People

Todd WarnerSeptember 13, 2016 Harvard Business Review

Change initiatives fail when senior leaders don’t see that their organizations are social systems. Organizations are defined by bundles of local relationships, social expectations, and unwritten rules that exist between thousands of people. These people create tribes — collections of like-minded people who build patterns around how work gets done. This system is highly resistant to change, as anyone who as ever tried to “impose” change on a system knows.

Tribes in organizations tend to be small, and may cut across formal silos; they have their own power structure, and their own ways of rewarding and punishing deviance.  You see this in on-boarding, for example, where new joiners take cues and align with co-workers, replicate work habits, and learn day-to-day, despite organizational attempts to “teach” through formal on-boarding programs.

The secret to changing an organization is to understand the fundamental units that make up the social system — these local tribes — and to invert the change process so that tribes own the change. To influence tribes in organizations, you have to give up control, and recognize that every change always goes through a process of localization as it gets executed.   ....  " 

Saturday, November 19, 2016

AI, Robots Won't Try To Kill Us

I mentioned this Stanford study when it came out.   Worth the read.  Now Fast Company takes a closer look.   'Try' is the operative word. Long view is right, with some interesting but hard to calibrate forecasts.

Robots Won't Try To Kill Us, Says Stanford's 100-Year Study Of AI
Stanford University is taking the long view on the future of AI with an ambitious project that will unfold throughout the century.  .... " 

Minds, Machines, Precision and the Internet of Things

Timothy Chou writes:

" ... On the heels of the successful Minds+Machines 2016,  the IoT Institute has launched a class, which uses my new book, Precision, as the textbook.  It's online, video, easy-to-consume and complimentary for the next 90 days. Register at:

www.crowdstory.com/precisiontheclass

On completion they will receive a certificate.   ....   Thanks,   Tim

p.s. We launched the book Precision: Principles, Practices and Solutions for the Internet of Things (https://amzn.com/1329843568) in London this summer. One of the chapters is available at https://www.ibm.com/blogs/internet-of-things/precision-race-car/.   Penton called it one of the best books on IoT (http://www.ioti.com/iot-trends-and-analysis/50-top-iot-authorities-twitter)  and SAP did a really nice nice video (https://www.youtube.com/watch?v=TxYu9qlk1CM). Foreward by Bill Ruh.   .... " 

Paro Therapeutic Robot

A friend, who works in home health care, sent me a note about the Paro Therapeutic Robot.  She saw it live and was impressed.  Comes out of work in Japan, where related work in eldercare has been going on for decades.

PARO Therapeutic Robot
PARO is an advanced interactive robot developed by AIST, a leading Japanese industrial automation pioneer. It allows the documented benefits of animal therapy to be administered to patients in environments such as hospitals and extended care facilities where live animals present treatment or logistical difficulties.

PARO has been found to reduce patient stress and their caregivers
PARO stimulates interaction between patients and caregivers
PARO has been shown to have a Psychological effect on patients, improving thier relaxation and motivation
PARO improves the socialiazation of patients with each other and with caregivers
World's Most Therapeutic Robot certified by Guinness World Records
PARO is the 8th generation of a design that has been in use in Japan and throughout Europe since 2003.

PARO has five kinds of sensors: tactile, light, audition, temperature, and posture sensors, with which it can perceive people and its environment. With the light sensor, PARO can recognize light and dark. He feels being stroked and beaten by tactile sensor, or being held by the posture sensor. PARO can also recognize the direction of voice and words such as its name, greetings, and praise with its audio sensor.    .... " 

New Chip Usage to Power AI

Re-using chip designs to perform the methods used for AI applications.  Now new chip designs are in play.

Intel Looks to a New Chip to Power the Coming Age of AI
Microsoft Researchers recently built an artificially intelligent system that seems to recognize conversational speech as effectively as a human. Yes, this research comes with caveats, but it’s part of a very real and very rapid leap in artificial intelligence over the past several years, a leap driven by deep neural networks.

These sweepingly complex algorithms can teach themselves very particular tasks by analyzing vast amounts of data. Microsoft’s system learned to recognize words by looking for patterns in old tech support calls. But it’s not just the algorithms that are driving the recent revolution in AI. It’s also the hardware behind these algorithms. Microsoft’s speech rec system relies on large farms of GPU processors, chips that were originally designed for rendering graphics but have proven remarkably adept at running artificial intelligence models. .... 

Internet giants like Microsoft, Google, Facebook, and Baidu typically train their deep neural nets using GPUs. But they’re moving towards other, more specialized chips that can help accelerate not just the training but the execution of these systems. Google recently built its own AI processor. IBM is building another.

So Intel, the world’s largest chip maker, is doing the same. Yesterday, the company unveiled a new AI processor called Nervana, saying it plans on testing prototypes by the middle of next year, and if all goes well, a finished chip will come to market by the end of 2017. At the moment, the market for AI chips is dominated by nVidia, the primary supplier of GPUs.  ... " 

Recommender Systems Beyond Matrix Completion

Recommender Systems: Beyond Matrix Completion/Abstract
By Dietmar Jannach, Paul Resnick, Alexander Tuzhilin, Markus Zanker
Communications of the ACM, Vol. 59 No. 11, Pages 94-102
10.1145/2891406

Abstract: The use of recommender systems has exploded over the last decade, making personalized recommendations ubiquitous online. Most of the major companies, including Google, Facebook, Twitter, LinkedIn, Netflix, Amazon, Microsoft, Yahoo!, eBay, Pandora, Spotify, and many others use recommender systems (RS) within their services.... 

These systems are used to recommend a whole range of items, including consumer products, movies, songs, friends, news articles, restaurants and various others. Recommender systems constitute a mission-critical technology in several companies. For example, Netflix reports that at least 75% of its downloads and rentals come from their RS, thus making it of strategic importance to the company. ... " 

Requires registration to see full aricle.

Friday, November 18, 2016

Putting Coffee Machines on the IOT

It has long been a challenge to connect coffee machines to the internet.   Something comfortable about making coffee available reliably at all times and letting us know it.   Even inspiring the worlds first webcam in 1991.  Now we are seeing it commercially, clearly a better way

 Nestlé picks Telefónica to build IoT for coffee machines by Alex Scroxton
Nestlé’s professional business services unit will incorporate internet of things technology into its coffee machines in pursuit of greater efficiency and customer satisfaction

International food and catering services supplier Nestlé has chosen Telefónica Business Solutions to provide a 50-country internet of things (IoT) communications network for its Nestlé Professional coffee business as part of a major digital transformation.  ... "  

New Products for a CPG Giant

I no longer work for consumer product CPG Giant Procter & Gamble.    I did analytics and competitive analysis innovation work for them for years.  I still have quite a few readers there,  I still own some of their stock,  so this post is particularly aimed at them.    Your global competitors are doing some interesting things to develop new product ideas.   See the article linked to below:

Lysol owner and Indiegogo team up to find the next... Lysol
Crowdfund a new health product and you might be discovered by a major chemical company. ... " 

Gartner on Augmented Marketing Reality

In the Gartner Marketing Blog.  Good piece,   See my posted piece yesterday on an impressive Hololens demo. The idea is a decade old, we tested it then, but the technology has now evolved to usability now.  It will require some big events to launch it, and Pokemon was not enough and faded too fast. ....

Our Augmented Marketing Reality
By Noah Elkin | November 18, 2016 

Chances are that you’re one of the 100+ million users who have downloaded and experimented with Pokémon Go. But even if you haven’t thrown a Poké Ball into a stranger’s yard to catch a Charmender, you’ve no doubt witnessed friends, family members or coworkers heads-down in their smartphones, engrossed in Niantic’s popular augmented reality (AR) game.

If so, you’ve caught a glimpse of the future of commerce. By the end of 2017, Gartner predicts (client subscription required), one in five leading global retail brands will use AR to enhance the shopping process — by layering digital information — text, images, video and audio — on top of the physical world. .... " 

What Does it Mean to be Digital?

Some good ongoing posts on the process of digitization in the Cisco blog.    Asks 'So what are you trying to do?    The ultimate business process question.    And I would add:   How is it being done today?   By you and your competitors?  And how do you define competitors?  .... 

Teaching Algorithms Right vs Wrong

Do we know what is right and wrong?   In obvious cases, yes, but am starting to wonder if the label can be much abused.  And the underlying mechanism,  hidden in an algorithm, is difficult to check or test.  So this appears to be a tough call for the average consumer.

Teaching an Algorithm to Understand Right and Wrong  by Greg Satell.  In the HBR.  Good discussion.
"  .... Today, as we enter a “cognitive era” of thinking machines, the problem of what should guide our actions is gaining newfound importance. If we find it so difficult to denote the principles by which a person should act justly and wisely, then how are we to encode them within the artificial intelligences we are creating? It is a question that we need to come up with answers for soon. .... " 

AT&T Offers Limited Text Chat via Alexa

A first, integrating multiple communications channels via Alexa.  Here from AT&T.  Still quite limited, 10 contacts only, but I like the direction.   Closer to a more universal assistant.  Lets see better bot interaction with the chat.   I use hands-free calendar schedule set up frequently using Echo.  Now would like to see further VOIP and text capabilities hands-free.

Amazon’s Alexa will be able to send texts for AT&T customers starting tomorrow by Chris Welch

For when you’re “busy with your hands”

AT&T just announced that beginning tomorrow, its mobile customers will be able to compose and send text messages using Amazon’s Alexa assistant. So if you’ve got an Echo, Echo Tap, or Echo Dot somewhere in the house, you can ask Alexa to “have AT&T text” a contact’s name (Alexa can store up to 10), followed by your message. You’ve got to take that first step of adding people to the skill before Alexa can do anything; you can’t just call out the name of anyone in your contact list, for example. ... " 

Via Walter Riker.

Bots in the Azure Cloud

More capabilities for building Bots.   I think the definition is still too loose.

 Microsoft launches Azure Bot Service to bring more bots and AI into the cloud
 by Khari Johnson  in VentureBeat.
 .
Microsoft launched its Azure Bot Services today to give bot creators the option to move their bots into the cloud and let Microsoft handle server and storage issues, the company announced today.

Alexa Web Service is cloud-based, but until today, none of the big three cloud providers — Microsoft Azure, Amazon Web Services, and Google Cloud Platform — have provided a solution for developers to create bots in the cloud.

The service was created, wrote Microsoft FUSE Labs general manager Lili Cheng in a blog post today, to “give you the operational agility to run and handle scale as your bots grow in popularity and you only pay for what you use. Moreover, you also don’t need to worry about provisioning or managing servers that run your bots. Patching and infrastructure maintenance is handled for you — you focus on writing code.” ... ' 

Thursday, November 17, 2016

Hololens Moving Towards Real World Applications Beyond Games

Saw today a video demo from a Dutch company called Wortell, which is a Microsoft partner developing 'practical' applications of the Hololens.

Using head mounted displays that let you see-through to the real world while navigating and interacting via gestures with augmenting objects. Comfort and ease of use are claimed to be much improved.  No motion sickness was discerned in the demonstration.

The demo can be seen at this Windows Weekly show, starting at about minute 33:50.  In English.

Includes a real time demo view of data using a feed from Schipol airport as a kind of air controllers view of the airport using gesture interaction.  Based on this you can see the potential for maintenance, design and data interaction applications.

The software is from Wortell, their web site is in Dutch at this time.

Turning Data into Leverage

You have data, there is pressure to gather and archive more.  What is its value?

Turning data into leverage
Ozan Turgut discusses how to use visualization and analytics to apply data to decision making.

Video by Ozan Turgut September 22, 2016

Medical Device Security

In the CACM:  Was brought directly to my attention recently.   This is video history, but useful as an introduction as well.  Covers well beyond data security, but also physical security due to hacking.

This Blog Gets more Security

This is useful for better security.  I have verified its operation. Google informs me:

" .. Your HTTPS settings have changed. All visitors are now able to view your blog over an encrypted connection by visiting https://eponymouspickle.blogspot.com. Existing links and bookmarks to your blog will continue to work. Learn more.  ... " 

Examples of Python Implementation of Machine Learning

Jim Spohrer retweeted  by  Randy Olson @randal

This is usefully instructive
Minimal and clean #Python implementations of #MachineLearning algorithms. #DataScience

https://github.com/rushter/MLAlgorithms

Wednesday, November 16, 2016

Gamification as Tactic

Has started to come up more often as a tactic to deliver results.    A long followed approach.  Make sure you have the values and the culture of the participants right.  In Retailwire: 

Gamification reaches the next level
by Guest contributor  COLLOQUY staff

Through a special arrangement, what follows is a summary of an article from COLLOQUY, provider of loyalty-marketing publishing, education and research since 1990.

For some brands, gamification has emerged as a productive tactic in an era when keeping customers engaged has become increasingly difficult. Though the approach has been in practice for years, the new wave of brand-based gaming builds on learnings from previous campaigns.

Recent launches by major brands demonstrate increased sophistication, investment and commitment to engaging consumers:  ... " 

Information Extraction


Dear Professors, Students, IBMers and others, 

This a reminder about our Cognitive Systems Institute Group Speaker Series tomorrow, Thursday, November 17, 2016 at 10:30 am ET US (7:30 am PT).  Our presenter is Laura Chiticariu, from IBM Research - Almaden, who will be presenting "Declarative Multilingual Information Extraction with SystemT.“

Please point your web browser to https://apps.na.collabserv.com/meetings/join?id=2894-8491 password=cognitive.   Use audio on computer or dial 855-233-7153 (other countries numbers here) PIN Code: 43179788    Non-IBMers please use the "guest" option instead of entering your email.   

Slides and recording.  (may or may not be posted before the talk) 
Followup Discussion.

Thank you and please let me know if you would like to be a presenter on an upcoming call!  

Dianne Fodell

IBM Global University Programs
fodell@us.ibm.com  ... " 
   

Business Process for Analytics Modeling

I Posted in the Wiki for my Columbia Class:

Business Process Modeling ... Franz Dill

I have often mentioned business process modeling (BPM) in my case study comments.  It is often useful to understand how the business works. .  It lets you understand how the parts you want to improve connect, and where to insert the analytics or intelligence needed.  Its also a good way to communicate your plans to management.   It can also be used as a means to simulate solutions.   There are formal ways to do BPM, with specific computer packages, or informal using concept maps to sketch out the process.  I always use process mapping before doing a project.  In the enterprise we cataloged analytics systems via their process models.

More in my posts on business process mapping (BPM).

Its Not Always About Quick Wins

It used to be an old saying that you should look for quick wins, so that the analytics team could establish itself. Yes, not a bad idea, but better yet look for wins along the way to consistent value, and creating an organization that learns to win.    Big wins are good too.   And small quick wins on the way to big ones.  Things you can learn from and repeat.  Techtarget also discusses.  

Tuesday, November 15, 2016

What AI can and Cannot Do Today

Useful thoughts by a practitioner.  Which gets you back to the term AI again.  It is still a considerable overreach.    Augmenting, improving, extending, replacing and assisting our work is still the best view today.  Which is still not universally easy.   And also, when is now and when is our dreamed of future?   Have been asked the question by many people already.

What Artificial Intelligence Can and Can’t Do Right Now   by Andrew Ngmate  

Many executives ask me what artificial intelligence can do. They want to know how it will disrupt their industry and how they can use it to reinvent their own companies. But lately the media has sometimes painted an unrealistic picture of the powers of AI. (Perhaps soon it will take over the world!) AI is already transforming web search, advertising, e-commerce, finance, logistics, media, and more. As the founding lead of the Google Brain team, former director of the Stanford Artificial Intelligence Laboratory, and now overall lead of Baidu’s AI team of some 1,200 people, I’ve been privileged to nurture many of the world’s leading AI groups and have built many AI products that are used by hundreds of millions of people. Having seen AI’s impact, I can say: AI will transform many industries. But it’s not magic. To understand the implications for your business, let’s cut through the hype and see what AI really is doing today. .... " 

Alexa Gives Birth to Yumi?

At Indiegogo: Unclear how much different this is than the Echo, or Google Home  Except that it includes embedded screen and camera and video calling.  This allows applications like surveillance and video calling.  shows how people are taking up the framework of existing assistant systems for specific robotic applications.  Here in the home.  Beyond to the office?  Interesting technical details at the link.

Omate Yumi - The Most Advanced Home Robot

Omate Yumi - The World's First Family Robot Powered by Android with Amazon Alexa

Meet Yumi, the World's First Home Robot powered by Android and Amazon Alexa Artificial Intelligence. Yumi can hold a conversation with anyone in the family, kids and adults alike. Yumi has been built on AI from the ground up; it is an incredibly smart and cute robot, a personal assistant, a music hub and a smart-home control point all in one. ... " 


A Learning Analytical Organization

I wrote for my Columbia Course:

A Learning organization is one of the fundamental lessons of this course. Our world is changing quickly, so our methods and implementations must adapt.  Business focus, competition, business process, finance, technology, risk and likely rewards are all rapidly evolving.  So our implementations, and how they are integrated must change as well.

That learning can make our methods more efficient to implement, and produce more value. A learning organization adapts to all parts of its environment. That learning can come from academia, publications, consulting, and online resources. 

And also importantly from continually retesting our proposed analytical methods, to make sure they continue to be correct under change. 

A learning analytics organization also learns from previous projects. that can only happen if they have been well documented, cataloged and made readily available to colleagues. .... 

AI Based Search for Research

Making search better with semantics ...

New AI-Based Search Engines Are a 'Game Changer' for Science Research
Scientific American (11/12/16) Nicola Jones 

Artificial intelligence (AI)-based academic search engines such as Semantic Scholar and Microsoft Academic could transform scientific research and inquiry, according to proponents. Semantic Scholar from the Allen Institute for Artificial Intelligence (AI2) is designed to sort and rank academic papers with more refined content and contextual understanding than keyword-reliant search engines. Stanford University neurobiologist Andrew Huberman calls Semantic Scholar a "game changer," noting "it leads you through what is otherwise a pretty dense jungle of information." The AI2 search engine's creators say they are growing its database to encompass about 10 million research articles, mostly on computer science and neuroscience. Meanwhile, Microsoft Academic was released in May as a replacement for Microsoft Academic Search, and Microsoft Research's Kuansan Wang contrasts the tool with Semantic Scholar in several respects. He notes Semantic Scholar is more deeply invested in natural-language processing to drive searches, while Microsoft Academic, powered by Bing's semantic search capabilities, covers far more publications--160 million. Wang says the tool's recursive algorithm evaluates the most influential scientists in each sub-discipline according to whether their papers are cited by other important papers. Microsoft Research says the development of a personalizable version of Microsoft Academic also is underway ... " 

Monday, November 14, 2016

Cancer Research with Deep Learning

In the CACM:

" .... Oak Ridge National Laboratory (ORNL) researchers are applying deep-learning techniques to automate how information is collected from cancer pathology reports documented across a nationwide network of cancer registry programs.

Georgia Tourassi, director of the Health Data Sciences Institute at ORNL, led a team focused on software that can identify valuable information in cancer reports faster than manual methods.

The machine-learning technique leverages algorithms, big data, and the processing power of the Titan supercomputer at the Oak Ridge Leadership Computing Facility. Using a dataset of nearly 2,000 pathology reports, researchers trained a deep-learning algorithm to simultaneously carry out two closely related tasks. .... " 

Learning Lessons from Building Bots

Thoughtful video from recent conference.   I am continuing to look at bot style solutions.  Still not very impressed with solutions I have seen.

Lessons learned from building bots
Andy Mauro discusses pitfalls, opportunities, and the future of conversational bot interactions.

Video by Andy Mauro November 14, 2016.

On the Ethics of AI

In O'Reilly.  Well put.  Its not required right away, but we need to think about it because it is moving into some areas where compliance and ethics are important.

The ethics of artificial intelligence, A framework for thinking about AI.

By Mike Loukides November 14, 2016

In this industry, it's a tired old cliche to say that we're building the future. But that's true now more than at any time since the Industrial Revolution. The proliferation of personal computers, laptops, and cell phones has changed our lives, but by replacing or augmenting systems that were already in place. Email supplanted the post office; online shopping replaced the local department store; digital cameras and photo sharing sites such as Flickr pushed out film and bulky, hard-to-share photo albums. AI presents the possibility of changes that are fundamentally more radical: changes in how we work, how we interact with each other, how we police and govern ourselves. ... " 

Robots Working Together


A topic of interest that addresses applications like cooperation and teamwork in robots.   Working with each other or with people.

From GaTech:

Team of Robots Learns to Work Together, Without Colliding
Georgia Tech News Center (09/06/16) Jason Maderer 

New algorithms created by Georgia Institute of Technology (Georgia Tech) researchers enable robots to move within inches of each other, without colliding, to complete their task. The team, led by roboticist Magnus Egerstedt, director of Georgia Tech's Institute of Robotics and Intelligent Machines, is the first to create such minimally invasive safety algorithms. A set of safe states and barrier certificates ensure each bot stays in its own safe set throughout the entire maneuver, the researchers note. In a demonstration involving four robots, the machines approach from four different areas, meet in the middle, circle counterclockwise within inches of each other, then fan out into opposite directions. In another demonstration, eight robots perform the same task, this time circling clockwise before dispersing. Instead of keeping their distance and taking the long way around their neighbors, the robots move very independently wherever they wish. "Robots are very conservative--they want to make sure they're safe," Egerstedt says. "You couldn't pack the interstate with self-driving cars with today's technology." He says a minimally invasive safety controller similar to the algorithms could be used for the next generation of air traffic control.  ... " 


IBM Plans for Watson's Future

Watching this closely.  That they are continuing to plan for hardware is interesting.

As Watson matures, IBM plans more AI hardware and software
IBM releases faster hardware and software that make it easier to train deep-learning systems  ... "

Sunday, November 13, 2016

Standards for Body Area Networks

Examined for possible wearable applications, and medical sensors and effectors.  

IEEE releases new standard for body area network
After five years of work, the IEEE announced a new standard, IEEE 802.15.6, for wireless communications supporting ultra-low power devices operating in or around the human body.
By Chris Jablonski for Emerging Tech  .... 

Even Physicists Fear Mathematics

 Well yes,  Because math is a deeper abstraction.   Not always connected to the see-able or tangible.  But also a very useful tool.   But a tool that needs to be reworked for the most difficult problems.  And that reworking then requires re-explaining. .... A barrier.

Even physicists are 'afraid' of mathematics
November 11, 2016
Physicists avoid highly mathematical work despite being trained in advanced mathematics, new research suggests.

The study, published in the New Journal of Physics, shows that physicists pay less attention to theories that are crammed with mathematical details. This suggests there are real and widespread barriers to communicating mathematical work, and that this is not because of poor training in mathematical skills, or because there is a social stigma about doing well in mathematics.  ... " 

Global Technology

50 notable global technology companies.

The New Game of Global Tech
As distinctions dissolve among the hardware, software, services, and telecom sectors, companies are changing how they compete. ,,,, " 

by Olaf Acker, Florian Gröne, and Germar Schröder .... "

Non Technical View of Machine Learning

Machine Learning: An In-Depth, Non-Technical Guide - Part 1
By Alex Castrounis • Jan 27, 2016

Chapters
Overview, goals, learning types, and algorithms
Data selection, preparation, and modeling
Model evaluation, validation, complexity, and improvement
Model performance and error analysis
Unsupervised learning, related fields, and machine learning in practice   ... " 

Sports Analytics not Always a Slam Dunk

In our Columbia course we have been using sports analytics as a disruptive example of the analytics technology.  Movies like 'Moneyball' imply it is fail safe, but of course it is not.  For my students and further reference to cautions  This piece addresses some of the issues.

Future of Data Warehousing

This came up in a recent course I gave.  What are the implications of better and more intelligent data availability?   Better process decisions based on data.   In DSC.  " ... To Thrive in the Age of the Customers – Businesses Must Merge Previously Separate Efforts ... Now that we are seeing real-time and streaming data, it is more important than ever before to create cohesive strategies for business insights. This means merging formerly separate efforts like traditional BI, agile BI, and big data.  ... " 

Cognitive Disrupting Enterprises and the C-Suite

Why is clear, but how and the implications are still unclear.  What are the all the dimensions of disruption.  What will the clearest use cases be? Likely increasingly valuable and engaging assistants.

AI, Cognitive Computing To Disrupt Enterprises: IDC

IDC is forecasting big growth for cognitive computing and AI in the next 5 years. This infographic shows the growth, industries, and use-cases for these technologies.

What used to be science fiction is now an accepted path for IT. Multiple IT analyst firms are predicting that artificial intelligence technologies will become important components in future IT organizations. Indeed, at the Gartner Symposium, AI was simply another accepted factor in almost every system that will be created in the next decade   .... " 

Saturday, November 12, 2016

(Update) First Look at Google Home Assistant

As part of my long running look at virtual assistants, started to look at Google Home, in comparison to Amazon Echo for smart home applications.  Compared to my nearly two year to date view of the Echo.

Excerpt of my notes: 

 .... Google Home does have better bass in its own speaker,  can be bluetooth linked to other speakers but have not tried that.    Has some native podcasts,  but not many.   Has a number of integrated news casts, but no Tunein, (Yes it does have Tunein now, but not their podcasts)  which I have gotten used to for streaming local news.  Needs considerably more useful content.  In general is slightly slower to react to a command than Echo, annoying once  you have gotten used to echo.   Has a female voice very similar to Alexa, but a little less upbeat and less smooth.   When Google does language translations, words or phrases, another, different voice takes over .... which sounds odd ... that voice speaks too quickly, which does not help, since you are implying you don't know the translation .... and probably need it to be slower for recognition and repetition.  Does from-English, but not to-English translations. ... 

   Google/She has a similar problem in pronouncing constructed text, she trips up on odd words and proper nouns where general pronunciation rules don't apply.  She does do a better job of maintaining a conversation ... that is if you ask for some information during a stream of music, she goes back to the stream.  Echo often loses what it was doing originally.  Still have not seen obvious examples where she is going deeper or remembers context of a previous conversation.  Which would be hints at 'intelligence'.  Or the use of your Google Knowledge Graph, except for the simplest things like location.   Assume this is developing.  Or use of 'common sense' semantics'.  Still seems very search oriented.  She still can't create a calendar entry by voice, though she links to Google's calendar.     All interactions show up in text on the App, which is good for reference.      ......  

Is Samsung's Otto next? No pedestal Siri in sight.

See Google Home site.  

Friday, November 11, 2016

Real Time Shelf Analytics

In Retailwire, requires sign-in information to get the report.

Aberdeen Study: The Power of Real-Time Analytics on the Shelves
This study, produced by Aberdeen Group, highlights the impact of real-time analytics. See why top performing brands are 50% more likely than others to use real-time technology to empower their back-office with the actionable insights needed to drive operational efficiency, increased sales, and superior customer experiences.

Download the study to see why achieving marketplace performance gains requires that brands empower their back-office with mobile analytics. ... " 

Virtual Advertising

 An area we looked at when it seemed like virtual worlds would become common.   Augmentation can have an ad floating in front of us wherever we are.  In AdAge: 

Virtual Advertising: Uncharted Legal Waters  By Jim Gatto.

As augmented reality and virtual reality applications continue to proliferate, virtual advertising and related legal issues will grow too.

AR applications layer information or graphics on a real-world view. As such, AR could enable a virtual ad to cover advertisements that exist on billboards and other structures, or insert an ad where no real-life one exists. Both uses raise legal questions for creators and brands using AR technology.

In VR, scenes are often created to mimic real-world locations. Famous buildings and locations can be depicted, and virtual ads may be incorporated. Some of VR's legal issues mirror well-known ones, such as depicting actual products in movies, but there are new twists.  ... " 

Thinking About Cooperation

What the implications between people, machines, assistants?   Can we include a cooperative ethic inside an assistant?

The Cost of Cooperating

A Conversation With David Rand ... 

Why is it that we care about other people? Why do we have those feelings? Also, at a cognitive level, how is that implemented? Another way of asking this is: Are we predisposed to be selfish? Do we only get ourselves to be cooperative and work for the greater good by exerting self-control and rational deliberation, overriding those selfish impulses? Or are we predisposed towards cooperating, but in these situations where cooperation doesn't actually pay, if we stop and think about it, rationality and deliberation lead us to be selfish by overriding the impulse to be a good person and help other people?     

DAVID RAND is an associate professor of psychology, economics, and management at Yale University, and the director of Yale University’s Human Cooperation Laboratory. David Rand's Edge Bio Page