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

Thursday, December 31, 2015

Cognitive Sciences Exchange

An intriguing piece in Mindhacks.   Which points to cognitive science resources that might be useful for building systems that aim to use cognitive models.    Always question cognitive-social models, compare them with alternatives.   This site might give you such alternatives.

Topic Modeling in R

Topic Modeling.  A kind of textual analysis we did long before R, using cluster based 'Content Analysis'.  An  examination and R code for how this can be done in the modern era. An example of unstructured text analytics.     Their definition:  " .... A statistical approach for discovering “abstracts/topics” from a collection of text documents based on statistics of each word. In simple terms, the process of looking into a large collection of documents, identifying clusters of words and grouping them together based on similarity and identifying patterns in the clusters appearing in multitude. .. " 

Linkedin Open Source Success

Despite having had a number of UI issues, you have to admire a company with over 400 million users.  A  system that I use by choice most every day.    Also one that needs to continue present a continuing evolving face on professionalism.  They talk about how they use open source.  Have heard that from many enterprises recently.

LG Announces Smart Home Hub

Looks like an Echo cylinder with a small screen and speaker.  But is not voice activated. AI / Cognitive?  In the past have been impressed by LG's experiments with smart appliances, we used and tested several in innovation settings.  before they could be networked.  Note the use of the Alljoyn system,  which has been mentioned for inclusion in Windows 10.   More next week at CES.

In the Verge: 
" .... The hub, announced today ahead of a demonstration at CES next week, is visually similar to Amazon's Echo, and like that device, features a large built-in speaker that can stream music. But where the Echo is only gradually moving from cool curio to useful gadget, LG is pitching the SmartThinQ as the "gateway" to your automated home, connecting to smart refrigerators, washing machines, and air conditioners, and displaying their information on a 3.5-inch color LCD display. ...  The hub uses the open AllJoyn "internet of things" platform, connecting to other appliances — both smart and conventional — via Wi-Fi, Bluetooth, or Zigbee ....  " 

Wednesday, December 30, 2015

Machine Learning is Optimization

It is nice to be reminded that the math part of Machine Learning is optimization.  Because that was my original training, in the form of 'Operations Research'.   Course its not all about math.  Its also about decision process, data and understanding constraints.   And formulating the right model.  See below.  Nicely put.  Its NOT just finding a magic formula.  It is establishing a complete context of the problem at hand.   O'Reilly also introduces the open software framework KeystoneML.

" ... Textbook machine learning problems can be boiled down to solving a problem in mathematical optimization, of which there are many libraries and packages to choose from. In practice, applying an algorithm is hardly the only thing a data scientist needs to do. The reality is that data scientists need to optimize data pipelines (acquire, wrangle, featurize, fit a function) which means chaining together primitives and systems and optimizing them across interdependent steps. With recent successes in computer vision, speech, and machine translation, deep neural networks provide an approach for optimizing such pipelines via gradient descent. But there’s no reason to believe that other algorithms can’t become competitive with deep learning. For alternative strategies to emerge, frameworks and platforms for comparing and optimizing data pipelines need to get better.  ... " 

Need Better Metadata Services

Favorite topic stemming from recent consulting interaction.  Creative look at the implications. I always more broadly define metadata as also including contextual data.   In O'Reilly:   " ... We need open and vendor-neutral metadata services.  Comprehensive metadata collection and analysis can pave the way for many interesting applications. ... "   By Ben Lorica

Deep Learning for Decision Making and Control

Robot Deep Learning.
Deep Learning for Decision Making and Control
Published on Mar 18, 2015
A remarkable feature of human and animal intelligence is the ability to autonomously acquire new behaviors. This research is concerned with designing algorithms that aim to bring this ability to robots and simulated characters. Levine will describe a class of guided policy search algorithms that tackle this  ... " 

No UI is the New UI

A novel look at conversational, messenger style interfaces.   Implies that the US can well understand context and even intent.  What are the implications of error?

No UI is the New UI
On the rise of UI-less apps and why you should care about them as a designer.

If you don’t know about these apps, what make them special is that they don’t use a traditional UI as a mean of interaction. Instead, the entire app revolves around a single messaging screen. These are called ‘Invisible’ and ‘Conversational’ apps, and since my initial post, a slew of similar apps came to market. Even as of writing this, Facebook is releasing M, a personal assistant that’s integrated with Messenger to help you do about anything. ... " 

Face Watch: Thief Recognition

Already much in use in Vegas casinos.  Categorizing facial images in real time based on stored images.   Implications of false positives?  Experiments described in the BBC.

"  ... A scheme widely used by UK stores to identify criminals is testing facial recognition technology, the BBC has learned.

Until now, Facewatch has provided local groups of businesses with a way to share their CCTV images of shoplifters and other potential offenders.

It is now giving shops the ability to generate alerts if a face recognition system matches individuals in the shared pictures to customers in their stores, as the BBC PM programme's Chris Vallance reports. ... "  

Tuesday, December 29, 2015

Fedex Holiday Glitch

Over holiday,  Fedex apparently unable to meet increased demand.  Poor forecasting of demand, resource allocation?   Awaiting more information.  Might be an interesting case study for highly varying demand.

Free NYC Gigabit Wifi Hubs

Saw a proposal for this quite a number of years ago.  Finally NYC putting public Wifi into many old 'phone booths'.  In the Verge.    " ... The full network will install more than 7,500 public hubs throughout the city, each replacing a pre-existing phone booth. Once completed, the hubs will also include USB device charging ports, touchscreen web browsing, and two 55-inch advertising displays. The city estimates that ads served by the new hubs will generate more than $500 million in revenue over the next 12 years. ... " 

Look at Amazon Echo as Assistant

Have much enjoyed the Amazon Echo's capabilities over the last year.  As the article suggest, it works well in unusual, sound challenging environments.  That is a plus.  It does quite well as an always-on attentive assistant.  But is it the best sound activated assistant?  My own observation to date is that it makes more semantic mistakes than the Apple Siri.  And usually it has to deal with shorter commands that drive a specific narrow context,  like music selection or appliance interaction.  It appears to be more trigger-word sensitive, so its actions can collapse due to any mis interpretation. Many of the skills seem to be 'state' driven, so again, if the state is not correctly determined, it fails.  It too often then must start again from the beginning. Responding to:   in Android Central. 

R for Big Data: PivotalR

Just examining.    Addressing the usability of languages like R and Python in Big Data computing environments.

" ... When discussing data science tools, it’s common for folks to passionately debate about algorithm breadth, scalability, and performance among the many available options. Yet one of the most important aspects to consider when choosing a data science tool—usability—is often ignored in these discussions.

We believe that usability is perhaps the most important aspect to consider when selecting data science tools. In day-to-day settings, a data scientist should be focusing on what she wants to do with the data, rather than having to determine the technical aspects of how she is going to get there. Unfortunately, we’re at a stage where the how takes up a large chunk of a typical data scientist’s workflow.  ... " 

Winning at Work

Useful thoughts:  In the HBR:  Research-Based Ways to Win at Work in 2016  ....  by Sarah Green Carmichael

None of these points are that surprising.  Each suggestion has a link that leads to related research.   All from other articles in the HBR.  So this is a kind of meta description. Hard to easily tell what the depth of each study might be or its exact context.

Monday, December 28, 2015

Media Spending Slower Growth

From McKinsey:   A key measure.   Overall expenditures rose marginally year on year in 2014, but slower growth is coming.  

BMW Pushes Gesture Control

Always had my doubts about gestures control in cars, but BMW is pushing it:

In TheVerge:
Last CES, BMW teased the future of its human-machine interface with a gesture control demo — a capability that debuted later in 2015 on the new 7 Series. Once again, BMW will be showing a new form of system control at CES 2016 next week: AirTouch, they're calling it, which sounds like a much more advanced version of last year's gesture control system.

The system that's currently in the 7 Series is limited to just a small handful of basic controls — volume, for instance — but AirTouch "allows the display in a vehicle to be operated like a touchscreen without actually having to make contact with the surface" using an array of sensors in the interior. Basically, it sounds like you can use your outstretched hand like a mouse on the car's oversized display. Once you're focused on the on-screen item you want to select, the driver can use a button on the steering wheel to make the selection, and the passenger can do the same with a button on the door.  ... " 

Disease Diagnosis, Data Analysis

Instructive, technical piece on disease diagnosis using Bayesian analysis.  Includes R code for implementation.  Cautionary analysis of the results.  Particular attention paid to false alarm rate. Useful for analyzing many kinds of data.

Philip Tetlock and Super Forecasting

Recently covered in the BBC with an introductory article and broadcast.    About the gradual improvement of forecasting and the use of tournaments to drive them.  This topic has been covered here for some time.    Precision forecasting is fundamental for strategic and tactical organization planning. " ... Most intelligence agencies rely on the informed hunches of trained specialists. But there are people from ordinary walks of life who routinely do far better at predicting events than the experts. Tara Isabella Burton meets some of them. ... " 

Google Introduces Open Location Codes

Long time fascination with maps and location.  Google Earth and Google Maps,  linked to satelite imagery  amazed me in the way they brought interactive maps to anyone.  Further Google has recently added open location codes ....

Sunday, December 27, 2015

Why Book: On Causes

Reading excellent book that addresses a favorite, fundamental topic. Non Technical.

Why: A Guide to Finding and Using Causes
by Samantha Kleinberg

"Kleinberg expertly guides readers on a tour of the key concepts and methods for identifying causal relationships, with a clear and practical approach that makes Why unlike any other book on the subject. Accessible yet comprehensive, Why is essential reading for scientific novices, seasoned experts, and anyone else looking to learn more from data." 

Homekit Thermostat for a Smart Home

The humble home thermostat has always been a model for the sensing device.  Even for the future of pattern matching AI.   Have been examining examples like the Nest as a starting point for the smart home.   And Echo as a means of voice interaction.  Nest is still not very interconnected as a part of an internet of things, or doing enough analytics to leverage the data it gathers.  Now an example of a Homekit ready thermostat.   Lets see more.

Whats a Data Lake?

Nicely done overview.   In a world of cheaper storage, more data and metadata, better BI and deep analytics,  toward automated modeling methods .... makes much sense.  How about an automated way to scour the lake to find leverage for specific value plays?  Lower my raw material costs!

On Big Data Visualization

From InsideBigData:
In this special guest feature, Jans Aasman Ph.D., psychologist and expert in cognitive science as well as CEO of Franz Inc., discusses the use of semantic visual discovery to make big data meaningful. Dr. Aasman was an early innovator in Artificial Intelligence and leading supplier of Semantic Graph Database technology. .... 

Data visualization, according to Nathan Yau, a UCLA statistician, is “a representation of data that helps you see what you otherwise would have been blind to if you looked only at the naked source.”
Businesses are increasingly turning to visualization-based data discovery tools, with Gartner estimating a 30 percent compound annual growth rate through 2015. As organizations around the world collect more data than ever, and as Big Data gains popularity as a means of expanding entrepreneurial horizons, visuals can help us understand connections within the data with greater clarity. .... " 

And I add, while data viz is often presented as 'just' a better descriptive mechanism,  it is often a key part of discovery, and should always be used prior to analytics and machine learning exercises.  Start with the basics.

Digital Signage 2.0 Advances as Data Art

Good overview of the advance of digital signage.  We spent much time trying to understand its effective in store use.   Beyond retail grocery retail?  Perhaps.   " .... Their secret weapon is "data art," a new breed of highly creative data visualisation that uses business information to create real-time visual effects. So rather than resorting to a relatively dull amber ticker tape to communicate the latest stock market movements in real time, this data can now be presented in a far more engaging way through stunning visuals; for example — liquid gold to represent commodity fluctuations. ...  " 

Saturday, December 26, 2015

Localized Web Pages for Beacons and Search

In Retailwire: With obvious applications to beacon interactions and segmentation attached to locations.  Other integrations with analytics?    What is the best segmentation linking online and physical retail.    Or?

" ... For big chains, the web presence of individual stores is often limited to an address on a "Store Locator" page or a handful of store-specific Yelp reviews. But as retailers look for greater interaction between physical stores and their digital presences, Brickwork is trying to change that.

Already partnering with almost 25 retailers, including Urban Outfitters and J. Crew, the just launched startup creates store-specific websites for individual stores in big chains.

According to the New York Business Journal, Brickwork CEO and Founder David Muczinski pointed out in an interview that 90 percent of sales occur in physical stores, but 80 percent of customers start on the path to purchase by researching online.  ... " 

Science of Skylines

When suggestions were being made as to the change of the Cincinnati skyline I recall talking to an architect on the implications.  And how technology of the time could be used to interactively display the design alternatives.  An example of visualization of the macro effects of a design.  This article relates to that question and related research since then.    Implications beyond the aesthetic, to the smart city?

A Workshop for Network Theory

A recent meeting on network theory by the Santa Fe Institute.  We used these approaches for supply chain models, and for marketing decisions.   Looking for more that came out of this conference.  Some at the link.  " ... Networks are everywhere – from social interactions to species feeding relationships to the algorithms that pull information from large datasets. Because of its broad utility in quantifying interacting systems, network theory now finds application in many disciplines.  ... But network theory, like any emerging field, needs to keep up with the times. ... " 

Robot Bartenders

We are getting closer,  A step forward?   Despite the complexity, still a narrow dimension of human interaction and need.   " ... Sure, we've seen machines that can mix drinks with the help of an app and robot bartenders that can create cocktails out of Keurig-esque pods, but they all require you to actually push the control buttons and carry your own beverages. The horror! Fortunately for lazy drinkers everywhere, members of the CSAIL (Computer Science and Artificial Intelligence Laboratory) team at MIT recently presented a system of robots working together in a "bar" scenario that consisted of a PR2 robot "bartender" and two four-wheeled open-source "Turtlebot" kits acting as waiters. That... must have been a mighty tough day at work.  ... " 

How to be a Good Boss

In FastCompany:  Googler impression of what it takes to be a good boss.  " .... Kim Scott has built her career around a simple goal: Creating bullshit-free zones where people love their work and working together. She first tried it at her own software startup. Then, as a long-time director at Google, she studied how the company’s leaders created an environment where the joy that people took in their work felt almost tangible....  " 

Thursday, December 24, 2015

Google Planning Chatbots in Messengers Leveraging Cognitive AI

I recently mentioned the linking of chatbots and messenger systems.  Recall we addressed this in early assistance systems.   See the Mr Clean tag below.

In Forbes:  Google To Launch An Artificial Intelligence Messenger Service To Rival Facebook M   .... Theo Priestley

" .... According to sources close to the Wall Street Journal, Google is looking into launching a new mobile based messenger service underpinned by artificial intelligence. “For its new service, Google, a unit of Alphabet Inc., plans to integrate chatbots, software programs that answer questions inside a messaging app,” the article claims. ...  " 

Sharing Your Data on Your Terms

In FastCompany:

 MIT's Enigma Project would allow you to sell the rights to use your data in aggregate without having to grant access to your raw information.     By Steven Melendez  .... 

A new Massachusetts Institute of Technology project could soon bring much greater flexibility to cloud data sharing—potentially helping companies analyze consumer data without giving employees access to individual customers' personal information or letting loan applicants submit data for automated underwriting without ever sharing their information with a human being.

Consumers could even be able to sell access to their own data for research purposes without having to worry about it leaking across the Internet or showing up in unexpected hands. ... " 

What if Ford and Google Partner?

Most car makers are examining the space.   Implications are still unclear.  Would direct partnership with a car maker radically change the industry?

Who Wins If  Ford and Google Partner?    
Phil LeBeau, correspondent at CNBC

Industry insiders have long speculated that Silicon Valley's technology firms would eventually team up with Detroit's big automakers to develop self-driving vehicles. ... " 

If Everything had a Chip?

In the early days of working with the MIT lab on the 'Penny Tag' project, a thought experiment often posed was:  What if everything manufactured were tagged, or more radically,  intelligently chipped?   What could be done with that idea using modern analytics?   Marc Andreessen says it is coming.

Wednesday, December 23, 2015

Changes in Managing Risk

Noting the need to connect risk directly to relevant decision process.

How Managing Risk Has Changed    (Podcast and Transcript)
The problem with many catastrophic risks isn’t just that their impacts, when they hit, are so massive. It’s also that their odds of occurring in any given short time frame are very small, so that planning for them has to be handled as a long-term priority while the proverbial sun is shining. And neither companies nor individuals are particularly apt at taking serious, long-term action to prepare for low probability, high consequence events.

Enter the Wharton Risk Management and Decision Processes Center, which was created 30 years ago to help individuals, businesses, governments and global organizations to be better prepared for those longer range, more unpredictable dangers.

Knowledge@Wharton spoke with Howard Kunreuther and Robert Meyer, co-directors of the Wharton Risk Management and Decision Processes Center, and executive director Erwann Michel-Kerjan about the center’s research and how managing risk has changed over the past few decades. ... " 

Chatbots are Back for Services

Google working on a chatbot filled messaging service.  We experimented with chatbots.  And I see that the 'Slack' messaging service claims one as well.   It is a natural place to place a service that interprets interactions and attempts to provide value.   NLP based AI has improved considerably, so it is useful to think of their inclusion in service channels.  An example of an attentive system.  Also an example of a 'UI Free' system.    More to follow here.   Also consider their similarity to Facebook M,  where messages can be passed to humans, or chatbot agents.

InHalio: Towards an Internet of Scent

From a long time co-participant in retail.   Our innovation center tested a number of sensory interaction devices in retail environments using smell, touch, vision and sound.  The data gathered during such interactions is also valuable.

Entertain your senses.  A digital Gateway to Fragrances.  An Internet of Scent.

5th Screen Digital enables a bold new world of smell based on its Inhalio platform. Years of development led to this disruptive and scalable solution to add scent to the digital experience. Inhalio brings a personalized scent experience to the connected world…the Internet of Scent.

Inhalio Retail enables a retail experience unlike any other. Consumers are greeted with gorgeous animations, and they touch the screen to select a fragrance. The dry fragrance is dispersed for them to smell. Consumers can determine their individual preferences in a personal experience. Inhalio platform creates a new engagement for digital signage.

While engaging and entertaining the consumer in a remarkable new manner, the Inhalio platform monitors the system, tracks usage and provides invaluable feedback and data to the retailer and brands. ... The Inhalio platform can create a mood scent that just like music or lighting allows the dispersion of scent to help create a mood.

The sense of smell is the most powerful memorable sense that we have. The olfactory cortex is embedded within the brain’s system where emotions are born and emotional memories stored. That’s why scents, feelings and memories become so easily and intimately entangled.

We look forward to how you will connect the Internet of Scent with the Internet of Things!   ... " 


Background on Digital Scent Technology

Modern Innovation Leadership.

Modern Innovation Leadership
September 2, 2015 in Innovation Philosophy

Lord Kelvin’s Ampere Balance

There are two main typical approaches to leadership. Some people think that a leader is a person with strong convictions and an own vision that can convince people to accept his own ideas. Other ones think that a leader is a person who understands the needs of the group and has the capability to organize and coordinate it in order that the group can reach its own objectives. The main differences are related to the how the vision about the group is created, the former one proceed from a single person, and the latter one is gathered from all the people in the group ..... " 

Tuesday, December 22, 2015

Microsoft Acquires Metanautix for Data Supply Chains

In CIO Today: Addressing the data supply chain.  Not necessarily about company product supply chains, but about getting and delivering the right data to operations and analytics methodologies,  Good concept, its often a major challenge.

".... big push to beef up its enterprise analytics offerings before the end of the year with yesterday's acquisition of startup Metanautix. The three-year-old company specializes in integrating data sets throughout the supply chain, with a particular focus on servicing large enterprise clients with heterogeneous data types.

“Companies continue to generate enormous volumes of information and aspire to be more data-driven in their strategies and operations,” Joseph Sirosh, corporate vice president for Microsoft’s Data Group, said in a blog post. ... " 

Interview with a GE Predictive Analytics Practitioner

A new esteemed colleague of mine, Matthew Pietrzykowski,  who does predictive analytics at GE was recently interviewed by Predictive Analytics World.  This does a nice job of answering the question:  What do you do?   Below is the intro and the first question.  See also the PA World Blog.

 " .... In anticipation of his upcoming conference co- presentation, Advanced Analytics and the Corporate Audit Function at Predictive Analytics World San Francisco, April 3-7, 2016, we asked Matthew Pietrzykowski, Senior Data Scientist at General Electric, a few questions about his work in predictive analytics. .... 

Q: In your work with predictive analytics, what behavior or outcome do your models predict?

A: The predictive models that tend to be generated in GE’s Corporate Audit Function are heavily focused on classification outcomes, forecasting and optimization. The types of models used range from logistic regression to random forest classification models. Typically, the models are built to help auditors assess whether there is evidence to support an auditable event or find the optimal or reasonable outcome. These models tend to be of mixed data types and some are augmented with the results of text mining short form narrative fields. .... "       Read the complete article ...

GePhi Updates to version 0.9

I see that the Gephi Graph Analytics system has been updated to version 0.9  ... Creeping ever so slowly up to version 1.0   Used it for several structural analyses of supply chain designs. The popular, open source, free package is worth exploring.   Worth updating as older versions had issues with Java.  " .... Gephi is an interactive visualization and exploration platform for all kinds of networks and complex systems, dynamic and hierarchical graphs. ... Runs on Windows, Mac OS X and Linux. Gephi is open-source and free. ... " 

Machine Learning Marketing Trends

Christopher S. Penn talks machine learning and marketing.  The technical future of marketing?

" ... In this multi-part series, we’ll look at upcoming trends in marketing in 2016 you should be prepared to address. Today, we’re looking at Machine Learning.

Machine learning has been a goal of computer scientists and engineers since the 1950s, but only in the last 5 years has it become accessible to more people. Prior to the democratization of machine learning, advanced technologies were limited to the biggest computer science companies and laboratories, only for corporations with the biggest budgets. ... " 

Monday, December 21, 2015

Robotic Vacuum has Security Functions

In Engadget:  Was waiting for some of these kinds of obvious multitasking capabilities.    Only way we will see their more general home use.

First Web Site is 25 Years Old

In Endgadget:  I remember first seeing the concept being used by folks in research.  I had used the Internet in the form of the Darpanet years before.  The idea of having hyperlinks in documents had existed for some time, but making them universally bridge documents and remote resources was new and exciting.   It threaded and stitched together knowledge in new ways.

Personalizing Maps

In the CACM:

For individuals, or for commercial entities. This method is often used when models are run and the results are overlaid on a map.  Interaction with a map is also a means of personalization for a context.

" ... Geographic maps constitute a ubiquitous medium through which we understand, construct, and navigate our natural and built surroundings. At the intersection of the explosion of geographic information online, data-mining techniques, and the increasing popularity of Web maps, a novel possibility has emerged: Instead of generating one map for large numbers of users, user profiling and implicit feedback analysis can support creation of a different map for each person. The automated personalization of the map-making process is still in its infancy but has the potential to provide more relevant maps to millions of users worldwide. ... " 

Loyalty Rewards Discounts Work better Online

More studies on loyalty in retail.  Also good comments.  Read the article details.

" ... With consumers typically not needing the incentive of a coupon to return to their favorite physical store, discounts from loyalty reward programs tend to be more effective in encouraging repeat customers online, according to a new university study.

Internet shoppers are "more variable," intently looking for bargains and less loyal to particular websites, writes Sanghee Lim, an assistant professor at Johns Hopkins Carey Business School in her study, "Loyalty programs and dynamic consumer preference in online markets," published in Decision Support Systems. ... " 

The Internet of Things and AI

In the CACM:  A look at how AI should connect with the Internet of things.   " .... Oblong Industries CEO John Underkoffler does not believe a connection between the Internet of Things (IoT) and artificial intelligence (AI) will become a threat to mankind, provided basic questions are answered prior to considering imposing constraints on AI capabilities. ... " 

Sunday, December 20, 2015

Wal-Mart Trip Type Classification

From DSG:  Nice example of how data science can be used for a real world classification problem, with real data.  Including 1.2 million observations with 6 features.    Here using a Random Forest method for classification of retail trip types and comparing it to other methods.    Includes all the Python code from a Kaggle competition. Visualization of modeling progress.   Nicely done example. via Kirk Borne.

China Customer Data Opens Potential

Remember looking for just this kind of detail data just a few years ago to model the Chinese Consumer in more detail. Quite a potential here.  This is big enough and detailed enough to drive some real value.

Chinese Internet Giant Tencent Opens Its Trove of Data to Agencies
Dentsu Aegis Network and GroupM Are the First Partners ... 

Chinese internet giant Tencent is massive in scope and in the variety of ways it reaches users. Its mobile social platform WeChat has about 650 million monthly-active user accounts; another platform, QQ, has 860 million. It's big in games, music, video and news. And it has the data that goes with all that. ...  " 

Smart Talking Toys as Virtual Assistants

A kind of extension to the virtual assistant idea.   Toys that can engage, inform, aid, teach.  It is already here with advisors like Cortana, Alexa and Siri on a phone or tablet.  But is a toy a different kind of channel that requires more care?  Yes, I believe.   Will also get kids used to the idea of using virtual assistants.

In the WSJ:  (Registration, etc)
Talking Toys Are Getting Smarter: Should We Be Worried?
Internet-connected talking dolls like Hello Barbie that can actually converse are bewitching kids but unsettling parents. Are they really the menace critics have made them out to be? ... " 

One of the toy examples, a dinosaur called Dino, from Cognitoys, is said to be based on some Watson technology, and claims to provide a 'true companion'.   The article talks to MIT device psychology commentator Sherry Turkle about how such toys can mislead children.  The article discusses how the conversational data provided by children could be mis used.

Consider also the strong engagement factor formerly seen by the Tamagotchi toys.  Or from attempts at leveraging apparent intelligence in the toy.  With little help from biomorphic forms .   Previously talked about here.

Big Data as a Trillion Dollar Asset

A favorite topic.  It is not only about the data, but how the data can be combined with other data in given contexts.   It need not be 'big' in any sense, but valuable in context.  A year ago spent some time looking at methods for predicting the value of data.  See Data Assets tag below.   Any pointers to others doing work in this area?

Glucose Tracking Device

An example of new health sensor development by Google and Dexcom.  Disposable device to track glucose levels. 

Saturday, December 19, 2015

Stanford Winter Crowd Research Program

Via ACM.  Register until the end of December.

Stanford Crowd Research Winter Program
Are you a designer, hacker, engineer, aspiring researcher? - come join the Stanford Crowd Research initiative - a first of its kind massive open online research initiative - and do research with Stanford faculty and researchers - get recommendation letters, publish papers, build a unique experience. Everybody welcome to apply. Questions? Comment below or email me at rvaish@cs.stanford.edu. ... " 

Half of Psych Studies Cannot be Reproduced

So what does reproduce ability mean?    Look first at the kinds of bias involved.

In the Verge: " A surprising number of psychology studies can’t be reproduced ... More than half of 100 tested studies produced different results .... " 

Future of Virtual Customer Experience

According to research by Gartner, 89 percent of executives are betting on customer experience as their primary mode of competition before the end of 2016.

However, over the last few decades, customer experience and service delivery veered away from personalized support and adopted a “call and wait” system while companies cut costs and installed voice-activated phone prompts. Today’s trends are all about personalization with a twist.

We want our personalized customer service to encompass a wide variety of platforms from social media to mobile while still being easy to use with lightning-fast responses. ... " 

Cognitive Legal Aide

Another example of cognitive assistance:

" ... Boaz Carmelli gave an excellent presentation today on "Argumentor" as part of our Cognitive Systems Institute Speaker Series. Please continue the conversation here with Boaz, who's published works are posted here     Talk slides   Further discussion,  ... "

Apple Pay Goes to China

In CWorld:  " ... Apple partners with UnionPay to bring Apple Pay to China   ....  "China is an extremely important market for Apple and with China UnionPay and support from 15 of China's leading banks, users will soon have a convenient, private and secure payment experience," Eddy Cue, Apple's top executive for the Cupertino, Calif. company's mobile point-of-sale payment service, said in a Friday statement.  ....  " 

Friday, December 18, 2015

An Architecture of Data Science

Nicely done piece by Vincent Granville, about the 'architectural' considerations of data science.  In roughly the order you should progress, but understanding you may have to revisit many elements more than once.   Mostly nontechnical.  This is useful for the entire spectrum of project participants, from scientists to decision makers.  A template for understanding?

" ... In this article, I summarize the components of any data science / machine learning / statistical project, as well as the cross-dependencies between these components. This will give you a general idea of what a data science or other analytic project is about. .... " 

In comparison, related architecture, see also CRISP-DM.

Innovation Management

Thoughts on the management of innovation.  Not enough about the promotion of creativity.   Measurement is emphasized, which is good.

Customers are Talking, Companies are Listening

This as news surprises me.  Our company always listened.  In fact pioneered methods for listening and analyzing what we heard.   Hardest part is then, linking that knowledge to some action.  Complete the loop.   From Bain.

The Scale of Telco Data Services

I find this obvious, but the numbers are interesting.   " ... The $24 Billion Data Business That Telcos Don't Want to Talk About  ... Mobile Carriers Are Working With Partners to Manage, Package and Sell Data ... " 

More on Google's D-Wave Quantum Machine

Lots of implications for doing things like AI and machine learning faster.   But also breaking encryption codes faster?   Good, extensive,  non technical piece on the topic.  Have written much on this, see tags below.   How long before this is commonly applied?.    In Technology Review.  " .... As recently as last week the prospect of a quantum computer doing anything useful within a few years seemed remote. Then NASA summoned journalists to its Ames Research Center in Mountain View. ... "  

Thursday, December 17, 2015

Watson for Social Media

Another example of focused Watson application ...

Get accurate social media insights with Watson Analytics for Social Media
Watch this video for a deeper dive into how Watson Analytics for Social Media can Identify the pulse of an audience with guided analytics that offer suggestions and abstract snippets from your social media data.  ... " 

Assistants in a Watch

The IBM/Apple connection first was positioned as one to enhance the enterprise.  Now a look at the consumer side.

IBM and Apple can put Watson's A.I. insights inside Apple Watch
I spoke with IBM’s Vice President for the IBM/Apple Partnership, Kathryn White, to get a little more background on news Apple and IBM are extending their partnership to put business intelligence across the digital ecosystem, from the artificial smarts of Watson to augmented cleverness strapped to any Apple Watch user’s wrist. The partners are in this for the long game. ... " 

Combining Multiple Scoring Systems

Cognitive Systems Institute Group Speaker Series call this week, on Thursday, December 17, 2015 at 10:30 am ET (7:30 am PT).  Our presenter on Thursday is Professor Frank Hsu, Fordham University, who will be presenting:  "To Fuse or Not to Fuse: Cognitive Diversity for Combining Multiple Scoring Systems."    Slides here.

  A link to slides and a recording of each call will be available on the CSIG website:
http://cognitive-science.info/community/weekly-update/  .... 

We encourage those who join the calls to add questions and comments to the https://www.linkedin.com/groups/Cognitive-Systems-Institute-6729452  on LinkedIn and we ask that you ask questions at the end of the call. ... " 

Automation Driving Retail Growth

Statistics on retail growth.  Advanced automation replacing in store and back end knowledge workers too.   Relates to think tank examination we did this week.  Voice commands.

" .... E-commerce platform provider Demandware has released its annual list of retail predictions for 2016, and many of them center on how machines will continue taking over tasks once done by people.

Out of the nine trends Demandware identifies, four are related to growing IT automation: machine learning driving voice-based intelligence assistants, in-store payment evolving dramatically, data-driven decision marketing replacing gut instinct, and last-mile fulfillment providers driving instant gratification.

Specifically, Demandware predicts that machine learning will expand beyond shopper personalization to provide retailers with a Siri-like intelligence assistant that pops up occasionally with insights and recommendations. Swiping, typing and tapping be largely replaced by voice commands. This will help make personalization such a fundamental part of the shopping experience that consumers will no longer notice it. ... " 

New Learning about Learning

Contrarian view that is of interest.  We do know much less than we claim we know about the brain. We do know it is complex.  We often try to adapt our knowledge of it to what we know about our own development of things that perform similar tasks.  How does it learn?  How does it store information? How can we leverage its model?

Celebrating Failures

From Knowledge@Wharton:    After action reviews?

The digital transformation of a company requires not a mere shuffling of the organizational chart, but rather a “chemical” change in the culture and business practices, says Ganesh Ayyar, CEO of Mphasis, a major IT services company. But it is easy to say and more difficult to do. One place to start is by encouraging experimentation through the celebration of failures, adds Wharton marketing professor Jerry (Yoram) Wind. Another is to learn to co-create with clients. As always, the CEO and other senior executives set the tone: The old command-and-control style of managing is becoming passe, replaced with a more collaborative model recognizing that good ideas can come from anywhere in the company. ... " 

Wednesday, December 16, 2015

Brain Inspired Computing

Technical.  From IBM Almaden Research center.   Dharmendra S Modha's Brain-inspired Computing Blog  ....  SyNAPSE: Systems of Neuromorphic Adaptive Plastic Scalable Electronics .... "  

A VC Perspective for AI

Everyone has been now been alerted to the re-emergence of AI.  But everyone still wants to know: What are the needs in our or our clients context that can now be effectively solved by AI solutions, and what resources will be needed to solve them?  I have been asked this question by a number of enterprises.  It's a big question. Where are we today to solve X?  What's off the shelf, what's open source?   It is still taking too much effort to get a complete answer.  Nathan Benaich of Playfair Capital provides a article below that is a good start, but we need more.

Watson IOT Research Center formed by IBM

A good direction to go with applications of Watson.   Considerable opportunity here.

IBM Bets on Watson With Global Research Center in Germany
The Internet of Things division will play a key role in driving revenue and profit within five years, Harriet Green, IBM’s vice president and general manager for Internet of Things and Education, told Bloomberg Television’s Ryan Chilcote in an interview. The company on Tuesday announced the opening of a new global headquarters and research lab in Munich for a division that will build Watson-based applications for Web-connected devices. The facility and eight other global centers are part of a $3 billion investment in the unit set out in March by Armonk, New York-based International Business Machines Corp.  ... “How this fits into the future of IBM is around profitable growth,” Green said. “We think it has real high growth, profitable growth potential.” .... ' 

Speed in Mobile Experience

In ThinkwithGoogle:  Google is continuing to provide some interesting snippets of data from their research about how people shop. Here another about how quickly people make decisions on mobile.   These could feed some useful predictive analytical models.   Worth following.

Buyer Decisions Not Journeys

I liked Journeys better, but ...

In CustomerThink:
" ... When marketing and sales leaders often think about how buyers make decisions, they are viewed through a prism of buyers making rational and process-driven decisions.  Leading to many strategies and tactics devoted to attempts to market or sell to the rationale behind decisions and buying processes.  Of late, the term “buyer’s journey” has come into vogue.  Essentially another name for mapping the buying process.  It is yet another example, however, of the search to understand the process of decision-making.

There is a significant problem with this prevalent perspective that exists in the world of marketing and sales: humans, thus buyers, do not make decisions in near-perfect rational, systemic, and predictable ways. ... ? 

Tuesday, December 15, 2015

Wolfram Programming Lab

I see that Wolfram has started this Beta.   Famously they created AI advisory system Wolfram Alpha, now part of Siri.  Start for free ....  Nicely done.

" ... .Knowledge-Based Programming .... Designed for the new generation of programmers, the Wolfram Language has a vast depth of built-in algorithms and knowledge, all automatically accessible through its elegant unified symbolic language. Scalable for programs from tiny to huge, with immediate deployment locally and in the cloud, the Wolfram Language builds on clear principles—and 25+ years of development—to create what is emerging as the world's most productive programming language. ....   " 

Evolution of Storytelling

 Beyond blogging,  hypertext links, technology.  How has this changed storytelling?  From my viewpoint it has made it easier to digress, but not much different.   When first introduced to hypertext, before the Web,  I remember thinking that all text would soon be heavily linked.    Experimented with knowledge management systems that could do that.  But the serial track of a story is still largely with us.   Making it effectively, usefully, responsively interactive is still hard.

 In Medium: Technology and Storytelling.

Storybooks and Data

Linking expert storybooks to data. Storytelling. It comes to mind that one way you could do an advisory system is to match to the 'right' storybook. Some kind of clustering or best fit. It then becomes what we called a 'case based system'. If you have enough of them. Here is more on what IBM announced at a talk.

Predicting Others Preferences

In HBS.    A classic element of market research.  Consider the analytic implications.  This is  a  Meta study:  Abstract: " ... Consumers readily indicate liking options that appear dissimilar—for example, enjoying both rustic lake vacations and chic city vacations, or liking both scholarly documentary films and action-packed thrillers. However, when predicting other consumers' tastes for the same items, people believe that a preference for one precludes enjoyment of the dissimilar other. Five studies show that people sensibly expect others to like similar products, but erroneously expect others to dislike dissimilar ones (Studies 1 and 2). While people readily select dissimilar items for themselves (particularly if the dissimilar item is of higher quality than a similar one), they fail to predict this choice for others (Studies 3 and 4)—even when monetary rewards are at stake (Study 3). The tendency to infer dislike from dissimilarity is driven by a belief that others have a narrow and homogeneous range of preferences (Study 5).   ... " 

How Microsoft Supports a Future with Countless Devices

In the NYT.      " .... Technology now allows for lots of powerful, compelling computing devices in a wide range of sizes and shapes. Just about everyone in the industry believes the smartphone will remain the dominant computing device for the foreseeable future. But second place looks entirely up for grabs, and the giants are jockeying to make that one other device that we’ll all use along with our phones. .. " 

Monday, December 14, 2015

An Algebra of Data

Had not heard of this,  or yet understand it.  But most data you mine in realistic domains does not have exact values, so isn't this more about data fuzzy logic?  Skeptical about what is claimed here.  " ... The Algebra of Data Promises a Better Math for Analytics, And More ... A company by the name of Algebraix Data is beginning to speak publicly for the first time about the algebra of data, an approach to storing and accessing data that it devised and patented. The company is using its data algebra now to provide big data analytics services to clients, but the uses could be much more far-reaching, including a universal language. ... " 

Smart Contracts and Block Chains

Continue to follow the concept and application of smart contracts.  In O'Reilly, here a simple but very real case: 

Deutsche Bank joins the blockchain gang
Deutsche Bank is the latest financial institution to test the potential of blockchain technology; its recent experiment used smart contracts to issue and redeem bonds. While the German bank has no plans to commercialize the results, this is the most recent in a series of successful proofs of concept, building confidence in blockchain's potential. ... " 

Passive Consumer Data Collection

Why are retailers turning to passive data for consumer insights? 

Custom surveys are considered the most important data source for the creation of insights today, but fast forward two years and they will be rivaled by passively collected consumer data, according to GfK's "Future of Insights" report. Do you see passively collected data rivaling consumer surveys within two years?  ... " 

Abundance Theory Applied to Analytics

Abundance Theory: Had never heard it called this, and was always as a general mindset rather than a working strategy.   Analytics usually works with constraints, can algorithms take us beyond this? Can we 'feature engineer' ourselves beyond constraints?

" ..... Scarcity theory, a term coined by Stephen Covey, suggests that everything in life has its limit. Whether that thing is a spot on the team roster, a scholarship, a job, customers, funding, promotions or something else, we need to hoard as much as possible for ourselves because there is simply not enough to go around. This same theory also says that there are limited ways to achieve success, and that anyone who wishes to make it must follow the same path and prescription that others have done previously.

In contrast, this coach, through her word and deed, demonstrated to me a living illustration of what Covey labeled abundance theory, or AT. Abundance theory is a mindset that looks at each glass as half full (at least) and sees the world as offering endless opportunity. ... " 

Cheaper Thermal Imaging Technology

How can the emergence of cheaper sensor technology segments change our world?   BBC looks at an example.  I had previously looked at uses of thermal cameras.

" ... But in the same way that GPS location tech has now found its way into cars, smartphones, cameras and many other devices, thermography, as it's more properly known, is on the brink of becoming a universal technology, too. The cost of chips and thermal detectors that enable us to see and measure infrared heat signatures from surfaces has plunged in recent years.  So in the future, that means more sensors in more places. Doing what exactly? ... " 

Sunday, December 13, 2015

A Bayesian Learning Framework

    I particularly like the premise here ... that the approach starts with causal decision descriptions of the real world.  With the potential of building from real world process.   Often use 'influence diagrams' to do this.  Or existing process models.  Sounds like there could be a means to iterate between real world and model.  And converge?

In Technology Review:
" ... The researchers used a technique they call the Bayesian program learning framework, or BPL. Essentially, the software generates a unique program for every character using strokes of an imaginary pen. A probabilistic programming technique is then used to match a program to a particular character, or to generate a new program for an unfamiliar one. The software is not mimicking the way children acquire the ability to read and write but, rather, the way adults, who already know how, learn to recognize and re-create new characters.

“The key thing about probabilistic programming—and rather different from the way most of the deep-learning stuff is working—is that it starts with a program that describes the causal processes in the world,” says Tenenbaum. “What we’re trying to learn is not a signature of features, or a pattern of features. We’re trying to learn a program that generates those characters.”  .... ' 

Processes of Memory and Learning. Machine Learning?

Several Blogs have recently picked up on the processes of learning.  Just this past week there have been a number of advances revealed on linking analytic and biomimicry processes.   Notably the use of Bayesian methods.  See this recent example.   In the Guardian a look at some classic studies of learning.

It should be noted that 'Machine Learning' (ML), as currently defined, is more about statistically based pattern recognition, than the kind of learning mentioned here.   A machine learning method can be used to determine patterns in data, and then those patterns could be stored away for later access and reference.  But that kind of full ML process is not usual today.   It should be more often considered.

'Deep Learning', is another term much in the news, but it is usually applied to more narrowly focused problems, like image recognition.   But it does use biologically inspired math constructs like neural networks.

Addressing Digital Gaps with CIO Insight

Pearl Zhu on the idea of digital gaps.  Have seen every one of these gaps.  CIO's should be about insight, process and delivery rather than just 'information'.  Please get us to the knowledge and wisdom level using the abilities of multiple people in the enterprise.

Many R Visualization Examples

In the past have always used advanced, interactive data visualization packages.   But open source analytics packages like R have considerable data viz abilities.  Here just an example of creating multiple plots on a singe panel from Excel data.  As I explore more,  amazed at the depth of resources on line. And it all seems to be self organizing for better value.

Saturday, December 12, 2015

Quantum Computing Aimed at Faster, Better AI

Note the emphasis on Quantum Computing directed at AI.  A new direction, it was not mentioned when we explored applications, which mostly addressed combinatoric optimization.   In the past, better algorithms (Software) were always emphasized over faster computing hardware.  Is this a major change in this thinking?  In the WSJ:

Why Google’s new quantum computer could launch an artificial intelligence arms race  By Dominic Basulto ...

The D-Wave 2X quantum computer promises exponential gains in computing power. (Stephen Lam/Reuters)  .... 

Google claims the D-Wave 2X is 100 million times faster than any of today’s machines. As a result, this quantum computer could theoretically complete calculations within seconds to a problem that might take a digital computer 10,000 years to calculate. That’s particularly important, given the difficult tasks that today’s computers are called upon to complete and the staggering amount of data they are called upon to process.

On the surface, the D-Wave 2X represents not just a quantum leap for computing, but also for the field of artificial intelligence. In fact, Google refers to its work being carried out at NASA’s Ames Research Center as “quantum artificial intelligence.” That’s because machine learning problems that today are too hard or too complex for computers could be solved almost instantaneously in the future. ... " 

Cities Become Digital Laboratories.

In the WSJ:  The City becomes a Digital Lab.   (May require registration)

New York: —Gregory Dobler is an astrophysicist who honed his craft by recording spectral images of quasars and black holes. Now, from a high-rise rooftop in Brooklyn, he is training his lens on the expanding universe of New York City.

Every 10 seconds for two years, Dr. Dobler and his colleagues at New York University’s urban observatory have taken a panorama of Manhattan. Across hundreds of wavelengths of light, they are recording the rhythmic pulse of a living city, just as astronomers capture the activity of a variable star. ... " 

First Droneport under Construction

Regulation is still  evolving for the concept, but given that the concept requires location and technology, it makes sense.  In FastCompany:  World's first Droneport under construction near Las Vegas.   The Aerodrome will provide education and certification for drone pilots, as well as facilities for drone racing and other services.  ... " 

Open AI Emergent

Why not?  Anything to get the ideas closer to many specific business purposes.   Following.   How can companies ally themselves for this kind of open innovation?   Their OpenAI introductory blog post.   See OpenAI.  Will be approaching.

Elon Musk + Sam Altman launch OpenAI Nonprofit that will use AI to 'benefit Humanity'.
Let by all-star team of Silicon Valley's best and Brightest, OpenAI already has $1 Billion in funding.   By Daniel Terdiman

Silicon Valley is in the midst of an artificial intelligence war, as giants like Facebook and Google attempt to outdo each other by deploying machine learning and AI to automate services. But a brand-new organization called OpenAI—helmed by Elon Musk and a posse of prominent techies—aims to use AI to "benefit humanity," without worrying about profit.  Musk, the CEO of SpaceX and Tesla, took to Twitter to announce OpenAI on Friday afternoon. ... " 

Friday, December 11, 2015

Watson Anaytics doing Social Media Analysis

" ... I was an Alpha and Beta user of Watson Analytics.  Continue to follow their application for specific industry areas.  Somewhat narrow, but impressive so far in ease of use and visual results presentation.  Another step forward in the automation of analytics.  About to look at the latest, see the link for more:

" ... by Holly Nielsen in Linkedin.
Social Strategist, Watson Analytics ... 

5 reasons you need deep social media analytics for better business outcomes
Introducing Watson Analytics for Social Media. Discover what matters to your audience by analyzing social conversations and applying social media analytics for true customer understanding. ... " 

Samsung and Automotive Systems

In CWorld: Samsung moves their technology into the auto industry.  This piece further looks at how other companies are in the automobile systems and sensors market, with links to further information. Implications for this particular kind of IOT will be considerable.

Shopper Scientist: Herb Sorensen

I want to remind people of Herb Sorensen's blog: Shopper Scientist.  He visited our innovation centers a number of times.   One of the most data oriented retail explorers I have ever met.  I hear he may be coming out with an update of his book,  anticipating it, follow him for more information, I will report on it when it comes out.  See my previous writings on his book: Inside the Mind of the Shopper.

3D Touch as a Scale

Always looking for new ways to convert mobile devices into sensors.   Here is a new one, perhaps obvious.  Using the 3D touch capabilities as a scale on the 6s phone.  Sometimes as part of a game. It seems that Apple is blocking Apps that do this.

IKEA's Future Living Innovation Lab

In their blog:  Future food and more.   Covered in the Verge.

" ... Space10 is a future-living lab and exhibition space in the heart of Copenhagen. Our mission is to investigate the future of urban living by detecting major challenges that will impact people on a global scale, and exploring possible solutions. The overall goal is to create opportunities for a better and more sustainable way of living in the future. ... " 

Open Source Software

The Open Source Era
An Oxford Economics Study Commissioned by Wipro

Open source is mainstream today.  Why are businesses shifting to open source software? What are the main challenges and how are they overcoming them? How can open source software benefit you?
Know more in the Oxford Economics study in collaboration with Wipro where a 100 global executives have shared their views on open source software adoption and development methods. ... " 

Thursday, December 10, 2015

MIT Reality Editor

MIT's Amazing New App Lets You Program Any Object.  The Minority Report Style App that makes programming your Smart home as easy as connecting the dots.  ...   (Video)  .... " 

The Reality Editor (and its free associated developer platform, Open Hybrid) aims to give users this power to fully control the smart objects in their lives. And if one object doesn't have the functionality they want? They can just link it to another one that does: a sort of meatspace IFTTT. Imagine having a lamp that turns down the volume on your TV when you dim the lights, a light switch that also turns off your television, or a car that switches on your home A/C in the summer when you leave the office. The Reality Editor makes all of this possible. ... " 

Download it now.

Wal-Mart Launches Walmart Pay

Payment systems can link together payment, behavioral segmentation, loyalty and mobile engagement channels.    Opportunity to link in new kinds of analytics is considerable.

In Retailing Today:
Shoppers at Walmart stores will soon be able to pay with their smartphones as the retailer has developed a unique solution to become the newest entrant in the crowded world of mobile wallets.

Branded as Walmart Pay, the service is a uniquely Walmart solution to the rapidly evolving world of mobile payments that is likely to increase downloads and utilization of the retailer’s mobile app. By adding payment functionality to its app, Walmart said it is the only retailer to offer its own payment solution that works with any iOS or Android device at any checkout lane, and with any major credit, debit, pre-paid or Walmart gift card. ... " 


Nautil on Caricatures.  I remember seeing the MIT Caricature Generator at work during a visit.    I did not think it was very good, compared to human work.  But I liked its premise.  Take a person's image, and figure out how much it was different in certain measurements, from the average human face.  Then exaggerate those dimensions.  An engineers 'definition' of caricature.  A design artist, though, did much better.  Made me think more about the creativity in design, and moved me a bit from my engineers view of expertise.

Making Predictive Analytics Work

Predictive analytics requires a customer-obsessed innovation culture
Posted by Fred Giron in Forrester.   Some very good points.   It starts and ends with business process and the ability to rapidily try multiple methods of solution.

Embedding Machine Learning in Data Visualization

Via Spotfire: From a survey of work to integrate data visualization packages with machine learning. With the intent to provide easily understood, extensible and standardized libraries of solutions against existing data.

Creative Leaders

Tom Kelley, co-author of Creative Confidence (Crown Business, 2013) and a partner at IDEO, a global design and innovation firm,  on what creative leaders do.   We worked with IDEO in a number of ways, see the tag below.    Well done short piece in the HBR.

Kroger and Predictive Analytics

Kroger and predictive merchandising.  Kroger's Improved Analytic Capabilities Powers Merchandising Decisions  ....

" ... One key area where the in-house analytics team is being leveraged is in analyzing consumer behavior and predicting demand. As the healthy and organic craze moves from the buzz and niche phase to an established long-term trend, mega grocers like Kroger are quickly joining the likes of Whole Foods and Trader Joes and offering healthier food options to meet rising demand.  

"We certainly do a lot of predictive analytics with data we see not just inside our stores but overall in the economy, trying to figure out where the consumer is going," Schlotman said. "It clearly does seem as though there's going to be persistence and shift to a more healthy lifestyle and maybe out of some of the traditional center-of-the-store categories into more fresh categories."

In typical Kroger style they are not just following the trend but setting the pace. The grocer not only sells natural and organic products but has its own store brand Simple Truth dedicated to providing healthy food options.... " 

Wednesday, December 09, 2015

Wearable Technology for Services

Some innovative wearable technology examples for customer service.    " .... The capabilities of wearables are just beginning to emerge for retail, enterprise, and consumers. While their sales are not expected to compete with those for smartphones and tablets for some time, their impact on technology is far greater. As Multichannel Merchant points out, wearables “will become the interface between body, apps, data, and last not but not least, services.  ....  "

Google says D-Wave Works

Recall the controversial claims of D-Wave in doing quantum computing for NASA and Google. Also, that we had a minor interaction with them regarding test problems.  Now is the tide turning in their direction?

' ... “We found that for problem instances involving nearly 1000 binary variables, quantum annealing significantly outperforms its classical counterpart, simulated annealing. It is more than 10^8 times faster than simulated annealing running on a single core. We also compared the quantum hardware to another algorithm called Quantum Monte Carlo. This is a method designed to emulate the behavior of quantum systems, but it runs on conventional processors. While the scaling with size between these two methods is comparable, they are again separated by a large factor sometimes as high as 10^8.”

So Google is claiming that D-Wave completed the task 100 million times faster than its non-quantum cousin.  .. " 

Dealing with Unstructured Data

A look at the value of unstructured data.  Worth a read.  Text is structured because cognitive beings created it.  But has structure that is driven by a number of influencing factors, that should have useful value.  Good read.  Good example.  Reminds me of our own experience in Content Analytics.

In CustomerThink: 
" ... What I want to focus on here is the nature of dealing with unstructured data. That’s where text analytics comes in. Text automation and analysis provides a way of harnessing unstructured information and incorporating it into insights and planning. I’ve been working with text for the last 12 years and have watched the field change a tremendous amount. In the earliest commercial offerings, text software was very time-consuming. It was heavily geared toward predictive modeling and typically used methods based on statistical and machine learning. Capabilities were far less robust for the most common analytic application today – survey verbatim classification through building taxonomies. (I remember working with one insurer whose team spent over 3,000 man hours building out text analytics just for categorizing claims notes). ... " 

Brand Destiny is Mobile

In K@W:
" ... The rise of mobile is accelerating faster than ever imagined. In many places in the world it is the only option to access the web. One big result is a leveling of the playing field for brand-advertisers large and small. Brands are getting cheap access to a unique reach unheard of with TV advertising. In this opinion piece, Aprajita Jain, a brand marketing evangelist at Google, and Gopi Kallayil, Google’s chief evangelist for brand solutions, outline the implications of this tectonic change. You can also read about Kallayil’s new book Internet to the Inner-net in our recent interview Connect with your “Inner-net”: Living and working with purpose. ... " 

Further in Retailing today:  Deloitte says mobile activity on the rise.

Cognitive Assistants Helping with Business Strategy

Via Jim Spohrer:

How might cognitive assistants help with business strategy?
" .... Dario Gil is a leading technologist and executive at IBM Research. As Director of Symbiotic Cognitive Systems, he leads a strategy focused on the creation of learning systems that are explicitly designed to collaborate with people to enhance and scale human expertise. The strategy is composed of leading experts in artificial intelligence, multi- agent systems, robotics, machine vision, natural language processing, speech technologies, human-computer interaction, social computing, user experience and interactive design. He received his Ph.D. in Electrical Engineering and Computer Science from MIT.

About one minute into this video, you will see an interesting scenario - two people collaborating with a Watson system on business strategy  ... " 

I add:  Strategy is a place where there is an assumption of longer term forecasts, and thus exactness is not expected. So I can understand an advisory boost might be useful. Yet for some of the same reasons people expect something magically intuitive and less quantitative from an expert strategist, and convincing people a machine can do this may be harder. Still a noble idea.

Emergence of the Buy Button

In Adage: Increased use of the 'buy button'.   Impulse like buying online   ... " ... While Smaller Retailers Jump on New Tech, Bigger Players Take Wait-and-See Approach ... "

Tuesday, December 08, 2015

Statistical Significance at Play

In Insight@Kellogg. What would the frequentists say then, Kenneth?  .... Blinded by Statistical Significance ... Putting too much stock in an arbitrary threshold may lead to bad decisions. ... Based on the research of Blakeley B. McShane and David Gal

Considering AI Breakthroughs

Bloomberg on a breakthrough year in AI.  Good description, though considerable value was produced during the last wave.  And it remains to be seen how well this wave will ultimately pay out to great generated expectations.

An Information Cartography

Information Cartography, illustrationIn the CACM: 'Metro maps' of information and its use in narratives.  Video.   A means to understand use architectures?  Fascinating, but I need to see it with real examples.

" .... Several methods have sought to summarize and visualize narratives.2,28,29 However, most work only for simple stories that are linear in nature. In contrast, complex stories exhibit a nonlinear structure; stories spaghetti into branches, side stories, dead ends, and intertwining narratives. To explore them, users need a map to guide them through unfamiliar territory.

We previously introduced a methodology for creating structured summaries of information we call "metro maps." The name is metaphoric; just as cartographic maps have been relied on for centuries to help us understand our surroundings, metro maps help us understand the information landscape. In this article, we explore methods we have developed for automatically creating metro maps of information. ,,, " 

Mining the Wikipedia for Influential World Universities

In Technology Review: Interesting way to look at existing data.    An alternative approach to ranking. The subject attracts,  how accurate depends on the goal of the method.

Gamification Gamble and Value

Good overview ... Fundamental idea is good, but how is it best inserted into cultures and values?

In Knowledge@Wharton:
Would you invest in a motivational program for your employees that has a success rate of only 60% or 70%, and is based on a field so new that researchers are still arguing about how to define it?

That field would be gamification: loosely, the application of motivational game elements to non-game situations such as business, education and fitness. The profound uncertainty around it isn’t stopping many companies large and small, who report that they are getting some impressive results. Experts from Accenture, Fidelity and other firms shared their successes and ongoing challenges at the recent Wharton conference, “Gameful Approaches to Motivation and Engagement.” ... ' 

Monday, December 07, 2015

Netflix, Analytics and What You Watch

In FutureStartup: Correspondent Bill Franks chief scientist at Teradata, talks about his analytics experiences with Netflix.    Further thoughts: " ... Teradata’s Franks said that if it wanted to, Netflix could almost be a stand-alone analytics firm.  ... “It is getting blurry out there as to what companies are in,” said Franks. “AT&T is not just a phone provider. It is providing TV, and then has its own data. Nike is now manufacturing high-tech electronics and housing data in its data center, not just doing knitted sportswear. Companies can commercialize the use of their data outside of their core business. It is a fascinating time, and a lot of companies will be morphing and twisting.” ... " 

Morphological Modeling

Tom Ritchey sends along an article about structural systems modeling.   Technical.  See previous writing about this in the tags below.   As an alternative to quantitative analytics.  Continuing to cover.

Principles of Cross-Consistency Assessment in Morphological Modelling

For those interested in the more detailed methodological issues of GMA/morphological modelling, the following article is now available at AMG:   Ritchey, T. (2015) “Principles of Cross-Consistency Assessment in Morphological Modelling”, Acta Morphologica Generalis, Vol. 4 no. 2.   Download at: http://www.amg.swemorph.com/pdf/amg-4-2-2015.pdf

Abstract: In General Morphological Analysis (GMA), the Cross-Consistency Assessment (CCA) both serves as a check on the integrity and clarity of the concepts being employed, and allows us to identify and weed out all internally incompatible relationships in order to reduce the total problem space of the morphological field to a smaller, internally consistent solution space. With computer support this solution space can be treated as an inference model. This article examines the methodological principles and practical procedural issues involved in the CCA process, and presents examples from a number of client-based projects.

Silicon Angle Examines the Analytics Market

Including mention of the Cortana BI and lead management release:

In Silicon Angle: With chief marketing officers expected to outstrip the technology spending of their peers in the IT department by the 2017, it shouldn’t come as much of a surprise that vendors are paying more attention to analytics, which has emerged as one of the hottest items on the shopping list. Microsoft Corp. joined the fray last week with a new release of its hugely popular lead management platform that uses machine learning algorithms to help organizations close deals more effectively.  ...  " 

Classifying Images with TensorFlow

From the Google research blog.  A non technical post on how to use their now open source TensorFlow capability for image classification.  Instructive as to capabilities.  Links to simple tutorials there. I see my former colleague, Admiral Grace Hopper is assisting.

Cognitive Capabilities for Online Merchants

Interesting to see Watson derived capabilities being used for  merchandising and marketing data insights.

" ... IBM ...  today announced new commerce capabilities that help online merchants easily gain the insights needed to evaluate category and product performance and make quick and effective merchandising decisions. Leveraging cognitive capabilities from Watson Analytics, IBM Commerce Insights allows practitioners to gain a real-time view into customer behavior and market factors that are impacting their ... "

Storytelling with Data

An increasingly discussed topic.  Here a site devoted to the topic.  Is it marketing or is it the presentation of data that allows it to be best used to present the patterns it contains?  Science or marketing?

Sunday, December 06, 2015

Service Design

Useful description of a need for design in service process delivery.      See also Service Design.

" ... SCAD Service Design students proudly (or should I say, courageously...) present the first product based on the Service Design Taxonomy. 

Actually, it all started with an analysis that tried to map all functions that a service blueprint could facilitate. One of the most interesting ones was "facilitating cross-functional communication in support of customer-focused solutions" (Bitner, Ostrom 2008). That it is why it was named "huddle," the Service Design Huddle.

Basically, it is a workshop led by Service Designers and promotes a conversation with a group of stakeholders about an existing or new service proposition. This conversation should have from two to four hours duration and discuss each one of the 68 words that compose the taxonomy. The main goal is to select a "size" for each word, represented by a folding paper card, following a specific category sequence. The size of the card  (1/1, 1/2 or 1/4) should be selected based on the consideration or effort that the participants believe is or will be exerted to that word within the specific service. ... " 

Trading and Exchanging Data for Services

How about a Hub of All Things?

" ... The HAT is a personal data platform created to do two things. First, the HAT acts as a personal data container and management system. This enables us as individuals to acquire our own data from Internet-connected objects or services and then to transform it by placing it into its wider context. This means that it becomes more meaningful and useful for making better decisions in our lives.

Second, it allows us to trade and exchange our own data for services in a standardised and structured manner. The HAT functions as a multi-sided market platform, in line with its objective to engineer a market for personal data. It aims to provide opportunity on both sides of the exchange  ... " 

How the Data Lake Works

Part of an exploration of useful linkages between data and analytics to promote easier data use and sharing.  Brought to my attention by Kirk Borne.  A PDF from Booz Allen.