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

Wednesday, July 24, 2013

Predictive Analytics Watch Outs

 In the article they say these are 'screw ups',  and I agree.  I think they are more broadly applicable than just to what is called 'predictive analytics' today. So consider scanning these as general watch outs.  Most of these examples are very broad and any analysts should know.

CFOs Thinking Fast

Brought to my attention by a collaborator.  I am a believer in quantitative analytic thinking and a critic of intuitive flash thinking, so obviously agree to these cautions,  In the WSJ: 

" When CFOs Think Fast ……They may get the company into a lot of trouble. Here's where they go astray—and how to avoid it. ... Daniel Kahneman, in his best-selling book, "Thinking, Fast and Slow," says there are basically two modes of thinking. Fast and slow. Fast thinking is intuitive, he says. It arises from experience and broad unconscious associations between ideas. Slow thinking, in contrast, relies on reason, analysis and deliberation. Both have their time and place. But the psychology professor and Nobel laureate in economics warns about the risks of too much fast thinking when thinking slowly would lead us to better decisions.... "

Expectation Gaps Between IT and Business Leaders

In the Forrester Blog by Dan Bieler.   More in the article:  " ... new Forrester survey data indicates several gaps in opinion about network infrastructure aspects between business and IT leaders. We see a risk that IT will purchase network and collaboration assets that do not address the demand by business lines. Similarly, there is a risk that business lines remain unaware of network and collaboration assets that IT has put in place. Under both scenarios, businesses waste valuable resources and end up with an inefficient network and collaboration infrastructure. ... " 

Tablet Market

A look at the current tablet market.  Tablets continue to replace laptops for personal computing and information access  power.  Apples share drops among other contenders.

Tuesday, July 23, 2013

Privacy and Compliance Expertise from P&G

Once again I point to Sandra Hughes' Web site.    She was previously chief privacy officer at Procter & Gamble.   We were colleagues in those years.  We met again today to discuss privacy issues caused by the recent NSA revelations, data security and over viewed compliance strategies and related technological methods.  Sandy Hughes would be particularly good as an adviser or board member to companies that have privacy exposure, as many companies do today.  Even if they don't think they do.  The link above also provides a nice graphic describing her areas of experience and influence. More about her and contact information.

Computing Table Tops

touch screen fingerprint readerWe saw demonstrations of tabletop computing nearly a decade ago at several labs.  Yet the format has not taken hold, though a few systems are still available.  At the time I did think that this would be the future of the interface,  along with walls that we could gesture to and use as large scale displays.  Yet integrating furniture with displays is not mobile, nor do tables and displays get updated at the same rate.    We saw a similar conclusion when a number of manufacturers put displays on kitchen devices like refrigerators.

Income Mobility in the US

MJ Perry writes and provides an informative map.  Shale oil is a big  driver.

Retail and Big Data

A case of retailer A&P using a big Data Portal.  " ... This cloud-based portal enables retailers to share their proprietary data – including point of sale (POS), inventory, loyalty and more – with their supplier partners. It allows a retailer's vendors to work with and analyze the same data the retailer does in order to collaborate on mutual business goals. ... " 

Crowdsourced AI Includes Everyday Knowedge

In New Scientist:  An AI that is apparently equivalent to a 4 year old in a verbal test.   A big part of intelligence is our ability to manage our knowledge of the world and apply it to decisions. " ... Step forward ConceptNet. Developed by Catherine Havasi and her team at the MIT Media Lab ConceptNet draws upon a crowdsourced database of millions of statements describing simple relationships between everyday objects, such as "a fawn is a deer" or "ice cream is capable of melting".  Havasi describes the system as containing "the kind of information that everybody knows about the world but that nobody ever writes down because we learn it too early". ... " 

3D Interactive Playground

Had not heard much from Disney of late in the digital world.  But now an interesting play. Reminiscent of other simulation world ideas, which are usually difficult to pull off.  Now can it be used to accurately mimic real spaces?  In one experiment we built a system that combined physical touch and a system to design store spaces.   " ... Disney,a company that knows a thing or two about bringing characters off the page and into the real world, has produced a prototype of a new kind of interface that does just that.  A tiny mobile projector beams out a figure that moves around a surface embedded with invisible infrared-absorbing sensors--a book, a desktop, a game board, a table, or a wall. A camera attached to the projector tracks the figure and reads the code printed in the infrared-absorbent ink, which causes the figure to interact meaningfully with the sensed object. A projected figure can track mud from an illustration of a puddle, or jump up and down on a paper trampoline. ... "

Monday, July 22, 2013

Datification of Business

Datification is a new term for me, but a natural next step.  In our own enterprise we drove the use of data for a long time.   Many other companies are now discovering the value of this.  And much more expertise will be needed in handling and analyzing that data, driving many opportunities.

Stouffer Mobile Marketing

In iMediaConnect:  An example of how a company links past behavior to promotion.   Apparently a big success:   " .... How do you reach customers right at the point of purchase? This case study shows how one food brand ... revolutionized digital shopper marketing. Food on the Table is a mobile app that provides recipes to consumers based on their grocery list and sales in the store. When the company approached Stouffer's for a partnership, neither of them could have imagined the success they would find when trying to deliver relevant, truly native experiences to consumers at the point of purchase. Here's how they did it.... "

Boost Creativity not Just Sales

In Fastcompany:  Do interesting things.  An early goal of our innovation centers.  Make a space that will boost innovaton for workers.  Connect it with lots of outside resources.  Bring in speakers and innovations and let workers experience them in detail.  Lesson learned: Do not let it become merely a place to market inbred ideas to your customers.  " ... Legendary ad creative John Hegarty is known for saying "do interesting things and interesting things will happen to you" and he walks the talk. Here, he shares some insight on getting, and staying more, creative.... " 

Data Points Mean Something

They do.  Book of interest brought to my attention: Data Points: Visualization That Means Something by Nathan Yau

"Whether it's statistical charts, geographic maps, or the snappy graphical statistics you see on your favorite news sites, the art of data graphics or visualization is fast becoming a movement of its own. In Data Points: Visualization That Means Something, author Nathan Yau presents an intriguing complement to his bestseller Visualize This, this time focusing on the graphics side of data analysis. Using examples from art, design, business, statistics, cartography, and online media, he explores both standard-and not so standard-concepts and ideas about illustrating data.

Shares intriguing ideas from Nathan Yau, author of Visualize This and creator of flowingdata.com, with over 66,000 subscribers. Focuses on visualization, data graphics that help viewers see trends and patterns they might not otherwise see in a table ... " 

Stat Methods

See Robert Kabacoff's web site: StatMethods.  And his associated blog.   These cover statistical methods delivered in R.  Devoted to learning R quickly and providing examples for its use for real statistical and visualization problems.

Data Preparation Mistakes

From Catlin Garrett via Vincent Granville.  Good thoughts worth reviewing.  Boils down to really understanding the source of your data elements, and how they were generated. Examine the elements, if there are too many to examine, visualize them and look for anomalies.  Get access to an expert on the business generation and use of the data and have them do the same.   Comments have some good points too.

Sunday, July 21, 2013

Retail Facial Recognition

Another example of using facial recognition for retail experience. Here apparently to detect VIPs. Had talked to people developing casino applications to detect people banned from the premises, where it was suggested that the same thing could be used to detect big spenders and personalize their presence.  But recognition may not be seen as a positive event.

And more generally Google Glass Facial Recognition.

Unique Silicon Valley

I used to give informal tours of Silicon Valley starting at Stanford, and often got the question: Why here?   I had a different perspective because I was usually a visitor as well.   But I was a student of industrial university collaboration.  This article has some good thoughts.  On why it is difficult to replicate.

Location Analytics from Placed

Location Analytics example from Placed.  Dated but interesting with broad statistics: " ..... In the largest showrooming study to date, Placed identified Best Buy and Target as high-risk retailers for Amazon customers to view offline and buy online. Surprisingly, Best Buy and Target were not the most at risk for showroomers taking over their aisles. The results from the Placed: Aisle to Amazon Study found that consumers who showroom and then purchase on Amazon are 20% more likely to visit Best Buy and 15% more likely to visit Target than average, but Bed Bath & Beyond, PetSmart and Toys ‘R’ Us all face greater risk. ... " 

Reliability of Big Data Predictions

In Knowledge Core:  Intriguing piece.  Philosophically oriented. A question we have asked of all analytics for many decades.  With the usual cautions and prescriptions.  But I ask:  Has this changed because you have more or more volatile data?  We always sought more data, and often did not have it.  Now we have it, and what does that mean?  Better predictions?  Or more difficult ones to explain?