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

Thursday, April 18, 2019

Free Graph Algorithms Book

My favorite analytics methods.

Hidden Data Patterns Only Relationships Can Find - Free O’Reilly Book

Graph analytics use relationships to reveal structural and predictive insights hiding in your data. Whether you are building dynamic network models, mitigating risk and fraud or forecasting real-world behavior.

Check out this great resource – just published today: The O'Reilly book, Graph Algorithms: Practical Examples in Apache Spark and Neo4j. Get your free copy now.

Pass this offer along to anyone on your team who would like to explore graph algorithm examples with working code and sample datasets for both Spark and Neo4j. Also included is a chapter on how graph algorithms enhance machine learning accuracy and precision.

Best,  Amy Hodler & Mark Needham, co-authors
@amyhodler & @markhneedham

Get your free copy now

Is it still about Paper Coupons?

Worked with paper coupon redemption analytics from the very beginning at very big CPG.   Is it about their physical nature?     Will there always be room for the  paper coupon?  Would not have guessed it, but the statistics still show it.

Will America’s love for paper coupons ever die?
by Guest contributor   MarketingCharts staff in Retailwire 

Through a special arrangement, presented here for discussion is a summary of articles from MarketingCharts, which provides up-to-the-minute data and research to marketers.

Despite a slight rise in the preference for paperless discounts, more adults still prefer paper coupons, according to the latest annual survey of coupon use from Valassis.

Fully half of adults in the U.S. prefer to get coupons in the mail, a figure that has continued to rise over the past three years. The proportion of respondents who prefer to get coupons from a coupon book found in a newspaper continues to rise as well, from 42 percent last year to 44 percent this year.

Some 42 percent agree that they prefer to get paperless discounts from the internet that they can download onto their store shopper/loyalty card. And even though a smartphone offers the convenience of carrying coupons anywhere, just 38 percent say they prefer paperless discounts on their smartphone/mobile device.  .... " 

WiFi as Hospitality, Retail Value

Now having experienced Wifi in Hospitality since its inception, I recall having a conversation with a Hotel Chain innovation group.  At the time they were unconvinced that Wifi had value, and there too many dangers.    Things have changed and continue to evolve.  Incentivizing repeat business.  Now a days I mostly wander through retailers checking stocking and possible product use, and the Wifi experience could be much better.  They need to make it very clean, and incentivize repeat business and value with the right data.  You are an edge of the retail IOT.  And how about making the edge more intelligent as an assistant?

Wi-Fi Without Reservations   By McCall Bunn in Cisco Blog

Partner Success Story

Providing high-quality Wi-Fi is an IT investment most businesses incur. Customers expect Wi-Fi at every destination. It is crucial to post pictures of appetizers, split checks, and look up the difference between ‘bucatini’ and ‘fettuccini’. It has become a necessity when dining out.

And a large hospitality group, previously known as Batali and Bastianich (B&B), with locations around the world, already knew this. Yet they had a problem offering quality connection in several of their Las Vegas restaurants. And, they were struggling to identify their customers. So, they came to a Cisco Solution Partner iValu8 for help.

B&B agreed to a one-month trial with iValu8 to pilot VivaSpotTM, a marketing campaign platform that runs over Wi-Fi, which then implements Cisco Meraki as the reliable and scalable way to power the solution.

Not only were their customers now able enjoy high-quality Wi-Fi during the trial, but they began to collect customer data to incentivize repeat business, gather feedback, and amplify the hospitality group’s social media presence. They could track the results in real-time and launch campaigns on any device from any location. And, free Wi-Fi became a key marketing assets.

With multiple locations around Las Vegas, VivaSpot also cross-promoted other properties of the hospitality group when customers logged-on to their current Wi-Fi. And the results? An overall increase in spend from customers’ during their stay at all of the properties.

The hospitality group found so much success with the pilot, they’re ordering more. They are bringing Cisco Meraki and iValu8’s VivaSpot to eleven different locations from NYC to LA. Taking the group from a one-month trial to over $100,000 in revenue.  ... "

DJ's Spin Code

I changed the title,  they will likely not write code in the future, they will create algorithms by some interface other than 'writing code'.   Coding, as it has developed, is far too inefficient and error prone.

DJs of the Future Don't Spin Records—They Write Code
in Wired  By Michael Calore

Artists in the underground electronic music culture are performing live-coding shows or "algoraves," in which they program software algorithms to create new forms of music. Musicians synthesize individual sounds on their computers, then direct the software to string those sounds together based on a set of predefined rules; the end product has the artist's signature, but is algorithmically sculpted. When the same routine is run again, the song will sound familiar and contain the same elements, but the composition will be structured differently. Performances often are enhanced with screens displaying the running code as trippy visuals. A popular venue for this emergent art form is the Algorithmic Art Assembly, a two-day festival in San Francisco ..... "

Spying on Your Smart Home

Have now had a smart home lab for a long time.  Could have used this for some time, in particular to understand how the home interacts with external contexts, like the car, the store, the Internet.  Like the idea.  So many things in the lab are IOT members.   Has to be marketing insight here.

Spy on Your Smart Home With This Open Source Research Tool 
In TechCrunch     By Natasha Lomas

Princeton University researchers have built an open source Web app that allows homeowners to monitor their smart home devices. The IoT Inspector is designed to help consumers analyze the network traffic of their Internet-connected appliances, mainly to determine whether those devices are sharing their information with third parties. The researchers said the IoT Inspector requires no special hardware or a complex setup, making it easy for consumers to deploy smart home monitoring. The team acknowledged it hopes to use data collected by the app to advance Internet of Things (IoT) research, including insights into privacy, security, and network performance risks of IoT devices. The researchers said the app can track the Internet activities of as many as 50 devices on a network. .... " 

Employee Surveillance is Data

Not an unexpected thing.  Its another form of data and is naturally part of efficiency considerations.  Even its 'architecture',  like those mentioned in recent silo readings, can be important to the accuracy of decisions being made.  Data exists in work contexts, decisions use data, which includes how people (and AIs) use that data.

Employee Privacy in the U.S. at Stake as Corporate Surveillance Technology Monitors Workers' Every Move     CNBC   By Ellen Sheng

The advent of technologies that let businesses track, overhear, and monitor employees on company time is raising issues about corporate surveillance. Gartner estimated last year that 22% of organizations worldwide in various sectors use employee-movement data, 17% track work-computer-usage data, and 16% access Microsoft Outlook or other calendar-usage data. Products of interest include Amazon's recently patented ultrasonic bracelet, which can localize warehouse employees and monitor their interaction with inventory bins via sound pulses. Meanwhile, last year Walmart patented a system for eavesdropping on workers and customers, which tracks employee "performance metrics" to ensure employees are on the job by listening for certain noises. Some makers of monitoring technologies are developing deployment guidelines to allay employer and employee privacy concerns, but advocates fear worker privacy could be compromised without appropriate regulation.  ... " 

Silo Syndrome

Its really always about silos.  The silos are formed in part by the data that exists within them.  But also more fundamentally by the trust that exists within decision making.    I have trust within a silo that makes decisions that influences agreed-to goals.   How might further automation change this?

Five-Fifty  in Mckinsey:
A quick briefing in five—or a fifty-minute deeper dive

In this edition:The silo syndrome  Working in silos can cause tunnel vision, tribalism, and weak corporate performance. What’s a silo-buster to do?  ... "

Wednesday, April 17, 2019

Automating Machine Learning with Azure

This example was sent to me, a straight forward example of using Azure.   Always looking for useful examples of better automating at least initial tests of a machine learning example.  Big proponent of quick, early,  cheap tests of even complex modeling efforts.      Look for ways to get the idea in front of other analysts, decision makers, data providers.   Even a sketch on a board is worthwhile, but showing something interactive gives a clear taste of the potential value.

How to forecast energy demand with Azure Machine Learning | Azure Makers Series
Use Azure Machine Learning to create a model and apply it to a real-world scenario: predicting energy demand and expected load on energy grids - a critical business operation for energy companies. The same principles apply across use cases, so you can adjust for your organization’s critical operations and needs.  

GitHub Repo: https://github.com/FrancescaLazzeri/A... 

Create your Azure free account: https://aka.ms/J1PxFdcK2tY

Follow Francesca on Twitter: https://twitter.com/frlazzeri

Faster and Smaller Neural Nets

Fascinating development.   Smaller usually means faster with training nets.   Smaller can also mean easier implementation at the IOT edge.  Now will they be as accurate?   It is all about more efficient perception.  Closer to human.   Technical piece in Google AI.  Intro below, more at the link:

MorphNet: Towards Faster and Smaller Neural Networks  in Google AI.   Wednesday, April 17, 2019

Posted by Andrew Poon, Senior Software Engineer and Dhyanesh Narayanan, Product Manager, Google AI Perception 

Deep neural networks (DNNs) have demonstrated remarkable effectiveness in solving hard problems of practical relevance such as image classification, text recognition and speech transcription. However, designing a suitable DNN architecture for a given problem continues to be a challenging task. Given the large search space of possible architectures, designing a network from scratch for your specific application can be prohibitively expensive in terms of computational resources and time. Approaches such as Neural Architecture Search and AdaNet use machine learning to search the design space in order to find improved architectures. An alternative is to take an existing architecture for a similar problem and, in one shot, optimize it for the task at hand.  .... " 

Distributed Ledger as new Enabler

Bold claim here in Supplychain Brain.

Why Supply Chain Technology Needs Blockchain
 Jon Kirkegaard, SCB Contributor

Blockchain: A New Enabler

" ... What’s encouraging is a growing wave of awareness of blockchain and distributed ledger technology (DLT). Interest in blockchain appears to be providing an avenue for the true scaling of real-world collaboration, by embracing decentralized technology, encrypted security and peer-to-peer networking technology.

Blockchain is increasingly being discussed in the context of supply chain, but the focus is often on the replacement of existing applications such as tracking. In my view, this is not where blockchain can be of the most value to the supply chain. The greatest need, and highest return on investment, lies with processes that can’t be automated using centralized technology.

As an example of this disconnect, despite decades of new supply-chain technology, as much as 90 percent of all real planning and coordination is still done via spreadsheet and e-mail. In particular, the coordination of S&OP build plans outside of one department or company is almost always carried out in that manner.

Perhaps you have witnessed in your organization attempts to deploy connected planning technology and kill the spreadsheet, in favor of one real-time instance of advanced planning and scheduling. Has that worked, or has it just created more underground use of spreadsheets, e-mails, whiteboards and conference calls?... "


New, brought to my attention, short non-tech articles on advancing tech.  With some interesting embedded links.

How AI will change your shopping habits   By Christine Persaud

Artificial intelligence (AI) is influencing everything from photography to gaming, home entertainment, education and autonomous vehicles. AI is also reinventing the way we shop.

AI has the potential to impact many industries, from healthcare to manufacturing. The University of Waterloo is just one of many companies and institutions that has opened an AI institute dedicated solely to complex and detailed research on the technology’s applications.  ... " 

About Futurithmic

Our editorial mission is to explore the implications of emerging technologies on society, business, politics and the environment of tomorrow. We aim to inform and inspire through thoughtful research, responsible reporting, and clear, unbiased writing, and to create a platform for a diverse group of innovators to bring multiple perspectives.  ... " 

Machine Teaching

Nice thoughts here.   If we can learn we should be able to teach.

Deep Teaching: The Sexiest Job of the Future 
Carlos E. Perez   in Medium

Microsoft Research has a recent paper (Machine Teaching: A New Paradigm for Building Machine Learning Systems) that explores the eventual evolution of Machine Learning. The paper makes a clear distinction between Machine Learning and Machine Teaching. The authors explain that Machine Learning is what is practiced in research organizations and Machine Teaching is what will eventually be practiced by engineering organizations. The teaching perspective is not only different from the learning perspective, but there are obvious advantages in that concept disentanglement is known a priori ... "

See the Microsoft Paper mentioned.

Tuesday, April 16, 2019

Simulation is Very Useful for Analytics Validation

I am a long time proponent and practitioner of simulation methods.   We simulated plant design and process, warehouse operations,  woodland growth and management, consumer in aisle behavior, coffee roasting,  advertising selection and delivery ... and much more.   Always a good approach to validate prescribed analytics and AI.    It needs be better linked directly to analytical solutions.

Advance Your Process Improvements with Simulation Technology  in the APQC Blog  By Lochlyn Morgan  Posted in Process and Performance Management

I recently spoke with Luis Lopez, manager of process improvement at the Port of Vancouver, to discuss the role that simulation technology has in process improvement, advice on piloting simulation software, and a few lessons learned from his hands-on experience with simulation technology.

What role can simulation technology play in process improvement work?

Simulation technology plays an important role in the improvement of complex processes as it provides a non-invasive, risk-free, and cost-efficient method to identify and analyze the underlying factors that may contribute to poor process performance and evaluate potential improvements. Simulation technology is key when testing improvements in the real process can be very costly, risky or lengthy.

Simulation models also provide a great way to engage project teams in the design and development of process improvements. Advances in 3D graphics have made it relatively simple to make detailed 3D simulation models of an operational process. These 3D renderings allow project teams to quickly visualize their ideas and identify potential benefits and implementation barriers. This is why I like to say that “if one picture is worth a thousand words, an animated 3D simulation model in a process improvement project is worth a thousand pictures.”  .... " 

Google has an AI Cloud Platform. Lets link it with BPM

Quite some detail for making AI applications work with the cloud in this new production factory for AI in the Cloud.  I like the idea of standardizing such learning projects and installed solutions. I would also like to see this kind of work linked with business process models like BPM.

AI Platform

Create your AI applications once, then run them easily on both GCP and on-premises.

Take your machine learning projects to production

AI Platform makes it easy for machine learning developers, data scientists, and data engineers to take their ML projects from ideation to production and deployment, quickly and cost-effectively. From data engineering to “no lock-in” flexibility, AI Platform’s integrated tool chain helps you build and run your own machine learning applications.

AI Platform supports Kubeflow, Google’s open-source platform, which lets you build portable ML pipelines that you can run on-premises or on Google Cloud without significant code changes. And you’ll have access to cutting-edge Google AI technology like TensorFlow, TPUs, and TFX tools as you deploy your AI applications to production.  ... "

A testimonial they provide:

" ... In retail, it’s important to provide customers with easy access to alternative products or recommended add-ons. We train our own machine learning models with TensorFlow on Google Cloud ML, and we automate the periodic retraining of these models with Kubeflow Pipelines. Together with AI Hub, useful for sharing models between data scientists, we can now iterate faster on our models, and automatically deploy them to staging and production. ... '    Lucas Ngoo, co-founder, CTO, Carousell

See also: https://techcrunch.com/2019/04/10/google-expands-its-ai-services/

Mathematica Expands

Been a long time since I worked with Wolfram's Mathematica.  Was always impressed with what the package provided.  Especially useful for people that already have a math background.    Also good to let people/students with a strong interest in Math expand their mathematical powers.

They are coming up with a considerable update I have started to scan.  Lots of new descriptive documentation that looks good.  Now includes examples about how to do DeepLearning and blockchains with Mathematica.  The article below has a long description of the capabilities.

Version 12 Launches Today! (And It’s a Big Jump for Wolfram Language and Mathematica)
April 16, 2019 — By Stephen Wolfram  ... '

Predicting Sales Behavor in Real Time

Sales Prediction in real time.

6sense raises $27 million for its marketing and sales predictive analytics tool  By Manish Singh @REFSRC

6sense, a San Francisco-based startup that uses big data to predict in real time when people are looking to buy products, has raised $27 million to grow its marketing analytics tool.

The funding round was led by Industry Ventures, with existing investors Bain Capital Ventures, Battery Ventures, Costanoa Ventures, Salesforce Ventures, and Venrock also participating. 6sense, which described the new round as “growth funding following series B,” has raised $63 million to date.  .... " 

AI for the Diagnosis and Management of Eye Disease

Another example of the use of AI for healthcare management.  Considerable detail at the link.

A major milestone for the treatment of eye disease  in DeepMind
We are delighted to announce the results of the first phase of our joint research partnership with Moorfields Eye Hospital, which could potentially transform the management of sight-threatening eye disease.

The results, published online in Nature Medicine (open access full text, see end of blog), show that our AI system can quickly interpret eye scans from routine clinical practice with unprecedented accuracy. It can correctly recommend how patients should be referred for treatment for over 50 sight-threatening eye diseases as accurately as world-leading expert doctors.

These are early results, but they show that our system could handle the wide variety of patients found in routine clinical practice. In the long term, we hope this will help doctors quickly prioritise patients who need urgent treatment – which could ultimately save sight.

A more streamlined process
Currently, eyecare professionals use optical coherence tomography (OCT) scans to help diagnose eye conditions. These 3D images provide a detailed map of the back of the eye, but they are hard to read and need expert analysis to interpret.

The time it takes to analyse these scans, combined with the sheer number of scans that healthcare professionals have to go through (over 1,000 a day at Moorfields alone), can lead to lengthy delays between scan and treatment – even when someone needs urgent care. If they develop a sudden problem, such as a bleed at the back of the eye, these delays could even cost patients their sight. .... " 

Monday, April 15, 2019

On Automated Help

Our own work in this space often touched on this:   A person does like to talk to a clear, knowledgeable person who can solve our problem quickly and directly.   But in what context would they live with less than that?   And how does that alter their later commercial behavior regarding the service?   Quite an important issue for any assistant role.

In HBS Working Knowledge.  An infographic style look at ongoing research. With some backup statistical results.

Infographic: Can I Please Speak to an Actual Person?
 by Katherine Vizcardo and Danielle Kost

Customers still want the option to access human help in automated service—even if they don't use it, says research by Michelle A. Shell and Ryan W. Buell.  .... ' 

AI Creates a Sport

Always been interested in creativity from AI.   Here an AI 'observes' sports  and creates a new one.   Is this creative, predictive, adaptive?  Is the idea useful for other kinds or contexts of human behavior?  Could this be redone for serious games?

Speedgate is based on data from hundreds of existing sports.

Jon Fingas, @jonfingas
12h ago in Personal Computing in Engadget

Many existing sports have their roots in hundreds (if not thousands) of years of human tradition. But what if you asked computers to create a sport? You now know how that can turn out. The design agency AKQA has introduced Speedgate, reportedly the first sport envisioned by an AI. The event has six-player teams competing on a field with three open-ended gates. Once you've kicked the ball through a center gate (which you can't step through), your team can score on one of the end gates -- complete with an extra point if you ricochet the ball through the gate. You can't stay still, either, as the ball has to move every three seconds.   .... " 

Converting Insight into Action for CPG

Useful description about the ability to integrate CPG and behavior.

Using AI to Translate Insight into Action
By Steven Hornyak, Symphony Retail AI, CPG Solutions - in ConsumerGoods

Today, consumer packaged goods brands are tasked with much more than conveniently connecting shoppers with their favorite items. Consumers want brands to anticipate their needs and make relevant products available to them wherever they shop.

But many CPGs have yet to crack the code. They’re spending countless hours and dollars investing in trade promotions even though 72% of those programs don’t break even. They also invest heavily in new product development even though 95% of new products fail. To be successful in today’s market, CPGs must acquire and learn from a deeper understanding of consumer behavior to make more strategic, more intelligent promotion decisions.

Channel blurring has led to more complexities than ever for both CPGs and retailers. Consumers have the option to shop in physical stores or online, a blend of the two with buy online and pick up in store, or to not even actively shop at all by using a subscription service. On top of this, the growth of today’s private label and regional-based players have led many product lines to hit a premature revenue ceiling. ... " 

Cases for AI in Advertising

We  used AI in the earliest days, but did not have the needed tools, there are many more now.

The Case for Investing in AI for Advertising in ChiefMarketer

Posted on April 15, 2019 by Sven Lubek

There are numerous examples of high profile brands experimenting with artificial intelligence (AI) in creative ways. For example, Lexus recently worked with IBM Watson to release the first AI scripted advertisement.

AI is ultimately here to improve people’s lives both at work and at home, yet many organizations are still timid about investing in the technology. Here are some strong cases for investing in AI for advertising today:

Achieving True Scale and Engagement

Marketers are investing in AI to deliver advertising that is relevant, contextual and hyper-personalized to individual consumer preferences. Automation is an important component of driving this capability.

Adam Powers, CEO at Tribal Worldwide, shared at Mobile World Congress 2019 how they use AI in advertising to create contextual experiences for users: “Offer an experience where the application of AI is an invisible factor—emotional engagement and conversion focus. Magic can happen in the details, the small things and looking at the practical application of AI. For example, a client in Indonesia uses machine learning to forecast fashion trends by feeding in various data points, and image uploading to forecast in which part of the region certain products will sell.”

“Brands need to try to keep up with changing consumer behavior,” added Neil Stubbings, CRO at IV.AI. “It’s the age of availability. A brand should be available on any platform that the customer is, and that’s the challenge and the opportunity for brands to transact with consumers…people are looking for things that feel more native.”  .... ' 

Forrester on Value of Blockchain for the Supply Chain

High level overview of value of blockchains and link to a video:

Blockchain and Supply Chain
George Lawrie, Vice President, Principal Analyst
Martha Bennett, Principal Analyst

Modern supply chains are faster, more dynamic, and more volatile than ever before. The digital economy thrives on scouring the globe for new markets and new sources of supply.  Furthermore, agile supply networks outpace rigid supply chains, anticipating demand from empowered customers. But in contrast with rigid legacy supply chains, binding known and trusted vendors,  an agile network of temporary suppliers,  might damage customer experience, by concealing the origin and characteristics of components or ingredients. Blockchain and distributed ledger technologies can help,  by increasing visibility of supply chain documentation. They can reduce inspections, boost trade credit,  and build trust even with new, or temporary, suppliers. .... " 

On the use of Microservices

Fascinating case study that I have passed the link on to several clients.  Below the intro:

How we moved from a giant monolithic system to microservices in Medium
By Arjun Dixit  @ Rebel Foods (Part 1/2)

Defeat them in detail: The Divide and Conquer Strategy. Look at the parts and determine how to control the individual parts, create dissension and leverage it. — Robert Greene

With this being said, we at Rebel Foods tech, strongly believe in the idea of breaking down a large monolith system into the small sets of microservices. Here I would be describing the purpose and the journey behind it.

Overview of the monolith giant system of Rebel Foods:

When we started in 2011 with one brand FAASOS the requirements were quite clear to us — “Build a system that can handle 15k transactions per day” and 15k was just the number of orders expected daily, behind this Rebel Foods manages the complete lifecycle of on-demand food right from the inventory to our supply chain to content management to user channels like our mobile apps, websites and our third-party partners to our cloud kitchens to finally making the last-mile fulfilment. Looking at the fast-paced market we had to deliver a system that can produce a flawless experience while being 99% fault tolerant. We came up with a central core system design that would interact with our main database and would expose the APIs to perform various operations required to satisfy the above needs. We called this system the V1 architecture of Rebel Foods.  ... "

Sunday, April 14, 2019

Cyber Agriculture

My own strong interest and background in botany makes me consider this closely, consider the implications.  In particular have been experimenting with basil flavors for years.   While the smells of basils can be considerable, they often do not perform as distinctly in dishes.   Can we program in robust flavors?  Following.

MIT’s ‘cyber-agriculture’ optimizes basil flavors   By Devin Coldewey @techcrunch

The days when you could simply grow a basil plant from a seed by placing it on your windowsill and watering it regularly are gone — there’s no point now that machine learning-optimized hydroponic “cyber-agriculture” has produced a superior plant with more robust flavors. The future of pesto is here.

This research didn’t come out of a desire to improve sauces, however. It’s a study from MIT’s Media Lab and the University of Texas at Austin aimed at understanding how to both improve and automate farming.

In the study, published today in PLOS ONE, the question being asked was whether a growing environment could find and execute a growing strategy that resulted in a given goal — in this case, basil with stronger flavors.

Such a task is one with numerous variables to modify — soil type, plant characteristics, watering frequency and volume, lighting and so on — and a measurable outcome: concentration of flavor-producing molecules. That means it’s a natural fit for a machine learning model, which from that variety of inputs can make a prediction as to which will produce the best output.

“We’re really interested in building networked tools that can take a plant’s experience, its phenotype, the set of stresses it encounters, and its genetics, and digitize that to allow us to understand the plant-environment interaction,” explained MIT’s Caleb Harper in a news release. The better you understand those interactions, the better you can design the plant’s lifecycle, perhaps increasing yield, improving flavor or reducing waste.  .... " 

Case Study for a Global AI

Considerable case study, contains lots of good things to consider,  but not enough about the goals involved.  Good details but not enough of them.   Its a global brand, so how much do we need to consider varying languages and cultures?   In our own example we found that the training data quality varied strongly because it was gathered based on those cultural differences.  In some cases there was insufficient data for a particular culture.  I like particularly the involvement of decision makers early seems to be carefully considered. 

Building an AI For A Global Brand  by Alexandre Gonfalonieri in Medium

Building an AI solution for a global brand is actually quite challenging. From figuring out the best business issue for AI to the release of the solution, many things can impact the project. Through this article, I wanted to share with everyone my own personal experience.

AI has become a buzzword in almost all industries and most decision-makers want to start or have already started implementing AI solutions.  ... "

A Better Way to Multiply

Useful for cryptography and methods that require fast hashing on IOT edge applications?

Mathematicians Discover the Perfect Way to Multiply  in Quanta Magazine

By chopping up large numbers into smaller ones, researchers have rewritten a fundamental mathematical speed limit.

On March 18, two researchers described the fastest method ever discovered for multiplying two very large numbers. The paper marks the culmination of a long-running search to find the most efficient procedure for performing one of the most basic operations in math.

“Everybody thinks basically that the method you learn in school is the best one, but in fact it’s an active area of research,” said Joris van der Hoeven, a mathematician at the French National Center for Scientific Research and one of the co-authors. ... " 

Consumers Breaking with Legacy Brands

Still nostalgia equity is a powerful thing for many brands.

Nostalgia Is Not Enough: Why Consumers Abandon Legacy Brands
Mar 25, 2019 Strategic Management  North America

Earlier this month, Sears ended a nine-decade presence in Lincoln, Nebraska, when it closed its store at the Gateway Mall. So it was, too, at Park City Center in Lancaster, Pennsylvania, where that town’s Sears store was one of dozens shuttered nationally in yet another wave of contraction by the once-mighty retailer.

The closings set off the expected misty-eyed recollections about the legacy brand and the cherished place it occupied in hearts across the country. In Colorado, where Sears closed two stores in Colorado Springs and one in Pueblo, a columnist for the Gazette mourned the loss. But she also admitted that her February visit to report on the closing was the first time she had been to Sears in a decade. “I left empty-handed, and a little heavier-hearted,” wrote Stephanie Earls.

Among legacy brands, Sears is in similar, troubled company. Payless ShoeSource is liquidating its 2,100 U.S. stores. Toys “R” Us — where many a young American parent remembers buying his or her first Transformer or Super Soaker – closed its 730 locations last year and is struggling to come back in some form post-bankruptcy.

You might have expected that the pull of nostalgia would have protected these brands from the retail re-sorting underway. Customers have emotional connections to certain stores — places where their parents brought them as children and where they did their first Christmas shopping, and developed certain buying habits and loyalties.

So what was the breaking point for customers? Price? Experience? Convenience? Why, in the end, are customers abandoning their shopping heritage and breaking up with brands? ... '

Saturday, April 13, 2019

Comparison of Analytic Methods

Becasue I often work with different kinds of practitioners, I often get questions like:  How is this different from statistical methods, from Operations Research .... ?  I was trained in earlier methods that often had the same goals, but with no claim of being a science.   Just math and data based techniques that embed some useful goals and constraints.    In prep for an upcoming presentation I reviewed the below to consider the differences.    There are lots of overlap here,  so you can't define them precisely.    Often the best approach can be the one that your clients best understand.    But its good to know and review the general direction of many of these.   Also,  hype does not mean right,  and can confuse the issue when over-emphasized.    Good piece.   Join DSC for more.

16 analytic disciplines compared to data science  in DSC

Posted by Vincent Granville on July 24, 2014 

What are the differences between data science, data mining, machine learning, statistics, operations research, and so on?

Here I compare several analytic disciplines that overlap, to explain the differences and common denominators. Sometimes differences exist for nothing else other than historical reasons. Sometimes the differences are real and subtle. I also provide typical job titles, types of analyses, and industries traditionally attached to each discipline. Underlined domains are main sub-domains. It would be great if someone can add an historical perspective to my article.  ... "

Robots Sorting Trash

An excellent example of robot use.  Its a classic example of solving a messy problem that can change the economics of a difficult problem.  How well would it work?

MIT’s bot sifts through trash to do your recycling for you in DigitalTrends

Engineers from the Massachusetts Institute of Technology’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed a new recycling robot that’s capable of automating the process of sifting through trash to distinguish between paper, plastic and metal items. In doing so, it could automate a dull — but entirely necessary — job that few people would want to carry out by hand if there was another option available.

“Although single-stream recycling is really convenient for people, it’s actually a time-consuming and expensive ordeal, requiring significant human labor,” Lillian Chin, a CSAIL Ph.D. student who worked on the project, told Digital Trends. “In developing countries, people have to pick out the recyclable materials from normal waste which can be quite hazardous. [But] even in the U.S., with more automated recycling centers, people are still needed to double-check the machine’s output and manually pick out unrecyclable objects like car engines and plastic bags.”  .... '

Best Data Science Youtube Channels

Good resources, did  quick scan and there are some useful things there,  somewhat varied in quality,  but there are gems.

4 of the Best Data Science YouTube Channels in Medium
Go to the profile of #ODSC - Open Data Science

There are a lot of ways that you can learn about data science and/or stay up to date with the latest trends, but the easiest by far is still Youtube. You could do a search for data science and come up with thousands and thousands of hits, but we’ve got our top four picks to help you get started. Let’s take a look:   ... '

Natural Language and Intent with Ambiguity

Good simple explanation from Tableau on Intent.   Have used the concept now in several projects, and of course there is ambiguity, beyond dictionary-definition,  in the use of many terms within a company.   The ambiguity in context is important to consider.

Machine learning, natural language meet to understand intent
 By Mark Jewett, VP of Marketing, Tableau

Machine learning and natural language processing promise to better translate human curiosity into pertinent answers. If true, these smart capabilities will broaden the use of analytics and reach people who are less comfortable dealing with data. It will all start with helping machines learn to interpret human intent. The key is semantics.

Sometimes intent is simple and explicit, like asking Siri or Alexa if a flight is delayed. This question has clear intention and a simple response—returning the flight status answers the question. Such simplicity is seldom the case when it comes to data analysis. Questions are usually more nuanced, making it hard to correctly assume what the user is really looking for. Natural language is even more tricky where ambiguous terms are common.

It’s also difficult for a machine to understand our intent within a limited context. The machine has the data itself but doesn’t grasp the bigger picture in the same way a person with domain expertise can. Asking “How are my sales doing in the Northeast?” is a lot more ambiguous than the flight status example above.

Ambiguity isn’t a new challenge in data analysis. Different groups within an organization may have different definitions or calculations for the same words: for example, the term “profitability”. Some organizations use central dictionaries (also called data catalogs) to reduce ambiguity and create consistency across the organization. These tools can help provide users with the context they need to understand more deeply. .... " 

Data vs Democracy?

Interesting view ...

Who needs democracy when you have data?

Here’s how China rules using data, AI, and internet surveillance.
by Christina Larson in Technology Review

"... In 1955, science fiction writer Isaac Asimov published a short story about an experiment in “electronic democracy,” in which a single citizen, selected to represent an entire population, responded to questions generated by a computer named Multivac. The machine took this data and calculated the results of an election that therefore never needed to happen. Asimov’s story was set in Bloomington, Indiana, but today an approximation of Multivac is being built in China. ... " 

Friday, April 12, 2019

The Value of Inefficiency

I like the general thought.     Pure efficiency can make us miss things.   Its why  had second thoughts about optimization methods,  they could rarely be implemented directly, and their methods where often   did not give you enough hints at creative alternatives in context.  Podcast and transcript in K@W:

Edward Tenner discusses his new book about how too much efficiency kills creativity, innovation and problem-solving.

Technology is the undisputed champion of efficiency. Tasks that were once complex and time-consuming are now completed in the blink of an eye. But there is a downside to an abundance of technology. In his new book, scholar Edward Tenner explains how too much efficiency can kill creativity, which can turn off avant-garde thinking, innovation and problem-solving. He believes there is a better way to improve our lives through a combination of technology and intuition, and by exploring the random and unexpected.

Tenner, a distinguished scholar at the Lemelson Center for the Study of Invention and Innovation at the Smithsonian, spoke about his book, The Efficiency Paradox: What Big Data Can’t Do, on the Knowledge@Wharton radio show on SiriusXM. (Listen to the podcast at the top of this page.)

 An edited transcript of the conversation follows. ... 

Knowlege@Wharton: What’s so terrible about efficiency?

Edward Tenner: The problem with efficiency is that algorithms let us really learn from experience, they let us codify experience, they let us benefit, they recognize patterns. They are really tremendous at that. For example, I use the Google navigation program Waze. I first started out as a critic of it, but then I got into it more and more. However, the problem with Waze is that every once in a while, it will make a terrific blunder. If somebody relies completely on a system like that, no matter how brilliantly engineered, sooner or later some glitch is going to bite back. However, if they keep their awareness of where they are, if they keep their common sense, and if they keep trust in their common sense, then they can get the most of the program while avoiding those little disasters.

Knowlege@Wharton: Because we are so reliant on technology, are we losing something as a society, as a culture?    .... " 

IOTA Zeus for Retail Payment

A somewhat unexpected application for the novel IOTA 'blockchain' approach.  Following.

Zeux App Enables Use of IOTA For Cryptocurrency Payments at Retail Stores
By Gabriel M

The crypto payments startup Zeux has recently affirmed that it would team up with IOTA (MIOTA) in order to list the token for crypto payments at retail stores. According to the reports, the service will be available with stores that accept Samsung Pay and Apple Pay.

Zeux has revealed the partnership via social media, affirming that IOTA payments will be added for the new token and that people will be able to use it with several different merchants. At the moment, Zeux has a FCA regulatory license and this is probably part of the changes due to its upcoming launch in the European Union next month.

The company is also planning to launch its services in the United States soon. According to the announcements, a big number of shops from grocery stores to coffee shops will be able to receive payments using Zeux.  .... " 

Augmenting Quantum D-Wave Annealing

Mentioned before we connected with D-Wave Quantum Computing early on.   Have posted many items about their work.    Their annealing approach still has great potential for some difficult kinds of combinatorial problems.  Note how this addresses partitioning of sub problems to make the D-Wave approach most useful.  Here a new example of work from Japan in the automotive manufacturing space.  Not enough detail here, but taking a closer look.

Algorithm Optimizes Quantum Computing Problem-Solving 
Tohoku University in ACM

Researchers at Tohoku University in Japan have developed an algorithm that augments the ability of a Canadian-designed quantum computer to more efficiently determine an optimized solution for complex problems. The D-Wave quantum annealer uses the concepts of quantum physics to solve "combinatorial optimization problems;” Tohoku's Shuntaro Okada and Masayuki Ohzeki designed the algorithm with global automotive components manufacturer Denso and other collaborators, to improve this capability. The algorithm partitions a large problem into a group of subproblems, then the annealer iteratively optimizes each subproblem to solve the overarching one. The program also enhances another algorithm via the same concept, permitting the use of larger subproblems, and more efficiently arriving at optimal solutions. Ohzeki said, "As the number of [quantum bits] mounted in the D-Wave quantum annealer increases, we will be able to obtain even better solutions."  ... '

Extended Reality is for Employees for Now

Based on what we have seen form Microsoft and other players, VR and XR seem to be best for things the require a hands-free concept,  which accept he oddness of wearing some unusual and expensive head gear.   Do expect some exceptions,  but it will take a while for consumer transitions.  Also expect early work out of China, which is promoting wider use,  to see some advances.  Agree with Forester's look below.   See the full piece at the link.

Extended Reality (XR) For Employees, Not For Customers
By Samuel Stern,  in Forester

Just over a year ago, I was talking with my colleague Jenny Wise after she had returned from Mobile World Congress, 2018 in Barcelona. While we were talking about some of the virtual and augmented reality applications she had seen demonstrated at the event, it sparked a realization for us that the opportunity for altered realities, whether that be VR, AR, XR or mixed reality (MR) was much more immediate with employees than with customers.

That led to our latest report, The Extended Reality Opportunity Today: Your Employees, where Jenny and I share the three main employee use cases for XR:

Enhanced training. XR enables more employees to have more practice time in low-risk, virtual environments. Applications span everything from Walmart preparing its employees for the once-a-year but critical black Friday to  surgeons mastering surgical procedures. My favorite example? The FLAIM Trainer, which allows firefighters to simulate risky scnearios and get closer to the real things with little risk – and less expense too. This happens through a combination of VR elements to simulate a fire, along with real-world equipment like hoses and even haptic gloves (a huge feature for the interaction design) that simulated kick back from extreme water pressure.  .... "  

Wal-Mart Buys Polymorph Labs

Wal-Mart continues to push tech.  Now for online, digital targeted ads.

Walmart acquires adtech startup Polymorph Labs to scale up its ad business  By Sarah Perez   @sarahintampa 

Amazon has a large and growing advertising business, but rival Walmart’s own ad business is much smaller. It’s now working to change that. Earlier this year, the retailer consolidated ad sales for its stores and websites and said it was ready to start monetizing its shopper data on a grander scale. Today, its efforts continue as Walmart says it has bought the advertising technology company Polymorph Labs to help it better compete via online digital ads, targeted using shopper data.

Terms of the deal were not disclosed. ... "