Algorithm Helping Demystifying Networks

Interesting, but don't understand it directly.  But thinking it. Boolean models are easy.

Algorithm May Help Scientists Demystify Complex Networks  By Penn State News,  July 21, 2021

( note the PSU article does it better)

A new algorithm capable of analyzing models of biological systems can lead to greater understanding of their underlying decision-making mechanisms, with implications for studying how complex behaviors are rooted in relatively simple actions.

Pennsylvania State University (Penn State)'s Jordan Rozum said the modeling framework includes Boolean networks.

Said Penn State's Reka Albert, "Boolean models describe how information propagates through the network," and the nodes' on/off states eventually slip into repeating patterns that correspond to the system's stable long-term behaviors.

Complexity can scale up dramatically as the system incorporates more nodes, particularly when events in the system are asynchronous. The researchers used parity and time-reversal transformations to boost the efficiency of the Boolean network analysis.

Full PSU article.

Forecasting Ride Demand

Forecasting any kind of demand by consumers is always of interest, so this likely has applications beyond just ride ride prediction.

Hail Technology: Deep Learning May Help Predict When People Need Rides 

Penn State News   By Matt Swayne

Pennsylvania State University (Penn State) researchers analyzing a large dataset of ride requests to Didi Chuxing, a Chinese car-hailing company, found computers may be better at forecasting demand for taxi and ride-sharing services. The team used two types of neural networks to extract patterns of taxi demand, and then to predict the demand patterns with significantly better accuracy than current technology, explains Penn State's Huaxiu Yao. When users need a ride they make a request via a computer application, and the researchers think tapping these requests, instead of relying on ride data only, reflects overall demand better. With the historical data, which includes the request's time and location, the computer could anticipate how demand will change over time, and the researchers were able to visualize how that demand evolved by plotting it on a map. "Basically, we used a very complicated neural net to simulate how people digest information, in this case, the image of the traffic patterns," notes Penn State professor Jessie Li.  ... " 

Identifiying Plant Disease

Out of the fall garden and back to the possibilities of pattern recognition.  Back to my alternate botanic tech universe.  Penn State work using deep learning to identifying plant disease.  Great example:

 Phone Powered AI Spots Sock Plants with Remarkable Accuracy,   by Matt Simon, Science

Researchers at Pennsylvania State University (PSU) say they have designed a smartphone-based neural-network program that can automatically identify diseases in the cassava plant with near-flawless accuracy. The network is based on Google's open source TensorFlow machine-learning library, and Google's Pete Warden notes the TensorFlow Mobile app requires only about 25 million parameters, versus the hundreds of millions some networks need. "It only requires about 11 billion floating point operations to actually calculate its result, and some other networks require hundreds of billions of operations to do a similar job," Warden notes. He also says thanks to transfer learning, the network was trained to recognize cassava leaves on much less data. "It really comes down to the data, because garbage in, garbage out," says PSU's Amanda Ramcharan. She believes the growing affordability of smartphones and the continuing simplification of algorithms will combine to make such tools more widely available to farmers. ... " 

Thoughts on Data Engineering

All analytics is data engineering.  Manipulating data content in context to achieve useful results, using a proven process.

A Data Engineer's Guide To Non-Traditional Data Storage    by Irina Papuc:

Data Engineering
With the rise of  and data science, many engineering roles are being challenged and expanded. ... 

Originally, the purpose of data engineering was the loading of external data sources and the designing of databases (designing and developing pipelines to collect, manipulate, store, and analyze data).
It has since grown to support the volume and complexity of big data. So data engineering now encapsulates a wide range of skills, from web-crawling, data cleansing, distributed computing, and data storage and retrieval.

For data engineering and data engineers, data storage and retrieval is the critical component of the pipeline together with how the data can be used and analyzed. ... "   

Algorithmic Retailing

Gartner on Algorithmic retailing.   Pretty broad view.   I assume this is for particular parts of the business where process can be defined?   I don't have a Gartner subscription, so can't see the full report.  (now have the full report and reviewing)

" .... New Gartner research explores how retailers gain competitive advantage through the application of algorithms that reduce costs and grow top-line revenue. CIOs can use this research to identify use cases that will improve business performance in the unified commerce retail marketplace.

Gartner describes algorithmic business as the “enablement of business value through the action of algorithms on data” and regards algorithms themselves as a way to encapsulate and produce intellectual property, knowledge and insight in a reusable form. Algorithms are a set of rules for solving a problem in a finite number of steps, as for finding the greatest common divisor. New technologies create opportunities to advance algorithms, incorporating many more data inputs and steps and even decision-making capability.  ... " 

Building a Reco Engine with a Graph Database

Building a recommendation Engine ... Powering recommendations with a Graph Database.  I like the idea of decision process starting with a visualization.   " You May also like …" ... Product recommendations help businesses maximize their online revenue. It requires advanced technology, but this is now available off-the-shelf, and is already being used by Wal-Mart and other market leaders.

“You may also like” is a deceptively simple phrase that encapsulates a new era in customer relationship management. In offering the advice, businesses maximize the value they deliver by providing highly targeted, real-time product recommendations to their online consumers.

How to Moments on YouTube

This came to mind this past week as I searched for several very different how-to needs online.  I some cases I really needed a visual explanation of the technique, rather than a textual description.  An example of tacit knowledge.  Its not all on YouTube, but much is there.  As in all online resources you look for credibility, objectivity, and clear and complete visual explanation. Often too, how the task description is encapsulated in larger needs.   How then are such microtasks combined to form Jobs?    In Think With Google this week, satistics about YouTube use for the How-to moment.   And how that can link to marketing.

" ..... Do you know how build a treehouse? Cook the perfect pancake? Do a french braid? Neither do I. But life (and 3-year-old daughters) sometimes require us to do all of this and more in a given day. Learning how to do these sorts of things used to be time-consuming. We might've combed through an instruction manual, tracked down an expert, or called up good ole mom and dad—or worse, customer service. Today, we can find out instantly. We simply grab the nearest device and search—and not just on Google. Searches related to "how to" on YouTube are growing 70% year over year,1 and more than 100M hours of how-to content have been watched in North America so far this year   ... " 

Computational Thinking With Wolfram Language

An excellent transcript (with pictures) of the talk and detailed demonstration given by Stephen Wolfram at the recent SXSW on their Wolfram Language.    We looked at Wolfram Alpha when it was released, but I have not looked at their generalized computation language for some time.   Looks impressive, deserves a look.  

" ... From an intellectual point of view, the goal of the Wolfram Language is basically to express as much as possible computationally—to provide a very broad way to encapsulate computation and knowledge, and to automate, as much as possible, what can be done with them. ... " .  

All that being said, as a practitioner and leader of computational competencies in the enterprise, who interviewed many students, I saw relatively few students trained in Wolfram methods.  What we saw were typically those trained in the languages of the big computer software companies, or were open source methods now being used in schools,  like R or Python.   Wolfram seems to make a good case that their methods are better, now how can they show that to schools?   Is it better enough to make the case?

Perhaps better for generalized problem solving courses, but such courses are not common, rather than programming or machine learning courses.  This may well be the fault of CS curricula today, companies like mine now do little low-level programming, but need to do lots of problem solving.

Check out the examples in the demonstration above, these are things that would be very difficult to do readily with general purpose programming, or R or Python.

Shopping Through the Lens of IT

In the ACM    " ... Pennsylvania State University (PSU) researchers are studying how smart cameras could eventually guide visually impaired shoppers to find the items they need.

The researchers are using a $10-million U.S. National Science Foundation grant to replicate the human vision system using information technology as part of the Visual Cortex on Silicon project. "This project brings together the strengths and efforts of technical leaders in multiple disciplines," says PSU professor Vijaykrishnan Narayanan. ... " 

Butterflies for IT Management

Another old idea come around again.  Can you use it for IT management?  Perhaps in a theoretical sense, but I am skeptical about its predictive power.  " ...  So far so good — but is it enough? Do we have enough of theoretical engine power to encapsulate the intricacies, complexities and unpredictability of a) software development and, perhaps more crucially b) those human beings who will work to produce it in real world humanoid scenarios. The answer is no, we need chaos theory too. ... " ... Chaos theory: the behaviour of dynamic systems is highly sensitive to initial conditions ...  . 

Information Insertion at Point of Need

From the BBC: What a good ridiculously simple idea, a sticker applied to Mercedes vehicles that contains a QR code, which directs first responders to rescue methods to extract people from a specific car model.  They have put the idea into the public domain.

Reminds me of related ideas of tattooing bar codes on people for medical ID .... Not far different from RFID capsule injection, now common in veterinary applications and some human systems ...

Foodspotting What You Eat

FoodSpotting: Intriguing idea for Foodies like me.  You take a picture of a plate of food on an IPhone App where you are and add a capsule review.  I have known a number of people that do this informally in Facebook and elsewhere.  It can be entertaining.  The posted pictures are apparently cleared by the company.  They expectedly range from very good to very poor snapshots of what you may be eating.  Actual reviews are usually curt and not very useful.   Lots of fast food spots are included and since I usually know what to expect there, that is not too useful.   Agree completely that it is most useful as a 'celebration of local food',  but the fast food chains crowd that out. If there was a way to focus this a bit more. The interface is very simple, yet it is still too hard to do certain kind of things.  Most useful on mobile.  See their site.

" .... If you haven't tried the food photo social network app Foodspotting, you're missing out. The company announced today that it has raised $3 million in venture capital to fund further international expansion and to build what it calls "Pandora for food." ... I just started using Foodspotting a few weeks ago and have fallen in love with it. From discovering places to eat to seeing what a menu looks like on a plate before I order to reading it like a hyper-local foodie newspaper, there are lots of different ways to use the app. The six-person company will celebrate its one-year anniversary this weekend and has a compelling vision for the future ... '

Earth Project Seeks to Simulate Everything

From the BBC:  Quite a huge undertaking.  Including always problematic aspects of human behavior ... " ... Nicknamed the Living Earth Simulator, the project aims to advance the scientific understanding of what is taking place on the planet, encapsulating the human actions that shape societies and the environmental forces that define the physical world. "Many problems we have today—including social and economic instabilities, wars, disease spreading—are related to human behavior, but there is apparently a serious lack of understanding regarding how society and the economy work," says Dr Helbing, of the Swiss Federal Institute of Technology, who chairs the FuturICT project which aims to create the simulator.... "

Image Group at Penn State

I mentioned the ALIPR image tagging project before. Have had the opportunity to take a further look at James Z. Wang's image group at PSU: " ... Research interests of this group include automatic image tagging / learning-based linguistic indexing of images, semantics-sensitive image retrieval, image security, biomedical informatics, computational aesthetics, story picturing, art image analysis and retrieval, and computer vision ... " . Impressive directions. The site, linked to above, is particularly rich in relevant detail and links.