Tag Archive for Machine Learning

Our 5 Rules of Data Science

In manufacturing, the better the raw materials, the better the product. The same goes for data science, where a team cannot be effective unless the raw materials of data science are available to them. In this realm, data is the raw material which produces a prediction. However, raw materials alone are not sufficient. Business people who oversee machine learning teams must demand that best practices be applied, otherwise investments in machine learning will produce dubious business results. These best practices can be summarized into our five rules of data science.

For the purpose of illustration, let’s assume the data science problem our team is working on is related to the predictive maintenance of equipment on a manufacturing floor. Our team is working on helping the firm predict equipment failure, so that operations can replace the equipment before it impacts the manufacturing process.

Our 5 Rules of Data Science

1. Have a Sharp Question

A sharp question is specific and unambiguous. Computers do not appreciate nuance. They are not able to classify events into yes/no buckets if the question is: “Is Component X ready to fail?” Nor does the question need to concern itself with causes. Computers do not ask why – they calculate probability based upon correlation. “Will component X overheat?” is a question posed by a human who believes that heat contributes to equipment failure. A better question is: “Will component X fail in the next 30 minutes?”

2. Measure at the Right Level

Supervised learning requires real examples from which a computer can learn. The data you use to produce a successful machine learning model must demonstrate cases where failure has occurred. It must also demonstrate examples where equipment continues to operate smoothly. We must be able to unambiguously identify events that were failure events, otherwise, we will not be able to train the machine learning model to classify data correctly.

3. Make Sure Your Data is Accurate

Did a failure really occur? If not, the machine learning model will not produce accurate results. Computers are naïve – they believe what we tell them. Data science teams should be more skeptical, particularly when they believe they have made a breakthrough discovery after months of false starts. Data science leaders should avoid getting caught up in the irrational exuberance of a model that appears to provide new insight. Like any scientific endeavor, test your assumptions, beginning with the accuracy and reliability of the observations you started with to create the model.

4. Make Sure Your Data is Connected

The data used to train your model may be anonymized, because factors that correlate closely to machine failure are measurements, not identifiers. However, once the model is ready to be used, the new data must be connected to the real world – otherwise, you will not be able to take action. If you have no central authoritative record of “things”, you may need to develop a master data management solution before your Internet of Things with predictive maintenance machine learning can yield value. Also, your response to a prediction should be connected. Once a prediction of failure has been obtained, management should already know what needs to happen – use insights to take swift action.

5. Make Sure You Have Enough Data

The accuracy of predictions improve with more data. Make sure you have sufficient examples of both positive and negative outcomes, otherwise it will be difficult to be certain that you are truly gaining information from the exercise.

The benefits of predictive maintenance, and other applications of machine learning, are being embraced by businesses everywhere. For some, the process may appear a bit mysterious, but it needn’t be. The goal is to create a model which, when fed real-life data, improves the decision making of the humans involved in the process. To achieve this, data science teams need the right data and the right business problem to solve. Management should work to ensure that these five questions are answered to their satisfaction before investing in data science activities.

Not sure if you have the right raw materials? Talk to BlumShapiro Consulting about your machine learning ambitions. Our technology team is building next generation predictive analytics solutions that connect to the Internet of Things. We are helping our clients along each step of their digital transformation journey.

Berry_Brian-240About Brian: Brian Berry leads the Microsoft Business Intelligence and Data Analytics practice at BlumShapiro. He has over 15 years of experience with information technology (IT), software design and consulting. Brian specializes in identifying business intelligence (BI) and data management solutions for upper mid-market manufacturing, distribution and retail firms in New England. He focuses on technologies which drive value in analytics: data integration, self-service BI, cloud computing and predictive analytics

A Digital Transformation – From the Printing Press to Modern Data Reporting

Imagine producing, marketing and selling a product that has only a four-hour shelf life! After four hours, your product is no longer of much value or relevance to your primary consumer. After eight hours, you would be lucky to sell any of the day’s remaining stock. Within 24 hours, nobody is going to buy it; you have to start fresh the next morning. There is such a product line being produced, sold and consumed to millions of people around the world every day. And it’s probably more common than you think.

It’s the daily newspaper.

With such a tight production schedule, news printers have always been under the gun to be able to take the latest news stories and turn them into a finished printed product quickly. Mechanization and automation have pretty much made the production of the modern daily paper a non-event, but it has not always been that way.

150 years ago, the typesetter (someone who set your words, or ‘type’, into a printing press) was the key to getting your printed paper mass produced. With typesetters working faster than your competitors, you could get your product, your story, out to your consumers faster, gaining market share. However, it was still very much a manual process. In the late 1800s the stage was set for a faster method of setting type. One such machine, the Paige Compositor, was as big as a mini-van and had about 15,000 moving parts. (Samuel Clemens, a.k.a. Mark Twain, invested hundreds of thousands of dollars in the failed invention, leading to his financial ruin.) On a more personal scale and at the modern end of the spectrum, we think nothing of sending our finished work, perhaps the big annual report, off to the color printer or ‘office machine’, or upload it to a local printing vendor who will print, collate and bind the whole job for us in a fraction of the time it would take a typesetter to layout even the first page!

So why am I telling you all this? It’s certainly not for a history lesson. The point is that the printed news industry went through a transformation from nothing (monks with quill pens), to ‘mechanization’ (Gutenberg’s printing press), to ‘automation and finally to ‘digitalization.’ And, they had to do so as the news consumer evolved from wanting their printed subscription on a monthly basis, down to the weekly, to the daily and even to the ‘morning’ and ‘evening’ editions. Remember, after four hours, the product is going stale and just about useless. (We could debate whether the faster technologies was what drove news consumers to want information faster, or if the needs of the consumer inspired the advancements in technology, but we won’t.)

Data and reporting has followed the same phases of transformation, albeit not along a much accelerated time span. The modern data consumer is no longer satisfied with having to request a green-bar, tractor fed report from the mainframe, then wait overnight for the ‘job’ to get scheduled and run. They’re not even satisfied with receiving a morning email report with yesterday’s data, or even being able to get the latest analytics report from the server farm on demand. No, they want it now, they want it in hand (smart phones), and they want it concise and relevant. To satisfy this market, products are popping up that fill this need in today’s data reporting market. Products like Microsoft’s Power BI can deliver data quickly and efficiently and in the mobile format demanded due to the industry’s transformation to digital processing. Technologies in Microsoft’s Azure cloud services such as Stream Analytics, coupled with Big Data processing, Machine Learning and Event Hubs have the capabilities to push data in real time to Power BI. I’ll never forget the feeling of elation I had upon completing a simple real-time Azure solution that streamed data every few seconds from a portable temperature sensor in my hand to a Power BI Dashboard. It must have been something like Johannes Gutenberg felt after that first page rolled off his printing press.

Gutenberg and Clemens would be amazed at the printing technology available today to the everyday consumer, yet we seem to take it for granted. Having gone through some of the transformation phases with regard to information delivery myself (yes, I do in fact recall 11×17 green-bar tractor-fed reports) I tend to be amazed at what technologies are being developed these days. Eighteen months ago (an eon in technology life) the Apple watch and Power BI teamed up to deliver KPI’s right on the watch! What will we have in another eighteen months? I can’t wait to find out.

About Todd: Todd Chittenden started his programming and reporting career with industrial maintenance applications in the late 1990’s. When SQL Server 2005 was introduced, he quickly became certified in Microsoft’s latest RDBMS technology and has added certifications over the years. He currently holds an MCSE in Business Intelligence. He has applied his knowledge of relational databases, data warehouses, business intelligence and analytics to a variety of projects for BlumShapiro since 2011. 

Power BI Demo CTA

Internet of Things Modern Application Development

Over the past decade modern application development has shifted from mainframe computing to personal PCs, and now to smartphones and cloud services. These shifts required new software languages, new hardware and new application development solutions. The best illustration of this shift came with the 90’s “Internet Boom.”  This shift resulted in application development on mainframe and personal PCs to applications that can run on a web browser. The shift also included new tools such as Visual Studio, new languages like HTML/JavaScript, new Architecture Patterns such as MVC and new application life cycle processes like Agile/Scrum.

Then came the smart phone. This shift from personal PCs to small mobile devices such as iPhones forced modern application development to support multiple screen resolutions, and a need to operate off-line while remaining connected to cloud services. Our next shift is to the Internet of Things (IoT), once again giving a new meaning to modern application development. Now, applications need to be developed to run on different types of devices like thermostats, doorbells and small Bluetooth sensors. The application must be secure, cloud ready and able to perform predictive analysis using machine learning. Below are my thoughts on this latest shift in modern application development:

Devices

The IoT modern application development shift includes a multitude of devices that range from televisions to cameras, to refrigerators, to pretty much any device that is powered into an outlet. One of the more notable products in this space is the Amazon Echo which uses voice recognition as its main interface, and can provide control over your light switches, thermostats and even your music collection. Amazon Echo is an example of an IoT device which breaks away from the previous modern application development, as it uses voice as its interface, is always connected to the cloud, and can connect with other IoT devices. This changes everything about how we think of modern application development. No longer is it about supporting multiple device resolutions, but rather about what data can be captured via the latest IoT devices and how that data can be used to improve our lives. This means we need new software tools, new cloud services, new analysis software and new machine learning algorithms.

These applications do not always include fancy user interfaces, as they are often function specific. For example, an IoT device could capture changes in temperature on a farm, take soil sample readings or even capture images and video of the fields. This data can then be sent to cloud services where it can be analyzed and run through machine learning to produce an easy to understand update on the farm. The data from the disparate “things” needs to be collected in a common format for actionable insights. Of note, most of the “big data” being processed and collected today is machine-to-machine. Cloud services help to aggregate and display this data in ways humans can understand, analyze and take action on the insights delivered.

Cloud Services

Cloud services are at the heart of IoT. Devices are built to perform a simple purpose and leave all complex user interfacing, analysis and thinking to the cloud. Cloud services such as the Azure IoT hub provide both the software tooling and service for a device to talk to the cloud and the device to connect to other devices. For example, in the manufacturing industry, IoT devices using the Azure IoT hub can be developed to monitor the production line and equipment use, which is then submitted to cloud service which then can be interpreted by human intelligence to predict equipment maintenance.

With this shift to IoT modern application development software is developed to capture data from a range of sensors, submit that data to cloud services and then process that data using analytics services such as Business Intelligence dashboards for timely and relevant role based information.

Machine Learning

So what is the point of these IoT devices in our homes, cars and at work, capturing data and sending it to the cloud? Well that’s what machine learning is all about. We now need to develop algorithms that can learn based on data from the IoT. For example: home IoT devices using machine learning will learn the normal patterns in your house and only notify you when there is a disruption such as the lights staying on past a normal pattern or when you leave your windows open while you are away. Machine learning is one of the most important aspects of IoT and without it, all we would have is raw data in a cloud service with no meaningful way to utilize it.

BlumShapiro Consulting is a Microsoft Advanced Analytics partner, with experience building modern IoT apps. 

Technology Talks Newsletter CTA

About Hector: 

hectorHector Luciano, Jr. is a Consulting Manager at BlumShapiro, a Microsoft Gold Partner focusing on SharePoint, Office 365, mobile technologies and custom development solutions. Hector is an active member of the SharePoint community. His experience reflects a breadth of Microsoft .Net Technologies experience. With a focus on Software Application development, Hector has worked on various projects including architected and designed solutions for web, client/server and mobile platforms. He has worked closely with business owners to understand the business process, then design and build custom solution. Hector currently holds Microsoft Certified Solution Developer (MCSD), Microsoft Certified Professional Developer (MCPD).

 

 

Applied Machine Learning: Optimizing Patient Care in Hospitals, Profitably

Why are so many industries exploring Machine Learning as a means of delivering innovation and value?  In my view, the technology speaks to a primitive urge – machine learning is like having a crystal ball, telling you what will happen next.  For a business, it can convey information about a customer before they introduce themselves.  On a personal level, when I consider what I would like to have information about in advance, the first thing that springs to mind is obvious: my health.  Am I about to get sick?  How can I improve my wellness and overall health?  If you are wearing a Fit-Bit right now, then you probably agree with me.

In my last blog, I shared some real-world examples for how the Hospitality Industry applies Machine Learning.  What about Health Care and Hospitals?  While hospitals have similar challenges, in that they accommodate guests who stay overnight, the objectives in health care are quite different, and changing rapidly.   The Affordable Care Act is driving new business models, incentivizing outcome based reimbursement as opposed to volume based reimbursement.  Unlike hotels, today’s hospitals are interested in ensuring their guests do not have to return, at least not in the short term.   They also need to manage costs in a way they have not been incentivized to do in the past.  Hospitals across the country are considering how predictive analytics can have a  meaningful impact on operations, leading to improved patient health and improving the bottom line.

The cost savings opportunity for health care providers is startling. Here are just three examples:

Reducing Hospital Readmissions – in 2014, Medicare fined 2,610 hospitals $428 million for having high hospital readmission rates. Leaving actual fines aside, industry analysts estimate that the overall cost of preventable readmissions approaches $25 Billion annually. As a result, hospital systems all over the nation are mobilizing to intervene, using ML to identify risk factors which are highly predictive of readmission. Carolinas Healthcare System, partnering with Microsoft, did just that. Using data from 200,000 patient-discharge records, they created a predictive model deliver customized discharge planning, saving the hospital system hundreds of thousands of dollars annually. Read the article in Healthcare IT News.

Clinical Variation Management – Mercy Hospital is partnering with Ayasdi to find the optimal care path for common surgical procedures. Using knee replacement as an example, the Clinical Variation Management software helps hospital administrators find clusters of patient outcomes, then enables the exploration of those clusters in order to correlate a metric (i.e. Length of Stay) with a certain regiment or activity. Watch this video to learn how Mercy Hospital saved $50 million by applying Machine Learning to an extremely common procedure.

Improving Population Health – Dartmouth Hitchcock, a healthcare network affiliated with Dartmouth University, is piloting a remote monitoring system for patients requiring chronic care. 6,000+ patients are permitting the hospital to collect biometric data (i.e. blood pressure, temperature, etc.) in order that nurses and health coaches can monitor their vital signs, and machines can predict good days and bad days. Quite the opposite of hospitality: Dartmouth Hitchcock is trying to keep the guests from needing to checking in! Read more about the Case Study from Microsoft.

Are machines taking over for physicians? No. The Patient – Physician relationship remains (and I think will always remain) central to the delivery of personal health care. However, it seems clear that ACA is providing significant rewards to health care providers who manage population risk better. Machines can help here: through data, machine learning can find risk “hiding in the data”.

Contact Blum Shapiro Consulting to learn more about how Azure Machine Learning can curtail hospital readmissions, identify variations in common clinical procedures and improve patient population health.