Home » Artificial Intelligence » Mainstreaming Machine Learning, Part 2

Mainstreaming Machine Learning, Part 2

August 27, 2015


In the first part of this article, I discussed how business leaders’ eyes are just opening to the possibilities machine learning applications can have in the commercial world (e.g., activities as diverse as targeted marketing and route optimization). I also noted that as the Internet of Things matures and the amount of data needing to be analyzed increases, the usefulness of machine learning will only grow. In this article, I want to explore briefly some of the use cases for machine learning such as decision making, prediction, optimization, discovery, and fraud & abuse prevention. As Jorge Garcia observes, “Machine learning is escaping its containment from science labs to reach commercial and business applications.”[1] Garcia points out that machine learning is often discussed in two ways. First, machine learning associated with decision support management; and, second, machine learning associated with performance management and monitoring.


Decision Making


In Part 1 of this article, I explained how machine learning could be used to improve decision making in a business. Using a computer to make routine decisions allows a company to decrease human error and free human decision makers to focus on more important matters. Tribal knowledge (i.e., rules of thumb, exceptions, etc. that the best employees use when faced with non-routine decisions) can also be programmed into a computer to help the learning process and ensure that knowledge is not lost when employees are either no longer in the job or with the company. Garcia observes, “In some more advanced cases, decisions are made by the system with no human intervention, triggering the evolution of analytics systems, especially in areas such as decision management, and closing the gap between analytics and operations, which can mean boosting tighter relations between the operations, management, and strategy of an organization.”[2] As Bain analysts, Michael C. Mankins and Lori Sherer (), point out, if you can improve a company’s decision making, you can dramatically improve its bottom line.[3] “Not surprisingly,” they write, “companies that employ advanced analytics to improve decision making and execution have the results to show for it.”




As I noted in Part 1 of this article, Mike Yeomans, a post-doctoral fellow in the Department of Economics at Harvard University, asserts, “The most common application of machine learning tools is to make predictions.”[4] He provides a few examples of prediction problems in a business that can be addressed using machine learning. They are:


  • Making personalized recommendations for customers
  • Forecasting long-term customer loyalty
  • Anticipating the future performance of employees
  • Rating the credit risk of loan applicants


Yeomans explains, “These settings share some common features. For one, they are all complex environments, where the right decision might depend on a lot of variables (which means they require ‘wide’ data). They also have some outcome to validate the results of a prediction — like whether someone clicks on a recommended item, or whether a customer buys again. Finally, there is an important business decision to be made that requires an accurate prediction. … One of the easiest traps in machine learning is to confuse a prediction model with a causal model. Humans are hard-wired to think about how to change the environment to cause an effect. In prediction problems, however, causality isn’t a priority: instead we’re trying to optimize a decision that depends on a stable environment. In fact, the more stable an environment, the more useful a prediction model will be.”




Optimization is another common use for machine learning. It’s generally well-known that computer optimization is being used to help with the last-mile delivery challenge by assisting logistics companies with route planning. There are a number of activities, however, that could benefit from optimization. Garcia provides an example of how machine learning is being used to optimize energy usage.

NV Energy, the electricity utility in northern Nevada, is now using software from Big Data analytics company BuildingIQ for an energy-efficient pilot project using machine learning at their headquarters building in Las Vegas. The 270,000-square-foot building uses BuildingIQ to reduce energy consumption by using large sets of data such as weather forecasts, energy costs and tariffs, and other datasets within proprietary algorithms to continuously improve energy consumption for the building.”

Rebecca Merrett (@Rebecca_Merrett) reports that some researchers are attempting to marry game theory and machine learning to optimize situations involving competing interests.[5] Toby Walsh, an AI researcher at National ICT Australia, told Merrett that people often encounter situations where “there are multiple players coming together and they may behave selfishly. So how do we design mechanisms that, even if they are going to behave in selfish ways, … we get some optimal or good behaviors?” As you can see, the variety of ways that machine learning can be used to optimize activities is endless.




Merrett points out that advanced cognitive computer systems can use machine learning to assist in activities as complicated as drug discovery. IBM researcher, Ying Chen, told Merrett, “Our drug discovery process today is very time consuming and costly. It takes hundreds of millions of dollars to make one drug. And our failure rate is over 90 per cent still, partly because a lot of today’s diseases are very non-trivial … like cancer and multiple sclerosis, which are not very well understood. And the diseases themselves change. Once you make something, the disease adapts itself. So this process makes the discovery extremely difficult.” Computer scientists from Tufts University decided to let artificial intelligence (AI) have a go at solving the 120-year-old mystery of how sliced-up flatworms can transform themselves back into whole creatures — and it worked. Katie Collins (@katieecollins) reports, “For the first time ever a computer has managed to develop a new scientific theory using only its artificial intelligence, and with no help from human beings.”[6]


Fraud and Abuse Prevention


Fraud and abuse cost companies billions of dollars a year. Some of those losses can be mitigated using machine learning. As I wrote in a previous article, “Each of us benefits daily from predictive analytics in the area of fraud prevention. Financial services companies are always looking for suspicious activity that could indicate fraud is being perpetrated.” Spencer E. Ante reports that IBM is working on a new approach that involves “spotting threats before the crown jewels are stolen.”[7] It doesn’t take much imagination to appreciate how much money could be saved if fraud could be predicted before a loss actually occurs. The prospect is staggering.




Tim Negris, Vice President of Marketing and Sales at Yottamine Analytics, concludes, “Machine learning and the predictive modeling it enables is approaching the reach of most medium to large businesses thanks to the economics of cloud computing, advances in scalable software, and the growing availability and variety of big data.”[8] And as Yeomans noted, “Executives who want to get the most out of their companies’ data should understand what it is, what it can do, and what to watch out for when using it.” Most analysts believe we are witnessing the dawn of a new age in business analytics that will powered by actionable insights obtained using cognitive computing and machine learning technologies. Businesses that fail to use these technologies are likely to lose out to competitors that do use them.


[1] Jorge Garcia, “Machine Learning and Cognitive Systems: The Next Evolution of Enterprise Intelligence (Part I),” Wired Innovation Insights, 10 July 2014.
[2] Jorge Garcia, “Machine Learning and Cognitive Systems, Part 2: Big Data Analytics,” Wired Innovation Insights, 18 August 2014.
[3] Michael C. Mankins and Lori Sherer, “Creating value through advanced analytics,” Bain Brief, 11 February 2015.
[4] Rebecca Merrett, “Intelligent machines part 1: Big data, machine learning and the future,” CIO, 4 June 2015.
[5] Mike Yeomans, “What Every Manager Should Know About Machine Learning,” Harvard Business Review, 7 July 2015.
[6] Katie Collins, “Computer Independently Solves 120-Year-Old Biological Mystery,”Wired UK, 5 June 2015.
[7] Spencer E. Ante, “IBM’s New Cybersecurity Plan: Find Bad Guys Before They Steal,” The Wall Street Journal, 5 May 2014.
[8] Tim Negris, “Getting Ready for Machine Learning,” CITO Research, 2012.

Related Posts:

Full Logo


One of our team members will reach out shortly and we will help make your business brilliant!