X

Learn about data lakes, machine learning & more innovations

How Machine Learning Solves Real-World Problems

Michael Chen
Senior Manager

Machine learning (ML) represents one of the keys to thriving in the modern age of big data. Consider “just how big” big data is for most organizations. Even small and medium-sized businesses (SMBs) deal with multiple data sources such as CRM tools, social media feeds, and transaction platforms. That's not even including internal data from HR platforms, submission forms, emails and media, and cloud-based tools that transmit metrics.

All of that data means that unprecedented levels of insight are possible, but only if an organization can get a handle on the massive amounts of relevant data it has access to. In some cases, sources are transmitting records every second of every day, meaning that the scope of processing all that data quickly becomes unmanageable from a manual processing perspective.

Get the latest on machine learning straight to your inbox. Subscribe to the Big Data Blog now!

However, ML makes all of that easier to digest by processing records, identifying patterns, handling queries, and automating workflows. Each of these benefits chips away at the manual burden of tasks, resulting in data-driven insights and more efficient organizations that allow staff to focus on their most critical tasks.

Real-World Machine Learning Examples

This may be easy to understand from a theoretical perspective, but how does it apply in the real world? In the new O'Reilly ebook Getting Started with Machine Learning in the Cloud, the potential of ML is explored. In particular, this ebook looks at real-world examples such as:

Netflix: When you finish watching a film on Netflix and a suggestion appears for your next viewing, that's based on a recommendation engine. Recommendation engines take a combination of individual user data (history of clicks, searches, and engagement) and macro applicable data (demographic choices, geographic location), then process that through ML algorithms to generate each recommendation.

Danske Bank: The financial services industry has made great use of ML for accelerating and refining fraud detection. Danske Bank in Denmark represented a successful path for replacing their fraud system with one powered by ML. This was driven by a change on the ground, as transactions shifted from primarily in-person events to digital (phone or online). The result was a significant increase in accuracy: false positives dropped by 60% and genuine alerts increased by 50%. This improved process led the security staff to better utilize their time and resources.

Donatos Pizza: This Ohio-based pizza chain is one of the Midwest’s most popular choices for pizza, having grown from its initial 1963 location to 160+ franchises today. As a test run for a new ML-powered retention initiative, customer data was analyzed to identify potential flags before losing a customer’s business. Under this platform, Donatos was able to create criteria for at-risk customers, and the resulting initiative increased retention of this group by 45%. This pilot program was so successful that the company made it required usage for all of its franchise locations.

AGR: A leading global provider of design, build, and maintenance of oil/gas production plants, AGR has hundreds of locations around the world, each filled with valuable equipment used for intricate processes. Any time something breaks, the ripple effect on production and workflows creates losses that can add up to hundreds of thousands of dollars. AGR wanted to get ahead of this by using emerging technology to implement a predictive maintenance plan. Thanks to ML, AGR now has a tool for predicting the lifespan of its equipment—and when it can schedule maintenance to maximize a piece of equipment's usability.

Texas Children's Hospital: This leading children's hospital saw an opportunity to use machine learning as a means to minimize risk for children with Type 1 diabetes. A complication called diabetic ketoacidosis (DKA) is life-threatening, but preventative measures can minimize the impact. Using machine learning on its database of patients with Type 1 diabetes, the hospital was able to decrease its DKA admissions by 30.9%.

Learn More about Integrating Machine Learning

As shown from the examples above, just about any business or organization can benefit from machine learning. If you’re not sure where to begin, the new ebook Getting Started with Machine Learning in the Cloud provides an excellent foundation of technology basics, deployment options, best practices, and further case studies. Download a free copy today to see how you can get the most out of cloud-based ML.

And for more about how you can benefit from Oracle Big Data, visit Oracle’s Big Data page—and don't forget to subscribe to the Oracle Big Data blog to get the latest posts sent to your inbox.

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.