Site   Web

August 17, 2016

Understanding the Evolution of Machine Learning

To truly understand and take advantage of the opportunities that machine learning technology provides, the tech industry must first realize that machine learning as we know it is a result of a decade-long big data movement.

So how did we get to where we are today?

From the development of academic machine learning to the proliferation of commercialized data platforms, the evolution of big data and machine learning has created a perfect storm for the tech industry, one that allows for the creation of big data platforms that will forever change the way retailers and other large enterprise-based corporations interact with customers.These platforms allow for businesses to predict customer behavior with a new level of precision, which makes companies more efficient and profitable.

Three Waves of Machine Learning

The origins of machine intelligence can be broken down into three waves. Developers who understand how all three of these forces converge will be able to improve processes and replace legacy business software, such as CRM management and inventory optimization.

  1. The movement of traditional IT into a cloud-based infrastructure. The shift toward machine learning essentially began with the adoption of public elastic compute platforms, such as Amazon Web Services, Google Compute, Microsoft Joyent, IBM SoftLayer and Alibaba’s Aliyun. The result of these companies emerging was a trillion dollar industry fighting a war to get businesses to host everything to the cloud, which would ultimately allow them to cut costs, integrate disparate systems and operate more efficiently.
  2. Large amount of big data technologies emerge within the playing field. The “cloud war” eventually resulted in the proliferation of very tactically advanced platforms. These platforms, like Cassandra or SQuirrel, were open-source database companies designed to throw massive amounts of data into a new database and query it. However, they were highly technical and more suited for engineers or computer scientists.Because of the complex nature of these companies, large enterprise organizations such as banks, retailers or insurance companies were unable to move mission critical applications to the cloud. Most of these machine learning frameworks were simply too difficult for the average enterprise tech employee to use.
  3. Applied machine learning becomes commercialized and mainstream in the media. Triggered by Google’s acquisition of DeepMind, media and industry analysts began to pay attention to the tech industry’s race to develop the most innovative machine learning product. There was an outpour of media coverage and analysis of machine learning, and a wider audience began to listen to what computer scientists, such as Geoffrey Hinton of Google, had to say.As a result, enterprises are now demanding big data technology that the average tech employee can understand and use. We’re on the cusp of a wave of multi-billion dollar software companies that can embed all of the on-premise technology a retailer or financial organization needs to solve practical enterprise problems.

For example, a retailer will be able to use big data technology to merge systems like markdown software, marketing automation, business reporting software and price elasticity platforms into one platform. With the help of machine learning technology, that retailer can then use the data to make pricing and promotions decisions. The result? A more cost-effective and efficient way to drive sales and secure market share.

In this retail example, instead of using a gut reaction based on information from multiple disparate systems and competitors to determine pricing or promotions, retailers can use verified data to determine the most profitable route for promoting individual products to individual customers. Without an integrated data system, it’s nearly impossible for a retailer to make the most profitable decisions.

This concept can be applied to all types of enterprise companies originally built on ERP systems. However, it’s important to also remember the last phase of the evolution. IT leaders within enterprise-based corporations rarely possess the skills necessary to use machine learning platforms designed for the academic field. If a data platform is going to truly change the way a business interacts with customers, average IT employees must be able to understand how to navigate the platform with ease.

We are at a turning point in the big data movement. The tech industry has an opportunity to develop software platforms that can solve the biggest problems facing large enterprise companies today. And those that do so will certainly reap the rewards long-term.


Kerry Liu is the co-founder and CEO at Rubikloud, where he leads three important functions: people, sales, and technology disruption. Rubikloud helps retailers monetize their data to power personalized campaigns through the use of advanced machine learning techniques, effectively bridging the gap between marketing automation and retail management consultants. In his role, Kerry works to manage and maintain a thriving company culture that recruits the best and brightest in the industry, while also maintaining relationships with global retailers and investors. Kerry is passionate about machine learning and big data, and enjoys providing enterprise retailers with the tools and knowledge needed to enhance their overall business goals.