At Alacer, we’re interested in all facets of big data.  Lately, we’ve been fascinated by the broader tech community’s increasing chatter about machine learning.  It’s certainly a hot topic: enterprise personnel who previously yawned widely when they heard the words predictive analytics suddenly are intrigued by how a computer might become more intelligent over time and help solve business-related issues such as financial fraud and security.

The reality is, predictive analytics and machine learning are, in many ways, the same concept.  Machine learning centers around building computer programs that automatically improve a computer’s performance through experience.  Examples might include speech recognition systems that become more accurate over time based on the frequency of the computer’s interaction with the user, or a bank system’s analysis of potential fraudulent use patterns that can eventually differentiate between thieves and good customers.

Some analysts believe that machine learning technologies already outperform traditional analytics – the key is how a computer independently recognizes repetitive patterns and outcomes across a much wider dataset than what traditional data collection techniques can generate.

Fortune 100 companies such as Google and Amazon routinely utilize machine learning to help exploit data collection methodologies.  In fact, Amazon is heavily recruiting for machine learning scientists to join its teams in Seattle, Berlin and Bangalore.  The ML team develops algorithms that model patterns within data to drive automated decisions in ecommerce, web services and digital media.

So where are we heading?  Earlier this year, Microsoft sponsored a summit that assembled key researchers to talk about the dawning era of machine learning and the exciting developments that are causing scientists to believe that some of the greatest computer science-related problems may be solved in the very near future.  As an example, quantum computers with machine learning could dramatically speed up tasks that have previously eluded scientists – such as mapping the universe.

Big data and machine learning are facilitating a brave new world…what’s your take on the topic?