This content is part of the Essential Guide: Special Report: Artificial intelligence apps come of age

Essential Guide

Browse Sections
Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Deep learning applications could have big impact on big data

Listen to this podcast

Unleashing deep learning algorithms on sets of big data can help organizations analyze text, images, videos and other forms of unstructured data. Learn why in this Talking Data podcast.

Deep learning may not be the hottest topic in advanced analytics right now -- you may have heard this already, but the broader categories of artificial intelligence and machine learning are both getting a lot of attention from analytics software vendors and users alike. But deep learning applications could become important components in the big data analytics toolkits of many organizations.

At the 2016 Deep Learning Summit in Boston, several speakers explained how the machine learning variant is particularly good at interpreting text, images and video. Designed to delve into complex analytical problems, deep learning software is at its strongest, they said, when it's applied to the kinds of unstructured data -- from sources such as social networks and customer service notes -- that various enterprises have been piling up and looking to analyze in search of useful business information.

Of course, if you think it sounds too good to be true that one analytics process could solve all of your big data analytics needs, you're right. While deep learning is powerful, it's also complicated. A number of open source tools, such as Google's TensorFlow software, can be leveraged to help streamline deep learning projects. But even with their assistance, developing deep learning applications can be heavy lifting for organizations that are still trying to get going on implementing predictive analytics programs. Better bring your computer science Ph.D. and your top data scientists.

Despite the still-prevalent technical challenges of machine learning in general, there's a lot of work being done to embed deep learning algorithms in all kinds of applications and technologies, according to the conference speakers. In this edition of the Talking Data podcast, we identify analytics tasks for which deep learning tools and techniques are a good fit and discuss potential uses in which mainstream enterprises are likely to find the greatest business value. Listen to the podcast to learn what you need to know about this emerging analytics and AI discipline and the combination of deep learning and big data.

Next Steps

Deep learning applications may help make AI more human

The future of AI and deep learning still involves a personal touch

Machine learning examples are making their way to enterprises