The Fantastic World of Deep Learning
Jan 21, 2015, by Corey Padveen
I recently wrote an article about some emerging technologies marketers need to know about and I thought I would go into some detail about a particular field that is of great significance to all of these emerging industries: deep learning.
Deep learning really is as cool as the name suggests. Essentially, deep learning is a field of machine learning whereby series of complex algorithms develop systematic models out of non-linear actions. So, what in the world does that mean, and why is it so important for digital and social marketers?
Understanding Deep LearningExpanding on the complicated description I laid out above, deep learning is all about understanding patterns where there are none - at least none that can be seen on the surface. When it comes to deep learning, the machine looks for subtle hints and cues that allow it to refine its process. An example of basic machine learning would be something like talk-to-text, or intuitive autocorrect. Where things get exciting, is when deep learning starts to create models out of behaviors.
Some of the more exciting (and very real) fields of deep learning include work in the areas of computer vision, natural language processing (a field that I, in particular, am very interested in) and deep neural networks. Take, for instance, the app from startup Word Lens (above), or the latest update to the Google Translate app. Now, instead of taking a picture and waiting for the text to translate, or typing in your own text, you can simply hold your camera up to the foreign language and see the translation in its place. Incredible! That is an example of these fields in practice, and for some of them, there is a great degree of importance for marketers.
Why Should Marketers Care?Take the example I noted above about natural language processing. As digital and social marketers, almost everything we do revolves around conversations. Whether those are one-on-one conversations with a prospect sharing marketing materials, a brand advocate in the planning stages of a campaign, or a disgruntled client on social media, we are constantly talking to one or many. As markets expand thanks to the increasingly overarching reach of social media, how can we ever be certain that we are communicating with every interested party?
Of course, there are plenty of listening tools on the market. When we want to track conversations around our brand or the industry in general, we have our veritable pick of the litter. There is radian6, Brandwatch, Meltwater, Tracx, Mention and dozens more, and that is just for listening and monitoring. (And, yes, of course, these tools all have a plethora of other capabilities, as well.) But how can we be sure that the information we are getting is at all valuable? After all, we’ve seen that the rules of grammar and clarity sort of go out the window when it comes to social conversations. That is why natural language processing is so important.
As a marketer, with the use of a platform that integrates a proven and intelligent natural language processor (by intelligent, I mean one that is capable of adapting to my preferences) - my job becomes significantly easier.
On a larger scale, when the tools that I use become intuitive and begin to do some of the work for me, again, my job becomes significantly easier. This is why deep learning is such a fascinating space and one that should be exciting marketers.
ConclusionIt might seem like the applications are for the privileged big-budgeters now, but it won’t be long (as with all things in this space) that much of these capabilities are scaled so that they can be offered to virtually anyone. Just a few years ago, the idea of tapping into social conversations and extrapolating vast data sets seemed like a pipedream achievable only through the use of bottomless resource wells. Now, businesses of virtually any size have access to a number of tools.
As this space continues to expand and the interest in it grows, its availability will experience the same trend. And as marketers familiarize themselves with the capabilities that this fields can present, it shouldn’t be long before that happens.