Last few months were quite eventful at Google. Of course, we are not going to write another highly introspective research paper on its two major algo updates code named – Panda and Penguin. Nope, this time around we are going to take a deeper look at what is happening at Google and what is meant for the search communities.
Lately Google has acquired few companies and strangely, these acquired companies have got little to do its main search business. Take for example – Deepmind – it is a UK based artificial intelligence startup or Boston Dynamics – it is a Massachusetts-based manufacturer of military robots. What the heck Google is trying to do by acquiring these companies that have almost nothing to do with its search business.
The industry, as always, has been rife with speculations and as the experts are of the opinion that Google is trying to develop a real life counterpart of ‘Skynet’ – or Artificial Intelligence.  AI, though maybe termed as another , could redefine the concept of search altogether. Search could be more intuitive, more personal and more humane if Google manages to infuse a fraction of the power of Artificial Intelligence.
Do not Expect a Skynet Though
Okay, you can be a huge fan of Terminator series but sorry to say, there is almost zero possibility of a Skynet going to take revenge on the humanity. At least, not in our life time at least. Before making any kind of presumption, you have to realize the fact that we are not heading towards a world governed by artificial intelligence. But, the writing on the wall is that, Google is capable of developing an algorithm that can analyze patterns and by doing so, it would be able to make predictions and thereby would be able to anticipate the needs of the searchers.
It is Already Here
Believe it or not, machine learning is already here and the chances are quite high that it has helped you more than once. When you visit YouTube and play a song, you might have noticed that YouTube’s recommended videos match almost your listening habits. Some other cool real life examples of machine learning algorithm are –
Pushing up local businesses based on your geographical location in the SERP.
Generating Picture name by utilizing the mobile platform.
Calculating Distance between two places.
Showing you the weather and traffic condition based on the GPS.
How Machine Learning Fits into Search Engine Optimization
Google has already made a few baby steps towards making a search world without relying heavily on the links. As links are easy to manipulate, some experts like Dr. Peter of Moz believe that Google is gravitating towards more complex metrics like Bounce rate, user interactions and more such things which are hard to manipulate.
Say for example, [spoiler alert – this is just a presumption] Google can use machine learning to develop an algo that can detect whether a web page deserves to be on the top by using the aforementioned metrics. People like – Jay Soriano [marketing head at SorianoMedia] has been predicting a future where links will be playing less and less effective roles. In his words – using spammy tactic in this age of machine learning will be ‘the equivalent of driving a wrecking ball through your physical storefront.’
Whether a website is involved in purchasing links can also be determined by utilizing machine learning algo to some extent. It can look at other links and if they look like paid links, chances are quite high that the link in question is also paid. Therefore, it will be good if search engine can discard the value of the link juice passing through that web page.
These are some examples of practical use of machine learning and if it is integrated with the mail search algo, it may redefine the concept of SEO.
