Data mining: Dr Mike Lynch says the next IT revolution is happening in the “I”, the information, not the “T”
In his 1950 paper entitled Computing Machinery and Intelligence, computer scientist Alan Turing opens with the words: “I propose to consider the question, ‘Can machines think?'”
Sixty years on, the idea of intelligent computers seems a little less ridiculous.
Technology leaps forward year by year with ever greater processing powers, bigger clouds, faster internet connections, sleeker interfaces and cleverer self-learning algorithms.
It poses the question: are computers getting closer to human intelligence by being able to make value judgements, understand concepts and process a world that is not just black and white in real time?
The answer is yes.
Not only is this a significant step forward, but it also has the potential to fundamentally change the world we know.
If technology can give data meaning, whether it is voice, video, text or images, for instance, think of the prospect of using this insight to create a more predictable, and consequently less volatile, world.
For CEOs, using technology to spot the “unknown unknowns” during this time of economic uncertainty will be regarded as a “superpower”, helping them to make better business decisions and ultimately beat the competition.
Now think of the potential of applying this approach on a macro level to other forms of information, such as medical or seismograph data.
The reason why understanding meaning in information has generated a lot of excitement is because of its enormous potential.
Today, thanks to technology’s mass appeal and accessibility, on a daily basis we collectively produce 2.5 quintillion bytes of data, and the growth rate is so high that 90% of all information ever created was produced in the last two years alone. The value of this information to organisations who want to keep ahead of the curve is huge.
In the business world, information has traditionally been housed in databases.
This has been an effective way to keep information safe and stored in a uniform fashion. However, the database was created as a way to work around the limitations computers had half a century ago.
As the majority of information flowing through organisations today is unstructured or ‘human information’ (90%), such as text, email, video, and audio, and therefore impossible to box into traditional databases, being able to manage that information so you know where it is and process it so you know what it means is essential.
For instance, you could store customer-service call centre recordings into a database if you really had to, but you would have no idea whether the customer was happy or not. This is exactly the type of insight businesses want to know after all.
To understand them, organisations have to take a new approach to managing and uncovering the meaning of their data. The next-age information platforms make sense of, and understand the meaning of, information and use it to solve problems, in order to give businesses a better understanding of themselves, their customers and the competition.
This is what has the analyst community so excited, due to the sheer size of the market opportunity.
To see just how human information can add value, let’s think of a market for an online retailer.
The availability of vast quantities of data on current or potential customers on e-commerce sites offers online retailers huge value – if they can gather and analyse this information efficiently and in a timely manner.
Customer browsing and purchasing habits create sizeable data trails and, coupled with social-media integration, can allow content to be effectively targeted.
However, it has to be noted that businesses can only be effective if decisions can be made in real time.
Business data is of little use if companies are not quick enough to act on it, which is often the case.
Organisations that truly want to understand their customers and keep up with changing demands need to change their organisation culturally in order to do this successfully.
The challenge is that having tens or hundreds of thousands of customers, each with potentially thousands of data points connected with them, results in a very large amount of data being collected.
Part of this might be structured data that can be easily categorised, for example, gender, age, and geography, but the majority will be unstructured, human information.
The content and sentiment of a product review, for instance, is not something that can be slotted neatly into a standard database, but the ability to tap into information like this has huge implications for driving sales or safeguarding a reputation.
Unlocking real value
The key is to understand that data – leaving it exactly where it is – in order to create an infinitely scalable platform, and a powerful basis for analysis and action throughout the entire enterprise.
By applying meaning, we can cut through the gordian knot of trying to find that “perfect” database, and get to the heart of the issue – being able to process 100% of the information, structured and unstructured, to unlock real business value.
By applying the ability to understand meaning, businesses can get a view of all of their data, not only the 10% of “neat”, structured data, but the whole 100%.
What we can do now has never been possible before: the next IT revolution is happening in the “I” – the information – not the “T”.