By Jonathon Savill
We look at the little strings of code that are taking over the world — from writing music to dealing on the stock exchange
In its earliest years, science was investigative. In the seventeenth century we observed the march of the skies to learn how to calculate our position on the globe. Then we began to theorise and to add rules to our life. Finally we discovered DNA (Deoxyribonucleic acid molecules) - the very building blocks of our biological existence.
Not content with splitting the atom or discovering the fountain of life we began to change it. The power of the atom is already well-acknowledged and DNA has been used to develop genetically modified (GM) crops. But there is a quieter revolution that is largely unknown, yet potentially will change life in much the same way as the other two building blocks of life. Welcome to the world of the the unglamorous - but vitally important - algorithm.
In fact, there is a strong Middle East connection with algorithms. The word algorithm comes from Abu Abdullah Muhammad ibn Musa Al Khwarizmi, a Persian mathematician from the ninth century. The first recorded algorithm comes from Shuruppak, near Baghdad.
How do algorithms affect you? Well, those of you who came to work on the metro today were driven not by humans but by algorithms. If you send money home, algorithms will decide exactly how much you get for your dirham and convert it into your home currency. Those are the ways that you can see the computer code working. But this science is in its infancy and you will find it not just entering your life but controlling it in the very near future.
Where algorithms work best at their lowest level is at the point where repetition is likely to cause humans to make mistakes. In Wall Street trading, algorithms started with very simple instructions - ‘if a price is here buy, and if the price reaches here, sell’. This simple instruction can be repeated several times a second and never makes a mistake. It’s simple ‘bot’ logic is never distracted and never goes to the bathroom.
But it can work on more advanced platforms too. Let’s say you are in Europe and see that a large amount of money has been taken from your account without your knowledge. You ring the helpline and you are furious.
Algorithms can help here. At the first point of contact, certain key words can tell the system that you are angry. One company called eLoyalty has now analysed more than 750 million conversations and logged certain keywords to trigger operative reaction to your conversation. They have broken humanity down into six major types. It will quickly work out your personality type and put you in touch with an operative that has the same personality type as you.
The result is that busy people are dealt with quickly but old ladies are allowed to talk about their cats.
Sounds flaky? Vodafone took the eLoyalty model based on the six personalities and applied it to their marketing. Each of the personality types had a sales pitch directly targeted at them. The result was an 8,600 percent increase in sales. The types of personality are identified by key words.
NASA used the six personality types to analyse those who would do well in space. In such confined spaces, and in such close proximity, human frailties are likely to develop and cause friction. In order to save not only mutiny but endorse team-working, NASA employs rigorous testing of its crews.
Amazon uses algorithms to suggest books you may enjoy. At its simplest level an algorithm is just a set of ‘what if’ instructions. For example, users often make mistakes when they key in their credit card numbers online. The numbers identify the card, type, where it is issued and so on. The Lunn algorithm weeds out false or mis-keyed numbers by mathematical formula and thus saves millions of dollars in processing times.
UPS uses algorithms to help deliver millions of packages every day. If a driver has only three destinations to visit, he can take only six possible routes. But the number of possible routes explodes as the destinations increase.
There are more than 15 trillion, trillion possible routes to take on a journey with just 25 drop-off points — and an average day for a UPS driver in America involves 150 destinations. “Algorithms provide benefits when the choices are so great that they are impossible to process in your head” says UPS’s Jack Levis.
Dunnhumby, a data-analysis firm, uses algorithms to crunch data on customer behaviour. Its best-known customer, Tesco - a British supermarket - generates data on 13m clients and 55,000 product lines. Dunnhumby’s analysts invented an algorithm called the rolling ball. It assigns attributes to each of the products on Tesco’s shelves. These range from easy-to-cook to value-for-money, from adventurous to fresh. The rolling-ball algorithm starts at the extremes: ostrich burgers, say, would count as very adventurous.
The algorithm then trawls through Tesco’s purchasing data to see what other products (staples such as milk and bread aside) tend to wind up in the same shopping baskets as ostrich burgers do.
“All this sophisticated data analysis and it comes down to where you put the biscuits,” says Martin Hayward, of Dunnhumby.
If a customer calls an electricity utility from an area that has suffered a power failure, they are automatically put through to an operator who can deal with their queries. Algorithms can engender simplicity from complexity.
Algorithms have been used to analyse music and have produced orchestral pieces. They have been used to generate poetry. They have been used to analyse whether films will succeed. In short, they can tear apart the fabric of any consumable. Even the spell checker on your computer uses algorithms to give you the words you meant to type. The future of algorithms is assured and massive. Key business opportunities exist for those who realise this.
One of the key targets for algorithms is the medical and pharmaceutical industries. The United States spends $2.6 trillion a year on healthcare. Doctors now are largely driven by flowcharts. When a patient comes in ,the doctor asks a series of questions to elicit what is wrong with the them. But they are prone to inaccuracy and the proficiency of the doctor and the truthfulness of the patient. Consider how often the question ‘how much do you smoke?’ has ever been answered truthfully.
Dutch company Philips makes a 99cent iPhone app that takes several vital readings from the person looking at the screen. Using algorithms it will soon be able to measure blood pressure, temperature, blood oxygen levels and signs of concussion.
Using the devices we already carry, like mobile phones, to monitor our vital signs and the extended knowledge we have of DNA algorithms could predict what diseases we are likely to suffer, and crucially when they are likely to strike.
There could be a future where someone who was going to have a heart attack could wander into a clinic and avoid it. Algorithms are now available to diagnose cancer and determine the optimum heart management procedure using the latest worldwide research. Can human doctors compete in the longer term and will algorithms be better at applying game theory to determine the ethical outcomes of who should live or die?
In 2011, the University of California opened a pharmacy staffed by a single $15m robot built by Swisslog, a Swiss logistics firm. The machine receives information from the standard emails that already emanate from the doctor’s office. The robot then picks the pills from long rows of boxes built into the pharmacy.
So far the robot has dispensed over 2m prescriptions without making a single error. But more than that it does not need to keep up with new technologies, as these are fed in as updates. By comparison, a human pharmacist in California costs around $130,000 a year. But were these robot pharmacies to multiply, the price would drop massively.
Lots of things have to fall into place for algorithms to work. They tend to be highly complex; it is not easy to find people with the right skills to develop and refine them. The systems within which the algorithms run — the user interface — need to be intuitive to non-boffins. “This is rocket science but you don’t have to be a rocket scientist to use it,” says Jack Noonan, boss of SPSS, a predictive analytics software firm. “The inputs have to be right”.
Algorithms are obviously on the march across the globe. The more data they can process the more accurate the answer they can give. For that reason they are bound to take over the world. But as American computer scientist Alan J. Perlis observes” “The cybernetic exchange between man, computer and algorithm is like a game of musical chairs: the frantic search for balance always leaves one of the three standing ill at ease.”