Big data explosion will change human mind

The internet has reshaped the way people communicate. The big data is different: It marks a change in the way societies handle information. Over time, big data may change the way we think about the world.

Everyone knows that the internet has changed the way businesses operate, government operates and people live. But a new, less obvious technology trend has the same huge transformative power, the "Big Data" (http://www.aliyun.com/zixun/aggregation/13814.html ">bigdata"). The trend towards large data begins with the fact that today there is a lot more information being disseminated than ever before, and that this trend is being applied to extraordinary new uses. Big data is very different from the Internet, although the Internet makes it much easier to collect and share data. The meaning of large data is not just communication: the essence is that we can learn from a lot of information that we can't get from a little bit of information.

Will change the way people think

In the 3rd century BC, the library of Alexandria was thought to have collected all human knowledge. And if the information of the world today is divided equally among the living, then each person will have more than 320 times times the amount of the library in Alexandria. If all this information is engraved on the disc and stacked in 5 stacks, the discs can be stacked up to the moon.

This data explosion is a relatively new phenomenon.

Only 1/4 of the world's stored information was digitized in 2000, and the rest was stored on paper, film and other analog media. But as the number of digital data has grown very quickly-doubling almost every three years-this situation has quickly reversed. Today, less than 2% of all stored information is digitized.

Given the scale of such disparity, it is unavoidable to think only quantitatively when it comes to understanding big data. But that would be misleading. Another feature of large data is its ability to use data to represent many facets of the world that have never been quantified before-a feature that can be called "data". For example, the data of location information was first due to the invention of latitude and longitude, and recently with GPS. When a computer samples several centuries of books, the text becomes the data being processed. Even friendships and hobbies are being computerized--through Facebook, for example. With cheap computer memory, high-performance processors, intelligent algorithms, smart software and mathematical knowledge borrowed from basic statistics, such data are being used in incredible new uses. The new approach is not to try to "teach" computers to drive or translate such things, instead, it is necessary to enter enough information into the computer so that they can infer probabilities, such as the probability that the traffic indicates a green light, the red light is not bright, or that the term "light" means "light" rather than "light" in a particular context.

The use of large amounts of data in this way requires us to radically change attitudes to data in three ways. The first is to collect and use large amounts of data, rather than being content with a small amount of data or samples, as statisticians have done over the past more than 100 years. The second is to abandon our preference for structured and pure data, and instead accept the disorder--in an increasing number of cases, a little imprecision is tolerable. Third, on many occasions, we need to give up the case and replace it with the acceptance of the relevant sex. Using large data rather than trying to understand the exact cause of engine failure or the disappearance of drug side effects, researchers can collect and analyze a great deal of information about such events and all the relevant material to find out the laws that might help predict future events. Big Data helps answer the question of what it is, not why--usually that is enough.

The internet has reshaped the way people communicate. The big data is different: It marks a change in the way societies handle information. Over time, big data may change the way we think about the world. As we use more and more data to understand things and make decisions, we are likely to find that many dimensions of life are random, not certain.

(Responsible editor: Lu Guang)