What can be done with large data applications?

No matter how hot the big data is, unless it's big data for big data, there's no way to evade the question: what can be done with big data?

What do you do with the big data? Or is it going back to basics: What does big Data bring? It is said that it does not create value, it value is through the http://www.aliyun.com/zixun/aggregation/20826.html "> Traditional industry efficiency improvement." Large data as part of it is also consistent with such a rule.

If we think about it further, how can traditional industrial efficiency be improved with the help of large data? What's the big data? Improve decision making ability, what full data analysis, raise people's cognitive level. These are all theories, and since they are theories, they are often unassailable. But the actual value of the theory lies in the landing, combined with practice.

What new business models and values do you bring to us when the big data is combined with practice? In the article "Industry Big Data: Ideal fullness reality bone feeling", once said, RTB, DSP, SSP, Ad Exchange and adnetwork are all the Internet Big Data application pronoun, but for the traditional industry /Enterprises, these are still lack of reference value.

For traditional industries/enterprises, in fact, can also find some big data application keywords, such as: Bank card fraud analysis, bank card real-time monitoring, operator customer churn analysis, shopping basket analysis, pathological diagnosis. In fact, these applications are not large data created, but large data will make these applications fundamentally change.

To the loss of telecommunications customers for example, the operator's customer base is actually very weak, small to a tariff promotion, or mobile phone bundling, may lead to customer churn. I have received many friends to change number 186 text messages, I am afraid that is thanks to Apple. Now think about it, if there is a big data analysis warning, such things may not happen, at least can take measures to avoid the emergence of one-sided situation.

4G is the case, many people do not know how much faster than 3G, also do not know what the difference between LTE and WCDMA, but the WCDMA label is 3g,td-lte is 4G, most will choose 4G. Or what words, if there are large data analysis, Unicom, telecom may not be so passive.

I asked some colleagues why have been using China Mobile, the reason is very simple, its number is bound a lot of services, such as bank cards, QQ, Alipay, etc., too much to remember their own. But for big data, that shouldn't be a problem. Therefore, for operators, large data customer churn analysis is not only embellishment, but a part of the business.

For industry enterprises, rather than make large data dizzy brain swelling, may also study similar "customer churn analysis", "shopping basket analysis" Such applications, evaluate the value of these applications.