Before the millennium, ordinary people would not have imagined that the development of the Internet would be so rapid, rapidly occupying half of the country in various fields, so that in the past twenty years, all corners of Chinese social life have the "figure" of the Internet. The development of emerging industries is based on network derivation, such as the emergence of artificial intelligence, Internet of Things technology, big data and other technologies.
The term big data is familiar to young people after the millennium, regardless of whether they understand its specific content or not, but the frequency of this term in life is not low at all. So, what is big data?
First, the question is actually proposed at the beginning, the term big data is proposed for the development of enterprises, enterprises in the process of growth will continue to generate a large amount of unstructured historical data, such data is always accumulating, and then at a certain stage they will be used as an important decision-making basis for the company. Unlike before the millennium, corporate decisions are based more on the experience and business sense accumulated by the company's management over a long period of time.
But with the rapid development of network technology, various fields are constantly integrating with network technology, such as express delivery services, takeaways, transportation, financial management and even home appliances, twenty years ago, who would have thought that vacuum cleaners could decide where to clean by themselves? In this context, the term big data has also been given a very different meaning.
An increasing number of hardware devices, software interfaces, and various types of users are connected to the network link, and these "individuals" connected to the network have data generated from the network every minute. The "network user behavior data, etc." accumulated by these individuals in the network is massive data.
The scale of network user behavior data is not simple to describe with "MB" or "GB", according to IBM research data, the world Internet in a day, the entire content generated, can be engraved nearly 200 million DVD discs, about 300 billion emails per day, this data volume is no longer "massive" can be described.
These behavioral data generated by network users are collected and based on the business needs or areas of concern, and then screened and organized through scientific and mathematical modeling, and finally applied within each domain.
For example, you often search for the keywords "computer", "tablet", "pen display" and so on in some shopping apps, and then these "behavioral data" generated by you will appear, which will be collected by the APP or mobile phone or some other unknown program, and the collected data will be collected, sorted, analyzed, modeled and filtered by other applications before being used for you. Therefore, when you open the mobile phone again, even if it is not on the same application, the mobile phone will continuously push you "computer" and "tablet" related advertisements and photos, that is, big data accurate push.
Use user behavior data to analyze their user characteristics, such as when you log in, which sections you spend the most time on, which content you click the most, etc. These materials do not have any characteristics individually, but if you accumulate over a long period of time and synthesize these materials, then more than 90% of your characteristics as a user can be completely analyzed. Finally, use each user's characteristics to personalize the push, because of this, everyone on the network, can always see what they want to see.
Coincidentally, big data brings development potential to various industries, but it also brings many hidden dangers, such as privacy leakage, telecom fraud, pornography, gambling and drug concealment.
But these potential dangers are also slowly being overcome, such as legislation to protect data security and privacy, and big data to eliminate pornography. Big data anti-pornography is also a typical example of big data applications. Those with these types of "user characteristics" may be identified as "pornographic".
First, the user time characteristics, such as a certain account has hardly been used, but each use is a large capital transaction in a night period, and the amount of each fund transaction may not change.
Second, the characteristics of user sensitive words, for example, an account often has some sensitive words that can only be identified in the field of pornography, gambling and drugs, accompanied by capital transactions, then such users are undoubtedly the key targets in network supervision.
Big data brings convenience to various industries, as does anti-pornography and public security.
The use of big data to sweep pornography, the first can save costs, network monitoring can save a lot of police resources, without frequent police dispatch.
Second, network monitoring is also faster, and the accuracy of network monitoring and identification according to user characteristics is higher.
conclusion
The development potential of big data is not only these, but the hidden dangers it causes also need people's attention, and continuous improvement of policies and laws and regulations is a reliable guarantee for its benign development. At the same time, as "web users", we also need to take responsibility for the behavioral data generated and protect our privacy.