Summary What is "big data"? The straightforward explanation is: through a certain technology to obtain massive data at a reasonable time and analyze it, and then use this analysis result to do various research, decision-making and so on. The author of Big Data believes that the three priorities must first be clarified. &n
What is "big data"? The straightforward explanation is: through a certain technology to obtain massive data at a reasonable time and analyze it, and then use this analysis result to do various research, decision-making and so on. The author of Big Data believes that the three priorities must first be clarified. The scale of the data must be "big"
The larger the data size, the higher the accuracy of the analysis results, and the results of the data on the order of billions or even billions of bytes are relatively accurate. But if the data is not big enough, there is no way for many data mining and forecasting work. For a simple example, long-term tracking of a user's browsing habits and various operations on the Internet can be very accurate predictions for him, but if only one or two times of data is predicted, it will not be too accurate.
"Big" refers not only to scale, but also to the value of value data, first and foremost to research value, and secondly to business value. Only the amount of stacking can not reflect these values, but requires a strong correlation and structure. Relevance means that, for example, Taobao only has a record of buyers, sellers, goods, prices, etc., and the business value is very limited. But if you record the communication and social relationships between the buyer and the seller, and other behaviors before and after the purchase, then this data will be very valuable. The structure is that, for example, a data record the extent to which each tree on the earth grows taller each year, and the value of this simple pile of data is limited. But if the data becomes the location of each tree, the climatic conditions, the tree species, the age of the trees, the ecology of the surrounding flora and fauna, and the height of each year, then this data will add value because of its structural nature.
The ultimate value of "Big Data" is reflected in how the analysis of how big data is used for its analysis and use is key, not technology. For example, Tencent and Alibaba have about one billion users who use their products for daily communication or purchase transactions. So they understand all the user's communication habits or transactions. Imagine that if the big data of the two companies are merged in the future, they can draw a complete human behavior habit from different dimensions and predict his future development and trends. I further imagine that it can be combined with artificial intelligence. Is there a machine that can analyze and predict human behavior? The use of big data is endless, and it is very dangerous to use it wrong. This is also a scientific and ethical issue that needs to be explored in depth.
What is particularly important to note is that most people do not have real "big data" in their work, at most "large-scale data", or even "large-scale". However, in practice, it is still possible to suggest a "big data-like" analytical idea to assist in the promotion of the enterprise.
What data is analyzed?
By analyzing Internet data, we can see the volume of e-commerce, the comments on the social media, and the comments on the brand, so as to measure the effect of public relations work on social media. Analyze media reports to see which media are reporting on corporate brands and companies, and to understand what public praise and criticism are all about helping to resolve potential public relations crises. Analyze ad serving data and feedback, using analytics as a basis for optimizing ad serving.
How to use this data?
Regularly monitor and analyze big data, and act according to the analysis results, such as: replying to the message on the company Weibo and WeChat in time, interacting with the fans; once you find that the social media has negative news about the company, contact the other party at the first time. Correct the attitude to solve the problem, transform a dissatisfied customer into a satisfied customer and so on. In short, don't be afraid of big data coming from the face, but make better use of big data to serve the enterprise.
Iron 3 Oxide,Iron Lll Oxide,Ferric Oxide Red,Transparent Iron Oxide
TINOX CHEMIE GMBH , https://www.tinoxglobal.com