The application value and potential of large data are no longer underestimated.

The application value and potential of large data are no longer underestimated.

Moore, one of Intel's founders, discovered an alarming trend in 1965, when the number of circuits integrated into integrated circuits doubled every 18 months, a finding that the industry hailed as Moore's law. It was later described as a microprocessor with a 18-month increase in performance, or a price drop of half, or a computer performance (speed and storage) that could be bought at the same price every 18 months, and so on.

More than 40 years in the history of human life is just a flick of a finger, Moore's law has witnessed the computer data processing and storage capacity from K (kilobyte) to M (megabyte) to G (gigabyte) to T (terabyte) changes. In particular, the advent of the internet, let us rush into the big data age. The main driving force has the following points:

1, with the development of social economy and the increase of personal income, people's personalized needs begin to highlight. and enterprises need to be efficient to meet these personalized needs, it requires a lot of data support.

2. The advent of the Internet and the development of related technologies have made it possible to collect and analyze massive data. The characteristics of the Internet have led to the propagation of these data by high speed and large capacity.

3, the Internet introduced by the user to generate data patterns. This pattern is characterized by multiple sources, low-cost, and more timely. Of course, the authenticity and reliability of these data need to be certified.

4, building on the internet based on E-commerce and traditional retail comparison is one of the advantages of data availability. E-commerce can be real-time access to the source of visitors, in the site search, collection, purchase behavior, as well as the relationship between the purchase of goods. This data can help companies to more accurately customer service.

5. The development of artificial intelligence, information system and decision science has promoted the promotion of various analytical methods and tools, including data mining, customer behavior model, decision support, etc.

Data is primitive and fragmented, filtered and organized into information (information), integrating and effective presentation of relevant information becomes knowledge (knowledge), the deep understanding of knowledge to understand the nature of things and analogy to wisdom ( Wisdom). So data is the source, the cornerstone of decision making and value creation.

The application of data is roughly divided into the following steps: A. Data collection, verification and filtration; B. Classification and storage in the Data Warehouse; c. Data mining to find the underlying laws and correlations between data; d. Data model establishment and parameter tuning; E. Data based application development and decision support. Here is an example to illustrate.

1, the American medical website WebMD according to pregnant female user fills in the conception information to send the EDM regularly to the user, reminds the mother in this time point attention matter, needs to ingest the nutrition, the prenatal physiological change and wants completes the preparation, the post-natal recovery, the baby's rearing and the health, and so on ｡

2, shop 1th Use the analysis of large data to send customers personalized EDM. If a customer has ever viewed a product on the 1th store site without buying it, there are several possibilities: a. Out of stock, B. Not the right price, C. Not the brand you want or the product you want, d. Just look at ｡. If the goods are out of stock when the customer is looking, inform the customer immediately; If the goods are available and the customer is not buying it is likely because of the price, then notify the customer when the product is being sold at a lower price, and when the product is similar or associated with the goods, please inform the customer. In addition, by digging the customer's periodic buying habits, in the proximity of the customer's purchase cycle timely reminder customers.

3, Taobao launched in 2012 Taobao time Machine ｡ The application analyzes the behavior of the customer since registering for the user, and tells the customer the growth of Taobao with the humorous and vivid language, and the statistic behavior of other users who have similar preference to the customer, after analyzing the customers ' knowledge about their preferences and predicting their behavior, etc. With vivid manuscripts and personalized data, closer to the customer's distance ｡

4, Google AdSense to the customer's search process and its attention to the site of the data mining ｡ and in its alliance site to track the whereabouts of customers, on the Alliance website and customer potential interest to match the ads, precision marketing, improve conversion rate ｡

5. Amazon has introduced the concept of FDFC (Forward deployed fulfillment Center) in recent years to speed up the delivery of customers. Amazon's order fulfillment center is divided into two tiers: FC and FDFC, in which FC varieties are more complete, while FDFC in physical location closer to the target market, but the variety focus on the target market for the best-selling goods, most of the customer needs can be met through the FDFC, The unsatisfied long tail goods are met by FC. In this way, most of the goods that customers need most can be completed by FDFC with faster and lower cost logistics. Because the hot goods are changed with the time and the season, so the decision of what commodity is stored in the FDFC is dynamically adjusted, and the basis of this decision is to analyze and forecast the customer demand.

Examples of various applications are hard to take, but the trend is clear: the application value and potential of large data are no longer underestimated. But not all companies can really dig gold in big data. Only those who have vision, focus on the system, willing to invest, attract excellent analysis and system talent will have some gains.