Intelligent derivative variable construction method in data modeling

Abstract

The invention discloses an intelligent derivative variable construction method in data modeling. The intelligent derivative variable construction method comprises the following steps that a mobile terminal collects original data and divides the collected original data into static data, behavior data and transaction data; the mobile terminal receives the information sending priority information; the mobile terminal receives a mobile terminal identity identifier, and the mobile terminal identity identifier is generated by the transmission receiving point and distributed to the mobile terminal; the mobile terminal determines a sending sequence based on the information sending priority information; and the mobile terminal determines sending random waiting time and sends a random access requestmessage to the transmission receiving point according to the sending sequence after sending the random waiting time, and the random access request message comprises the sending random waiting time and the mobile terminal identity identifier. By classifying the mobile terminals, it is ensured that some time delay sensitive data can be transmitted preferentially.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a method for constructing intelligent derived variables in data modeling. Background technique [0002] Data modeling refers to the abstract organization of all kinds of data in the real world, determining the scope of the database to be governed, the organizational form of the data, etc. until it is transformed into a real database. After the conceptual model abstracted after system analysis is transformed into a physical model, the process of establishing database entities and the relationship between entities in tools such as visio or erwin (entities are generally tables). [0003] In data modeling, the construction of derived variables is a very complicated problem. Different businesses, different products, and different customers have different construction methods for variables under different analysis objectives. In the traditional mode, analysts need to have a de...

AI Technical Summary

Problems solved by technology

In the traditional mode, analysts need to have a deep understanding of the industry, business, and data they are engaged in before they can construct variable variables to generate characteristic variables, and no matter whether the rules of variables are simple or complex, they need to be completed manually, which requires a lot of analysts. time

There are two problems with this method: (1) The work of constructing variables has relatively high requirements for analysts, who must be familiar with the industry and business, and have rich experience in modeling technology, making such talents rare in the market. It is very scarce, which greatly reduces the mining and utilization of data value by enterprises; (2) Because analysts manually construct derived variables based on their own understanding of the business, for the sake of time investment and modeling efficiency, most of their own data will be discarded. For the data that is considered unimportant, only a small part of the key data is retained to generate variables using familiar rules. In this way, many data with high value but unfamiliar to analysts and data with low single value and high combined value are discarded and cannot be mined. Finding or digging deep into all possible features will greatly affect the accuracy of analysis and prediction of the final model

Images