Big data modeling is a process of data mining, which is to find problems from data, explain these problems and establish corresponding data models. Big data modeling is not just a technology, but a process to solve business process problems. If there are no goals or business problems, then there is no big data modeling.

Big data modeling should be based on business understanding of business knowledge, and it is necessary to know what the relationship between these related data and business problems is and how to relate them. In the final forming stage, it is necessary to use business knowledge to model, and the established big data model must pass the question and answer of business questions.

Big data modeling is not only an action of modeling, but also many links in the whole process are very important. In the process of modeling big data, finding a suitable data source is the key point, but it is difficult to preprocess the data source. Although there are many automatic data processing tools available now, these analysis tools and various analysis methods have been explored for a long time.

When modeling big data, don't be anxious in the data preprocessing stage, and find an analysis method suitable for data preprocessing. When modeling big data, we should pay attention to some original patterns of data;

For example, in the process of analyzing customers' purchase behavior, customers' subsequent purchase predictions may be related to previous purchase behaviors. Of course, this process is closely related to the operator's experience, especially after understanding the initial business knowledge, you may have a better understanding of this original model.

A model is established, and many people will make various predictions according to this model. If the prediction is accurate, the model is good and valuable. In fact, this model can not be used as a standard to judge value. A good big data model is to change the behavior of enterprises, improve the behavior of enterprises with the predicted results, and transmit new knowledge and insights. Whether it can adapt to the needs of business development is its yardstick.