Before discussing this issue, it is necessary to understand the meaning of model-driven and data-driven two words. Exactly what is model driven. From the perspective of understanding the world, we understand the system, etiquette, morals and so on, basically can be understood as a model, through these models, we can clearly understand what is good, what is bad, what to do, what not to do. However, in the data side, in the business understanding, can also be similar to understand. We need to comb a clear set of ideas to make the business better, called the business model, defined the specific participants, processes and other key factors. We need a set of components to implement a function, and so on, which is within the scope of our understanding of the model. The data analysis that is emitted from the model side is called the model driver. This is like, we analyze a situation, such as to understand the history of the highest consumption of the user groups have what characteristics, in response to this problem, we have a set of ideas and frameworks to portray these users, also known as the model, based on the data analysis, can be called model-based data analysis. However, what is data driven? To answer this question, let's take a closer look at what data is. Data, which is a quantitative or qualitative record of natural, social phenomena and scientific experiments, and is the most important basis for scientific research. Research data is the collection, classification, input, storage, statistical analysis, statistical inspection, such as a series of activities collectively. Data-driven can be understood as a data base to find the model, so that the data generate value, to achieve the goal of profitability.
Then, the problem comes.
Since the model is discovered by data-driven approach, it is the model-driven category to solve the real problem with the model. On the other hand, the model-driven approach to the problem-solving framework and scheme is based on data, which is the data-driven range. Data-driven and model-driven are inseparable, data-driven is holding data to find the model, and model-driven is to take the model to find data. Both Yin and yang coordination, the elimination of the long, after running-in, business value-oriented to the integration of ideas as an opportunity to achieve data and model of the return, thus counterproductive business, promote the value of Ascension. So, whether the data analysis is model-driven or data-driven, the problem is not established. But the role of data and model is actually in the process of business problem solving and data analysis. I think what we should do is specific issues specific discussion, in the actual work carried out, play the role of data and model, truly realize data realization.
