Crop Growth Model in the Era of Big Data

What should the crop growth model look like in the era of big data? This is a question that many people may not be understand, because we used to say mathematical model, we may guess that it is a calculation program expressed by fixed formula and data.

Today, in the era of big data, the mathematical model we call is no longer a computer program of fixed data and formulas. The important thing is that the computer program and its input data are random. A large number of regular variable data and programs show that the data of crop growth model is not a fixed parameter, but it still has certain rules. In other words, the basic function and basic application of big data is the upgraded version of the application of modern statistical technology in big data.

Before we talked about the mathematical model of crops, we just said that it is a kind of abstract and simplified growth law of crops in the objective world that can be expressed by mathematical formula and computer program. Now we can further say that crop mathematical model is the concrete embodiment of applying big data to crop growth model.

In fact, in the era of big data, people are no longer studying the objects that can be expressed by fixed formulas and programs. For example, the writing of this paper is completed by using voice input method, which is realized in the cloud platform by using the voice analysis technology of big data. In the past, the speech input method that IBM just invented and created is completely based on each person's different speech habit model to realize speech recognition, so it can only have a training process to recognize your voice. Moreover, after the training,that is the text can not achieve the effect you want, which is really unsatisfactory. Today, the speech input method is approaching maturity, although the content of the output text is not 100% of the text content you want. However, the recognition rate has reached more than 80% in the general environment for any person, any local accent. When I wrote this text, I used the voice input method, which is what you see in this article. 80% of it is automatically completed. Although my dialect is not standard, but it can automatically input the text content in most cases.

It shows that the background analysis of big data, voice input has been able to complete the basic requirements of human beings, who want to automatically input text content through voice. In the same case, at the beginning of the handwriting input method, it must be implemented by the hardware system. Under the current input method, you pick up your mobile phone and draw a letter by hand. As long as the letters you draw are not very scrawly, the handwriting input system can display it on the screen correctly.

These are the application of big data analysis and machine learning method to be used in input method. The same is true for the crop growth model, we are going to talk about today. It is impossible for you to imagine in the era of big data that the crop model is still a fixed formula and has fixed parameter. It can also be used to get a model that can reflect the real crop growth. We should imagine that our crop growth is impossible to have a fixed growth mode in the real world. Because crop growth is a complex process, it has many influencing factors, and these influencing factors are also random changes, so the degree of influencing factors, that is, the data of factors, is also a random variable. Today, what we are talking about here is not an immutable crop mathematical model. First of all, its parameters are some randomly changing parameters. At the same time, its model must go through continuous iterations to get the simulation of the real growth and truly reflect the real growth situation of crops.

So, in today's era of big data, the mathematical model of crops should be a model of continuous iteration and upgrading, but it has laws to follow. We can always find a statistical method. Now it's called big data, AI learning method, to find the real crop modelling, meet the acceptable confidence interval, and representative the real growth model.