1, big data computing model is a statistical data perspective entity model usually refers to statistical analysis or big data mining, deep learning, artificial intelligence technology and other kinds of entity model, these models are defined from the perspective of scientific research.

2, the main points of the big data computing model: dimensionality reduction: data mining of large amounts of data and large-scale data, often faced with "dimensionality disaster". The dimensionality of the data set is increasing indefinitely, but due to the limited processing power and speed of computers, in addition, there may be **** the same linear relationship between multiple dimensions of the data set. This will immediately result in a lack of scalability of the learning model, and even many of then optimization algorithms results will be invalid. Consequently, one must reduce the total number of dimensions and minimize the hazard of *** linearity between dimensions. Data dimensionality reduction is also known as data reduction or data approximation. Its purpose is to reduce the number of dimensions involved in data computation and modeling. There are two types of data dimensionality reduction ideas:one is feature selection based dimensionality reduction and the other is dimensionality transformation based dimensionality reduction. Regression:Regression is a method of data analysis that examines the data analysis of variable X on dependent variable Y. The most parsimonious regression model we understand is the one-dimensional linear regression (contains only one independent and dependent variable and drying out the relationship here can be represented by a straight line). Regression analysis is categorized into single and multiple regression models based on the number of independent variables. Depending on whether the effects are linear or not, they can be categorized into linear and non-linear regression. Clustering: We have all heard the phrase "things are grouped together, people are grouped together", which is the basic idea of cluster analysis. Cluster analysis is the basis of big data mining and measurement of daily tasks, cluster analysis is a lot of statistical data concentration with "similar" characteristics of statistical data points into a consistent type, and finally transformed into a number of classes of the way. There must be similar data points in a large number of data sets. Based on this assumption, it is possible to differentiate the data and to find the characteristics of each data set (classification).