Pearson correlation coefficient and test P-value _ machine learning

It depends on two aspects: the significant level and the correlation coefficient.
(1) The significant level is the P value, which is the first, because if it is not significant, the correlation coefficient is no longer useful, may only be caused by accidental factors, then how much is significant, the general P value is less than 0.05 is significant, if less than 0.01 is more significant, such as p value = 0.001, is a very high level of significant , as long as significant, it can be concluded that the rejection of the original hypothesis has nothing to do with the two groups of data significantly related to the fact that there is a clear relationship between them. It is generally necessary to have a P value less than 0.1, preferably less than 0.05 or even 0.01, before we can conclude that the two sets of data have a significant relationship It can only be explained that the degree of correlation is not obvious or irrelevant. At least not linear correlation.
(2) correlation coefficient, namely Pearson correlation (Pearson correlation coefficient), usually also known as R value, in the confirmation of the above indicators, and then look at this index, the higher the general correlation coefficient indicates that the relationship between the two more closely.
R>0 represents a positive correlation of a variable, that is, one becomes larger and the other becomes larger.