Pearson correlation Degree

When the data is not standardized can give better results, better to obtain the overall change of variables.

But the formula is not intuitive ...

#-*-coding:utf-8-*- fromMathImportSqrtprefs={}#Total Datap1={}P2={}#two data to compare, expressed in a dictionarysi={}#the same sequence table forIteminchPREFS[P1]:ifIteminchPREFS[P2]: Si[item]=1N=len (SI)#get the same numberifN==0:flag = 1sum1= SUM ([Prefs[p1][it] forItinchSI])#sumsum2 = SUM ([Prefs[p2][it] forItinchSi]) SUM1SQ= SUM ([Pow (prefs[p1][it],2) forItinchSI])#ask for the sum of squaressum2sq = SUM ([Pow (prefs[p2][it],2) forItinchSi]) Psum= SUM ([Prefs[p1][it]*prefs[p2][it] forItinchSI])#The sum of the productNum= psum-(sum1*sum2/N) den= sqrt ((Sum1sq-pow (sum1,2)/N) * (Sum2sq-pow (sum2,2)/N))ifDen==0:flat =0r= num/denifFlag! = 1:    PrintR

Pearson correlation Degree