Python实现pagerank算法

import numpy as np

M = np.array([[0,1,1,0],
             [1,0,0,0],
             [0,1,0,1],
             [1,0,0,0]],dtype = float)

# 1.定义转移矩阵
def move_matrix(m):
    num = m.sum(axis = 0) # 统计每一列的总数，也就是网页的链接数
    return m/num  # 返回建立的转移矩阵

# 2.定义V矩阵,初始的PR值
def V(c):
    pr = np.ones((c.shape[0],1),dtype=float)/len(c) # 初始化PR值矩阵
    return pr

# 3.迭代计算pagerank值
def PR(p,m,v):
    i = 0
    while 1:
        v1 = p*np.dot(m,v) + (1-p) * v
        if np.abs((v-v1).all()) < 0.001:
            break
        else:
            v = v1
        i += 1
        if i==30:break
    print('求pr值迭代%d次'%(i))
    return v

# 4.测试
M1 = move_matrix(M)
V1 = V(M1)
a = 0.85
print('最后迭代网页PR值结果为：\n',PR(a,M1,V1))