อัลกอริทึม PageRank ใช้ได้กับหน้าเว็บ หน้าเว็บเป็นกราฟกำกับ เรารู้ว่าองค์ประกอบทั้งสองของ Directed graphsare -nodes และการเชื่อมต่อ เพจคือโหนดและไฮเปอร์ลิงก์คือการเชื่อมต่อ การเชื่อมต่อระหว่างสองโหนด
เราสามารถค้นหาความสำคัญของแต่ละหน้าโดย PageRank และมันถูกต้อง ค่า PageRank คือความน่าจะเป็นระหว่าง 0 ถึง 1
ค่า PageRank ของแต่ละโหนดในกราฟขึ้นอยู่กับค่า PageRank ของโหนดทั้งหมดที่เชื่อมต่อ และโหนดเหล่านั้นเชื่อมต่อแบบวนซ้ำกับโหนดที่เราต้องการจัดอันดับ เราใช้วิธีวนซ้ำแบบบรรจบกันเพื่อกำหนดค่าให้กับ PageRank
โค้ดตัวอย่าง
import numpy as np
import scipy as sc
import pandas as pd
from fractions import Fraction
def display_format(my_vector, my_decimal):
return np.round((my_vector).astype(np.float), decimals=my_decimal)
my_dp = Fraction(1,3)
Mat = np.matrix([[0,0,1],
[Fraction(1,2),0,0],
[Fraction(1,2),1,0]])
Ex = np.zeros((3,3))
Ex[:] = my_dp
beta = 0.7
Al = beta * Mat + ((1-beta) * Ex)
r = np.matrix([my_dp, my_dp, my_dp])
r = np.transpose(r)
previous_r = r
for i in range(1,100):
r = Al * r
print (display_format(r,3))
if (previous_r==r).all():
break
previous_r = r
print ("Final:\n", display_format(r,3))
print ("sum", np.sum(r)) ผลลัพธ์
[[0.333] [0.217] [0.45 ]] [[0.415] [0.217] [0.368]] [[0.358] [0.245] [0.397]] [[0.378] [0.225] [0.397]] [[0.378] [0.232] [0.39 ]] [[0.373] [0.232] [0.395]] [[0.376] [0.231] [0.393]] [[0.375] [0.232] [0.393]] [[0.375] [0.231] [0.394]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] [[0.375] [0.231] [0.393]] Final: [[0.375] [0.231] [0.393]] sum 0.9999999999999951
