sklearn 相似度矩阵_基于Sklearn机器学习实战---基于Sklearn模块的链路预测

分手后的思念是犯贱 2023-01-08 06:24 142阅读 0赞

Sklearn简介

自2007年发布以来，scikit-learn已经成为Python重要的机器学习库了。scikit-learn简称sklearn，支持包括分类、回归、降维和聚类四大机器学习算法。还包含了特征提取、数据处理和模型评估三大模块。

sklearn是Scipy的扩展，建立在NumPy和matplotlib库的基础上。利用这几大模块的优势，可以大大提高机器学习的效率。

sklearn拥有着完善的文档，上手容易，具有着丰富的API，在学术界颇受欢迎。sklearn已经封装了大量的机器学习算法，包括LIBSVM和LIBINEAR。同时sklearn内置了大量数据集，节省了获取和整理数据集的时间。

项目简介

链路预测是通过历史连接信息预测未来可能产生的连接，即通过当前网络中的连边信息预测将来可能产生的连边信息。

项目源码

from sklearn.model\_selection import train\_test\_split \# 分割数据模块

from sklearn.neighbors import KNeighborsClassifier \# K最近邻(kNN，k-NearestNeighbor)分类算法

import pandas as pd

import numpy as np

from sklearn.model\_selection import train\_test\_split

from sklearn.neighbors import KNeighborsClassifier

from sklearn import preprocessing

import matplotlib.pyplot as plt

from sklearn.svm import SVC

from math import isnan

定义计算共同邻居指标的方法

define some functions to calculate some baseline index

计算Jaccard相似性指标

def Jaccavrd(MatrixAdjacency\_Train):

Matrix\_similarity = np.dot(MatrixAdjacency\_Train,MatrixAdjacency\_Train)

deg\_row = sum(MatrixAdjacency\_Train)

deg\_row.shape = (deg\_row.shape\[0\],1)

deg\_row\_T = deg\_row.T

tempdeg = deg\_row + deg\_row\_T

temp = tempdeg - Matrix\_similarity

Matrix\_similarity = Matrix\_similarity / temp

return Matrix\_similarity

定义计算Salton指标的方法

def Salton\_Cal(MatrixAdjacency\_Train):

similarity = np.dot(MatrixAdjacency\_Train,MatrixAdjacency\_Train)

deg\_row = sum(MatrixAdjacency\_Train)

deg\_row.shape = (deg\_row.shape\[0\],1)

deg\_row\_T = deg\_row.T

tempdeg = np.dot(deg\_row,deg\_row\_T)

temp = np.sqrt(tempdeg)

np.seterr(divide='ignore', invalid='ignore')

Matrix\_similarity = np.nan\_to\_num(similarity / temp)

Matrix\_similarity = np.nan\_to\_num(Matrix\_similarity)

return Matrix\_similarity

def file2matrix(filepath):

f = open(filepath)

lines = f.readlines()

matrix = np.zeros((50, 50), dtype=float)

A\_row = 0

for line in lines:

list = line.strip('\\n').split(' ')

matrix\[A\_row:\] = list\[0:50\]

A\_row += 1

return matrix

filepath = '3600/s0001.txt'

MatrixAdjacency = file2matrix(filepath)

similarity\_matrix\_Jaccavrd = Jaccavrd(MatrixAdjacency)

similarity\_matrix\_Salton = Salton\_Cal(MatrixAdjacency)

filepath2 = '3600/s0002.txt'

MatrixAdjacency2 = file2matrix(filepath2)

similarity\_matrix\_Jaccavrd2 = Jaccavrd(MatrixAdjacency2)

similarity\_matrix\_Salton2 = Salton\_Cal(MatrixAdjacency2)

filepath3 = '3600/s0003.txt'

MatrixAdjacency3 = file2matrix(filepath3)

similarity\_matrix\_Jaccavrd3 = Jaccavrd(MatrixAdjacency3)

similarity\_matrix\_Salton3 = Salton\_Cal(MatrixAdjacency3)

获取jaccard相似性矩阵的行数和列数

Jaccard\_Row = similarity\_matrix\_Jaccavrd.shape\[0\]

Jaccard\_Column = similarity\_matrix\_Jaccavrd.shape\[1\]

Jaccard\_List = \[\]

for i in range(Jaccard\_Row):

for j in range(Jaccard\_Column):

if i

index = similarity\_matrix\_Jaccavrd\[i,j\]

if isnan(index) == True:

index = 0

Jaccard\_List.append(index)

获取Salton相似性矩阵的行数和列数

Salton\_Row = similarity\_matrix\_Salton.shape\[0\]

Salton\_Column = similarity\_matrix\_Salton.shape\[1\]

Salton\_List = \[\]

for i in range(Salton\_Row):

for j in range(Salton\_Column):

if i

index = similarity\_matrix\_Salton\[i,j\]

if isnan(index) == True:

index = 0

Salton\_List.append(index)

获取jaccard相似性矩阵的行数和列数

Jaccard\_Row2 = similarity\_matrix\_Jaccavrd2.shape\[0\]

Jaccard\_Column2 = similarity\_matrix\_Jaccavrd2.shape\[1\]

Jaccard\_List2 = \[\]

for i in range(Jaccard\_Row2):

for j in range(Jaccard\_Column2):

if i

index2 = similarity\_matrix\_Jaccavrd2\[i,j\]

if isnan(index2) == True:

index2 = 0

Jaccard\_List2.append(index2)

获取Salton相似性矩阵的行数和列数

Salton\_Row2 = similarity\_matrix\_Salton2.shape\[0\]

Salton\_Column2 = similarity\_matrix\_Salton2.shape\[1\]

Salton\_List2 = \[\]

for i in range(Salton\_Row2):

for j in range(Salton\_Column2):

if i

index2 = similarity\_matrix\_Salton2\[i,j\]

if isnan(index2) == True:

index2 = 0

Salton\_List2.append(index2)

获取jaccard相似性矩阵的行数和列数

Jaccard\_Row3 = similarity\_matrix\_Jaccavrd3.shape\[0\]

Jaccard\_Column3 = similarity\_matrix\_Jaccavrd3.shape\[1\]

Jaccard\_List3 = \[\]

for i in range(Jaccard\_Row3):

for j in range(Jaccard\_Column3):

if i

index3 = similarity\_matrix\_Jaccavrd3\[i,j\]

if isnan(index3) == True:

index3 = 0

Jaccard\_List3.append(index3)

获取Salton相似性矩阵的行数和列数

Salton\_Row3 = similarity\_matrix\_Salton3.shape\[0\]

Salton\_Column3 = similarity\_matrix\_Salton3.shape\[1\]

Salton\_List3 = \[\]

for i in range(Salton\_Row3):

for j in range(Salton\_Column3):

if i

index3 = similarity\_matrix\_Salton3\[i,j\]

if isnan(index3) == True:

index3 = 0

Salton\_List3.append(index3)

获取邻接矩阵的行数和列数

Adjacency\_Row = MatrixAdjacency.shape\[0\]

Adjacency\_Column = MatrixAdjacency.shape\[1\]

Adjacency = \[\]

for i in range(Adjacency\_Row):

for j in range(Adjacency\_Column):

if i

index = MatrixAdjacency\[i,j\]

Adjacency.append(index)

获取邻接矩阵的行数和列数

Adjacency\_Row2 = MatrixAdjacency2.shape\[0\]

Adjacency\_Column2 = MatrixAdjacency2.shape\[1\]

Adjacency2 = \[\]

for i in range(Adjacency\_Row2):

for j in range(Adjacency\_Column2):

if i

index2 = MatrixAdjacency2\[i,j\]

Adjacency2.append(index2)

获取邻接矩阵的行数和列数

Adjacency\_Row3 = MatrixAdjacency3.shape\[0\]

Adjacency\_Column3 = MatrixAdjacency3.shape\[1\]

Adjacency3 = \[\]

for i in range(Adjacency\_Row3):

for j in range(Adjacency\_Column3):

if i

index3 = MatrixAdjacency3\[i,j\]

Adjacency3.append(index3)

data = np.zeros((1225,3))

data2 = np.zeros((1225,3))

data3 = np.zeros((1225,3))

for i in range(1225):

data\[i\]\[0\] = Jaccard\_List\[i\]

data\[i\]\[1\] = Salton\_List\[i\]

data\[i\]\[2\] = Adjacency\[i\]

for j in range(1225):

data2\[j\]\[0\] = Jaccard\_List2\[j\]

data2\[j\]\[1\] = Salton\_List2\[j\]

data2\[j\]\[2\] = Adjacency2\[j\]

for k in range(1225):

data3\[k\]\[0\] = Jaccard\_List3\[k\]

data3\[k\]\[1\] = Salton\_List3\[k\]

data3\[k\]\[2\] = Adjacency3\[k\]

data\_train\_X = data\[:,0:2\]

data\_train\_y = data\[:,2\]

data\_test\_X = data2\[:,0:2\]

data\_test\_y = data2\[:,2\]

data\_target\_X = data3\[:,0:2\]

data\_target\_y = data3\[:,2\]

knn = KNeighborsClassifier()

knn.fit(data\_train\_X,data\_train\_y)

print(knn.predict(data\_test\_X))

print(data\_test\_y)

clf = SVC()

clf.fit(data\_train\_X,data\_test\_y)

print(clf.score(data\_test\_X,data\_target\_y))

项目详细了解

如需详细本项目信息，可发送邮件至18770918982@gmail.com

sklearn 相似度矩阵_基于Sklearn机器学习实战---基于Sklearn模块的链路预测

发表评论取消回复

还没有评论，来说两句吧...

相关阅读

相关机器学习之Sklearn基础教程

相关基于sklearn的机器学习股票预测

相关 sklearn 相似度矩阵_基于Sklearn机器学习实战---基于Sklearn模块的链路预测

相关《菜菜的机器学习sklearn课堂》sklearn入门与决策树

相关基于sklearn 的auc 计算方法

相关基于sklearn的特征选择方法

相关 sklearn学习（三）

相关 sklearn 机器学习库案例

相关安装Python的sklearn,机器学习

相关 [python] 机器学习 sklearn和tensorflow