SWM/SWM.py at master · nishark/SWM · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
import numpy as np
from sklearn.decomposition import TruncatedSVD
import matplotlib
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn import metrics
from numpy import eye, asarray, dot, sum, diag
from numpy.linalg import svd
from scipy.stats import spearmanr
'''

Original files
usersFile = open("/Users/apurvabharatia/Desktop/SWM/sample_dataset/users.csv")
likesFile = open("/Users/apurvabharatia/Desktop/SWM/sample_dataset/likes.csv")
ulFile = open("/Users/apurvabharatia/Desktop/SWM/sample_dataset/users-likes.csv")

'''


#Following is the user footprint matrix of the users who have liked more than 50 pages
#and pages which have more than 150 likes
#along with the user info of the users present in the updated user footprint matrix

denseMatrixFile = open("/Users/apurvabharatia/Desktop/DenseMatrix.csv")
newUsersFile = open("/Users/apurvabharatia/Desktop/newUsers.csv")

#Get above files in a matrix format
M = []
i = 0
for line in denseMatrixFile:
    if i == 0:
        i += 1
        continue
    currLine = line.split(",")
    M.append(currLine[1:])

#Users file values: id, gender, age, pol, ocean
users, gender, age, political, o, c, e, a, n = [], [], [], [], [], [], [], [], []
i = 0
for line in newUsersFile:
    if i == 0:
        i+= 1
        continue
    currLine = line.split(",")
    gender.append(currLine[2])
    age.append(currLine[3])
    political.append(currLine[4])
    o.append(currLine[5])
    c.append(currLine[6])
    e.append(currLine[7])
    a.append(currLine[8])
    n.append(currLine[9])

    users.append(currLine[1:])


print("Dimensions of M: ", len(M), len(M[0]))
print("Dimensions of newUsers: ", len(users), len(users[0]))

M = np.matrix(M, dtype='float')
def varimax(Phi, gamma = 1, q = 20, tol = 1e-6):
    p,k = Phi.shape
    R = eye(k)
    d=0
    for i in xrange(q):
        d_old = d
        Lambda = dot(Phi, R)
        u,s,vh = svd(dot(Phi.T,asarray(Lambda)**3 - (gamma/p) * dot(Lambda, diag(diag(dot(Lambda.T,Lambda))))))
        R = dot(u,vh)
        d = sum(s)
        if d_old == 0 or d/d_old < tol: break
    return dot(Phi, R)

U, D, V = np.linalg.svd(M)
v_rot = varimax(V)
u_rot = M * v_rot

X = u_rot
Y = o
X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2, random_state=0)
regressor = LinearRegression()
regressor.fit(X_train, Y_train)

y_pred = regressor.predict(X_test)

coef, p = spearmanr(y_pred, Y_test)
print(coef, p)

sentiments = []
sentiments.append(political)
sentiments.append(gender)
sentiments.append(o)
sentiments.append(c)
sentiments.append(e)
sentiments.append(a)
sentiments.append(n)
result = []
for i in range(len(sentiments)):
    Y = sentiments[i]
    leny = len(Y)
    lenFold = leny // 10
    #Create folds and create model for each fold
    foldresult = []
    for j in range(10):
        start = lenFold * j
        end = min(start + lenFold, leny)
        X_train, X_test, Y_train, Y_test = X[:start] + X[end:], X[start: end], Y[:start] + Y[end:], Y[start: end]
        regressor = LinearRegression()
        regressor.fit(X_train, Y_train)

        y_pred = regressor.predict(X_test)

        coef, p = spearmanr(y_pred, Y_test)
        foldresult.append(coef)
    result.append(foldresult / 10)


print(result)
#print(y_pred)
#print(Y_test)


#SVD