-
Notifications
You must be signed in to change notification settings - Fork 0
/
SpeedDating_DataAnalyse.py
250 lines (200 loc) · 9.69 KB
/
SpeedDating_DataAnalyse.py
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
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
# -*- coding: utf-8 -*-
"""
Created on Tue Mar 3 14:31:27 2020
@author: valde
"""
from DataLoad import load_data
from dataCleanUp import scaleGroup, replaceGroup
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
raw_data = load_data()
data = raw_data.drop(['id', 'idg', 'partner', 'position', 'positin1', 'career', "career_c", 'field', 'undergra', 'tuition', 'from', 'zipcode', 'income', 'sports', 'tvsports', 'exercise', 'dining', 'museums', 'art', 'hiking', 'gaming', 'clubbing','reading', 'tv', 'theater', 'movies','concerts', 'music', 'shopping', 'yoga', 'income', 'mn_sat' ], axis=1)
#data = data[data.columns.drop(list(data.filter(regex='_3')))]
#%%
# For fun, see how many found a match
pd.crosstab(index=raw_data['match'],columns="count")
shar = raw_data[["shar","iid"]]
#%% Removing Nans
## Summerizing nans in every feature/coloum
#
## Chosing which features to drop. Based on number of NaNs, and interest for us.
#
#data1 = raw_data.iloc[:, 11:28]
#data2 = raw_data.iloc[:, 30:35]
#data3 = raw_data.iloc[:, 39:43]
#data4 = raw_data.iloc[:, 45:67]
#data5 = raw_data.iloc[:, 69:74]
#data6 = raw_data.iloc[:, 87:91]
#data7 = raw_data.iloc[:, 97:102]
#data8 = raw_data.iloc[:, 104:107]
#
#data = pd.concat([raw_data.iloc[:, 0],raw_data.iloc[:, 2],data1,data2,data3,data4,data5,data6,data7,data8], axis=1)
## Summerizing null values again
#data.isnull().sum()
## removing rows with a nan values. Okay, to do because the NaNs in the features are more likely 100 than 1000
#data2 = data.dropna()
#
## See if it works
#data2.isnull().sum()
#
## Looking on data types
#data2.dtypes
## Removing the object features. Maybe, we will onehot-encode them later
#data3 = data2.drop(['field', 'from', 'career'], axis=1)
## Make a heatmap
#plt.subplots(figsize=(20,15))
#ax = plt.axes()
#ax.set_title("Correlation Heatmap")
#corr = data3.corr()
#sns.heatmap(corr,xticklabels=corr.columns.values,yticklabels=corr.columns.values)
#%%
#Alternative way to filter away columns with to many NaN values.
#Preserve shar value
from sklearn.impute import SimpleImputer
imputer = SimpleImputer(strategy='mean')
imputer.fit(raw_data[["shar", "shar_o"]])
shar = pd.DataFrame(imputer.transform(raw_data[["shar", "shar_o"]]), columns=["shar", "shar_o"], index=raw_data.index)
raw_data = replaceGroup(raw_data, shar)
null_sum = raw_data.isnull().sum()
too_many_nans = null_sum[null_sum < 750].index.values
too_many_nans = [str(index) for index in too_many_nans]
data = raw_data[too_many_nans]
data = data.dropna()
data = data.drop(["field", "from", "career"], axis=1)
#%%One hot encoding
data = data[data.columns.drop(list(data.filter(regex="_3")))]
data.drop(["gender", "race_o", "race", "field_cd"], axis=1)
field_1hot = pd.get_dummies(data['field_cd'], prefix= 'field') #Encode fields
data = data.drop('field_cd', axis=1)
data = pd.concat([data, field_1hot], axis=1)
race_1hot = pd.get_dummies(data['race'], prefix='race')
data = data.drop('race', axis=1)
data = pd.concat([data, race_1hot], axis=1)
goal_1hot = pd.get_dummies(data['goal'], prefix='goal')
data = data.drop('goal', axis=1)
data = pd.concat([data, goal_1hot], axis=1)
date = data['date']
date = np.abs(8 - date)
data = data.drop('date', axis=1)
data = pd.concat([data, date], axis=1)
go_out = data['go_out']
go_out = np.abs(8-go_out)
data = data.drop('go_out', axis=1)
data = pd.concat([data, go_out], axis=1)
#%%
round_1_1 = ['attr1_1', "sinc1_1", "intel1_1", "fun1_1", "amb1_1", "shar1_1"]
columnsToScale = data[round_1_1]
scaledColumns = scaleGroup(columnsToScale, 100)
data = replaceGroup(data, scaledColumns)
round_2_1 = ['attr2_1', "sinc2_1", "intel2_1", "fun2_1", "amb2_1", "shar2_1"]
columnsToScale = data[round_2_1]
scaledColumns = scaleGroup(columnsToScale, 100)
data = replaceGroup(data, scaledColumns)
#%%Correlation bewteen what you see as important vs how you rate the other person and if this correlates to a match
self_look_for_before = data[['attr1_1', 'sinc1_1', 'intel1_1', 'fun1_1', 'amb1_1', 'shar1_1']]
date_score = data[['attr', 'sinc', 'intel', 'fun', 'amb', 'shar']]
diff_before = self_look_for_before.values - date_score
#diff_after_1 = self_look_for_after_date_1.values - date_score
def calcLength(row):
return (row.values ** 2).mean() ** .5
lookfor_before_vs_datescore_diff = diff_before.apply(calcLength, axis=1)
#lookfor_after1_vs_datescore_diff = diff_after_1.apply(calcLength, axis=1)
#Invert scaling
lookfor_before_vs_datescore_diff = 100 - lookfor_before_vs_datescore_diff
lookfor_before_vs_datescore_diff.name = "lookfor_before_vs_datescore_diff"
data = pd.concat([data, pd.DataFrame(lookfor_before_vs_datescore_diff)], axis=1)
#%%
self_look_for_before = data[['attr1_1', 'sinc1_1', 'intel1_1', 'fun1_1', 'amb1_1', 'shar1_1']]
date_score = data[['attr', 'sinc', 'intel', 'fun', 'amb', 'shar']]
scaled_date_score = date_score * self_look_for_before.values
scaled_date_score.columns = [ s + '_s' for s in ['attr', 'sinc', 'intel', 'fun', 'amb', 'shar']]
data = pd.concat([data, scaled_date_score], axis=1)
data_algorithm = data.drop(['match', 'iid', "id","idg", "condtn", "wave", "round", "position", "partner", "pid", "career_c", "sports", "tvsports", 'exercise', 'dining', 'museums', 'art', 'hiking', 'gaming', 'clubbing','reading', 'tv', 'theater', 'movies','concerts', 'music', 'shopping', 'yoga'], axis=1)
data1 = data_algorithm.drop(list(data_algorithm.filter(regex="field")), axis=1)
data1 = data1.drop(list(data1.filter(regex="goal")), axis=1)
data1 = data1.drop(list(data1.filter(regex="_o")), axis=1)
data1 = data1.drop(list(data1.filter(regex="race")), axis=1)
corr = data1.corr()
corr_dec = corr['dec'].sort_values(ascending=False)
plt.subplots(figsize=(20,15))
ax = plt.axes()
ax.set_title("Correlation Heatmap")
sns.heatmap(corr,xticklabels=corr.columns.values,yticklabels=corr.columns.values)
#%% Check to see what the different genders value most on paper
from Plots import PlotBarSeries
male_rows = data[data['gender'] == 1]
female_rows = data[data["gender"] == 0]
male_avg = male_rows.mean()
female_avg = female_rows.mean()
self_look_for_before_average_male = male_avg[['attr1_1', 'sinc1_1', 'intel1_1', 'fun1_1', 'amb1_1', 'shar1_1']]
self_look_for_before_average_female = female_avg[['attr1_1', 'sinc1_1', 'intel1_1', 'fun1_1', 'amb1_1', 'shar1_1']]
dataframe = pd.concat([self_look_for_before_average_male, self_look_for_before_average_female],axis=1).T
dataframe.index = ["male", "female"]
PlotBarSeries(dataframe, "Mean value","Attribute value mean by gender (round 1_1)", [0,30])
#%% Mean values by attribute for dec = 1
all_dec_1_male = data[(data["dec"] == 1) & (data["gender"] == 1)]
all_dec_1_female = data[(data["dec"] == 1) & (data["gender"] == 0)]
attrs = ['attr', 'sinc', 'intel', 'fun', 'amb', 'shar', 'like']
male_attrs_dec_avg = all_dec_1_male[attrs]
female_attrs_dec_avg = all_dec_1_female[attrs]
dataframe = pd.concat([male_attrs_dec_avg.mean(), female_attrs_dec_avg.mean()], axis=1).T
dataframe.index = ["male", "female"]
PlotBarSeries(dataframe, "Score mean value", "Date score mean value for dec=1", [0,10])
all_dec_0_male = data[(data["dec"] == 0) & (data["gender"] == 1)]
all_dec_0_female = data[(data["dec"] == 0) & (data["gender"] == 0)]
male_attrs_dec_avg = all_dec_0_male[attrs]
female_attrs_dec_avg = all_dec_0_female[attrs]
dataframe = pd.concat([male_attrs_dec_avg.mean(), female_attrs_dec_avg.mean()], axis=1).T
dataframe.index = ["male", "female"]
PlotBarSeries(dataframe, "Score mean value", "Date score mean value for dec=0", [0,10])
#%%Yes vs no scores
from Plots import PlotHeatmap
dec_yes = data[data["dec"] == 1]
dec_no = data[data["dec"] == 0]
dec_yes_attr = dec_yes[attrs]
dec_no_attr = dec_no[attrs]
dec_yes_mean = pd.DataFrame(dec_yes_attr.mean()).T
dec_yes_mean.index = ["Yes"]
dec_no_mean = pd.DataFrame(dec_no_attr.mean()).T
dec_no_mean.index = ["No"]
df = pd.concat([dec_yes_mean, dec_no_mean], axis=0)
PlotBarSeries(df, "Mean rating", "Mean rating of partner for yes and no", [0,10])
corr_yes = dec_yes_attr.corr()
corr_no = dec_no_attr.corr()
PlotHeatmap(corr_yes, "Yes", 0, 1)
PlotHeatmap(corr_no, "No", 0, 1)
#%% Check to see if you can predict your own score accurately. Which score predicts better? Prior or during the speed dating event?
attrs = ['attr', 'sinc', 'intel', 'fun', 'amb']
diff_scores = pd.DataFrame()
for i in np.arange(552):
#Get rows for current participant
rows = data[data["iid"] == i]
self_score_1 = pd.DataFrame(rows[list(rows.filter(regex="3_1"))].mean()).T
my_score = pd.DataFrame(rows[[attr + "_o" for attr in attrs]].mean()).T
self_score_diff = self_score_1 - my_score.values
result = pd.concat([self_score_diff], axis=1)
diff_scores = pd.concat([diff_scores, result], axis=0)
diff_scores_mean = diff_scores.mean()
df = pd.DataFrame([diff_scores_mean])
df.index = ["3_1"]
df.columns = attrs
PlotBarSeries(df, "Mean difference", "Attribute score prediction for 3_1", [0,2.5])
#%% Does int_corr correlate with shar? or are we not able to evaluate shared interests in 4 minutes?
int_corr = data[["int_corr"]]
shar = data[["shar", "shar_o"]]
df = pd.concat([int_corr, shar], axis=1)
corr = df.corr()
PlotHeatmap(corr, "Shared interests", 0, 1)
#%%How good are we at predicting the other persons answer
dec_yes_other = data[data["dec_o"] == 1]
dec_no_other = data[data["dec_o"] == 0]
dec_yes = data[data["dec"] == 1]
dec_no = data[data["dec"] == 0]
dec_yes_other_prob_mean = dec_yes_other["prob"].mean()
dec_no_other_prob_mean = dec_no_other["prob"].mean()
dec_yes_prob_mean = dec_yes["prob"].mean()
dec_no_prob_mean = dec_no["prob"].mean()
df = pd.DataFrame([[dec_yes_prob_mean, dec_yes_other_prob_mean], [dec_no_prob_mean, dec_no_other_prob_mean]], columns=["Own decision", "Other decision"], index=["Yes", "No"])
PlotBarSeries(df,"prob mean-value" ,"Average probability for other say yes for dec=yes and dec=no")