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
|
"""
====================================
Outlier detection on a real data set
====================================
This example illustrates the need for robust covariance estimation
on a real data set. It is useful both for outlier detection and for
a better understanding of the data structure.
We selected two sets of two variables from the Wine data set
as an illustration of what kind of analysis can be done with several
outlier detection tools. For the purpose of visualization, we are working
with two-dimensional examples, but one should be aware that things are
not so trivial in high-dimension, as it will be pointed out.
In both examples below, the main result is that the empirical covariance
estimate, as a non-robust one, is highly influenced by the heterogeneous
structure of the observations. Although the robust covariance estimate is
able to focus on the main mode of the data distribution, it sticks to the
assumption that the data should be Gaussian distributed, yielding some biased
estimation of the data structure, but yet accurate to some extent.
The One-Class SVM does not assume any parametric form of the data distribution
and can therefore model the complex shape of the data much better.
First example
-------------
The first example illustrates how the Minimum Covariance Determinant
robust estimator can help concentrate on a relevant cluster when outlying
points exist. Here the empirical covariance estimation is skewed by points
outside of the main cluster. Of course, some screening tools would have pointed
out the presence of two clusters (Support Vector Machines, Gaussian Mixture
Models, univariate outlier detection, ...). But had it been a high-dimensional
example, none of these could be applied that easily.
"""
# Author: Virgile Fritsch <virgile.fritsch@inria.fr>
# License: BSD 3 clause
import numpy as np
from sklearn.covariance import EllipticEnvelope
from sklearn.svm import OneClassSVM
import matplotlib.pyplot as plt
import matplotlib.font_manager
from sklearn.datasets import load_wine
# Define "classifiers" to be used
classifiers = {
"Empirical Covariance": EllipticEnvelope(support_fraction=1.0, contamination=0.25),
"Robust Covariance (Minimum Covariance Determinant)": EllipticEnvelope(
contamination=0.25
),
"OCSVM": OneClassSVM(nu=0.25, gamma=0.35),
}
colors = ["m", "g", "b"]
legend1 = {}
legend2 = {}
# Get data
X1 = load_wine()["data"][:, [1, 2]] # two clusters
# Learn a frontier for outlier detection with several classifiers
xx1, yy1 = np.meshgrid(np.linspace(0, 6, 500), np.linspace(1, 4.5, 500))
for i, (clf_name, clf) in enumerate(classifiers.items()):
plt.figure(1)
clf.fit(X1)
Z1 = clf.decision_function(np.c_[xx1.ravel(), yy1.ravel()])
Z1 = Z1.reshape(xx1.shape)
legend1[clf_name] = plt.contour(
xx1, yy1, Z1, levels=[0], linewidths=2, colors=colors[i]
)
legend1_values_list = list(legend1.values())
legend1_keys_list = list(legend1.keys())
# Plot the results (= shape of the data points cloud)
plt.figure(1) # two clusters
plt.title("Outlier detection on a real data set (wine recognition)")
plt.scatter(X1[:, 0], X1[:, 1], color="black")
bbox_args = dict(boxstyle="round", fc="0.8")
arrow_args = dict(arrowstyle="->")
plt.annotate(
"outlying points",
xy=(4, 2),
xycoords="data",
textcoords="data",
xytext=(3, 1.25),
bbox=bbox_args,
arrowprops=arrow_args,
)
plt.xlim((xx1.min(), xx1.max()))
plt.ylim((yy1.min(), yy1.max()))
plt.legend(
(
legend1_values_list[0].collections[0],
legend1_values_list[1].collections[0],
legend1_values_list[2].collections[0],
),
(legend1_keys_list[0], legend1_keys_list[1], legend1_keys_list[2]),
loc="upper center",
prop=matplotlib.font_manager.FontProperties(size=11),
)
plt.ylabel("ash")
plt.xlabel("malic_acid")
plt.show()
# %%
# Second example
# --------------
# The second example shows the ability of the Minimum Covariance Determinant
# robust estimator of covariance to concentrate on the main mode of the data
# distribution: the location seems to be well estimated, although the
# covariance is hard to estimate due to the banana-shaped distribution. Anyway,
# we can get rid of some outlying observations. The One-Class SVM is able to
# capture the real data structure, but the difficulty is to adjust its kernel
# bandwidth parameter so as to obtain a good compromise between the shape of
# the data scatter matrix and the risk of over-fitting the data.
# Get data
X2 = load_wine()["data"][:, [6, 9]] # "banana"-shaped
# Learn a frontier for outlier detection with several classifiers
xx2, yy2 = np.meshgrid(np.linspace(-1, 5.5, 500), np.linspace(-2.5, 19, 500))
for i, (clf_name, clf) in enumerate(classifiers.items()):
plt.figure(2)
clf.fit(X2)
Z2 = clf.decision_function(np.c_[xx2.ravel(), yy2.ravel()])
Z2 = Z2.reshape(xx2.shape)
legend2[clf_name] = plt.contour(
xx2, yy2, Z2, levels=[0], linewidths=2, colors=colors[i]
)
legend2_values_list = list(legend2.values())
legend2_keys_list = list(legend2.keys())
# Plot the results (= shape of the data points cloud)
plt.figure(2) # "banana" shape
plt.title("Outlier detection on a real data set (wine recognition)")
plt.scatter(X2[:, 0], X2[:, 1], color="black")
plt.xlim((xx2.min(), xx2.max()))
plt.ylim((yy2.min(), yy2.max()))
plt.legend(
(
legend2_values_list[0].collections[0],
legend2_values_list[1].collections[0],
legend2_values_list[2].collections[0],
),
(legend2_keys_list[0], legend2_keys_list[1], legend2_keys_list[2]),
loc="upper center",
prop=matplotlib.font_manager.FontProperties(size=11),
)
plt.ylabel("color_intensity")
plt.xlabel("flavanoids")
plt.show()
|
