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
|
package de.lmu.ifi.dbs.elki.utilities.scaling.outlier;
/*
This file is part of ELKI:
Environment for Developing KDD-Applications Supported by Index-Structures
Copyright (C) 2015
Ludwig-Maximilians-Universität München
Lehr- und Forschungseinheit für Datenbanksysteme
ELKI Development Team
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
import de.lmu.ifi.dbs.elki.database.ids.DBIDIter;
import de.lmu.ifi.dbs.elki.logging.LoggingUtil;
import de.lmu.ifi.dbs.elki.result.outlier.OutlierResult;
import de.lmu.ifi.dbs.elki.utilities.datastructures.QuickSelect;
import de.lmu.ifi.dbs.elki.utilities.datastructures.arraylike.ArrayLikeUtil;
import de.lmu.ifi.dbs.elki.utilities.datastructures.arraylike.NumberArrayAdapter;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.AbstractParameterizer;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.OptionID;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.constraints.CommonConstraints;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.parameterization.Parameterization;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.parameters.Flag;
import de.lmu.ifi.dbs.elki.utilities.optionhandling.parameters.IntParameter;
/**
* Outlier scaling function that only keeps the top k outliers.
*
* @author Erich Schubert
*/
public class TopKOutlierScaling implements OutlierScalingFunction {
/**
* Parameter to specify the number of outliers to keep
* <p>
* Key: {@code -topk.k}
* </p>
*/
public static final OptionID K_ID = new OptionID("topk.k", "Number of outliers to keep.");
/**
* Parameter to specify the lambda value
* <p>
* Key: {@code -topk.binary}
* </p>
*/
public static final OptionID BINARY_ID = new OptionID("topk.binary", "Make the top k a binary scaling.");
/**
* Number of outliers to keep.
*/
private int k = -1;
/**
* Do a binary decision
*/
private boolean binary = false;
/**
* The value we cut off at.
*/
private double cutoff;
/**
* The "ground" value
*/
private double ground;
/**
* The maximum value
*/
private double max;
/**
* Constructor.
*
* @param k
* @param binary
*/
public TopKOutlierScaling(int k, boolean binary) {
super();
this.k = k;
this.binary = binary;
}
@Override
public void prepare(OutlierResult or) {
if(k <= 0) {
LoggingUtil.warning("No k configured for Top-k outlier scaling!");
}
DBIDIter order = or.getOrdering().order(or.getOrdering().getDBIDs()).iter();
for(int i = 0; i < k && order.valid(); i++, order.advance()) {
cutoff = or.getScores().doubleValue(order);
}
max = or.getOutlierMeta().getActualMaximum();
ground = or.getOutlierMeta().getTheoreticalBaseline();
if(Double.isInfinite(ground) || Double.isNaN(ground)) {
ground = or.getOutlierMeta().getTheoreticalMinimum();
}
if(Double.isInfinite(ground) || Double.isNaN(ground)) {
ground = or.getOutlierMeta().getActualMinimum();
}
if(Double.isInfinite(ground) || Double.isNaN(ground)) {
ground = Math.min(0.0, cutoff);
}
}
@Override
public <A> void prepare(A array, NumberArrayAdapter<?, A> adapter) {
if(k <= 0) {
LoggingUtil.warning("No k configured for Top-k outlier scaling!");
}
double[] scores = ArrayLikeUtil.toPrimitiveDoubleArray(array, adapter);
QuickSelect.quickSelect(scores, k);
cutoff = scores[k - 1];
max = Double.NEGATIVE_INFINITY;
for(double v : scores) {
max = Math.max(max, v);
}
ground = Math.min(0.0, cutoff);
}
@Override
public double getMax() {
if(binary) {
return 1.0;
}
return max;
}
@Override
public double getMin() {
if(binary) {
return 0.0;
}
return ground;
}
@Override
public double getScaled(double value) {
if(binary) {
if(value >= cutoff) {
return 1;
}
else {
return 0;
}
}
else {
if(value >= cutoff) {
return (value - ground) / (max - ground);
}
else {
return 0.0;
}
}
}
/**
* Parameterization class.
*
* @author Erich Schubert
*
* @apiviz.exclude
*/
public static class Parameterizer extends AbstractParameterizer {
protected int k = 0;
protected boolean binary = false;
@Override
protected void makeOptions(Parameterization config) {
super.makeOptions(config);
IntParameter kP = new IntParameter(K_ID);
kP.addConstraint(CommonConstraints.GREATER_THAN_ONE_INT);
if(config.grab(kP)) {
k = kP.intValue();
}
Flag binaryF = new Flag(BINARY_ID);
if(config.grab(binaryF)) {
binary = binaryF.isTrue();
}
}
@Override
protected TopKOutlierScaling makeInstance() {
return new TopKOutlierScaling(k, binary);
}
}
}
|
