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
|
/*
* Licensed to Elasticsearch under one or more contributor
* license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright
* ownership. Elasticsearch licenses this file to you under
* the Apache License, Version 2.0 (the "License"); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package org.elasticsearch.search.aggregations.bucket.significant.heuristics;
import org.elasticsearch.ElasticsearchParseException;
import org.elasticsearch.common.ParseField;
import org.elasticsearch.common.io.stream.StreamInput;
import org.elasticsearch.common.io.stream.StreamOutput;
import org.elasticsearch.common.xcontent.XContentBuilder;
import org.elasticsearch.common.xcontent.XContentParser;
import org.elasticsearch.index.query.QueryParseContext;
import org.elasticsearch.index.query.QueryShardException;
import java.io.IOException;
public abstract class NXYSignificanceHeuristic extends SignificanceHeuristic {
protected static final ParseField BACKGROUND_IS_SUPERSET = new ParseField("background_is_superset");
protected static final ParseField INCLUDE_NEGATIVES_FIELD = new ParseField("include_negatives");
protected static final String SCORE_ERROR_MESSAGE = ", does your background filter not include all documents in the bucket? If so and it is intentional, set \"" + BACKGROUND_IS_SUPERSET.getPreferredName() + "\": false";
protected final boolean backgroundIsSuperset;
/**
* Some heuristics do not differentiate between terms that are descriptive for subset or for
* the background without the subset. We might want to filter out the terms that are appear much less often
* in the subset than in the background without the subset.
*/
protected final boolean includeNegatives;
protected NXYSignificanceHeuristic(boolean includeNegatives, boolean backgroundIsSuperset) {
this.includeNegatives = includeNegatives;
this.backgroundIsSuperset = backgroundIsSuperset;
}
/**
* Read from a stream.
*/
protected NXYSignificanceHeuristic(StreamInput in) throws IOException {
includeNegatives = in.readBoolean();
backgroundIsSuperset = in.readBoolean();
}
@Override
public void writeTo(StreamOutput out) throws IOException {
out.writeBoolean(includeNegatives);
out.writeBoolean(backgroundIsSuperset);
}
@Override
public boolean equals(Object other) {
return ((NXYSignificanceHeuristic) other).includeNegatives == includeNegatives && ((NXYSignificanceHeuristic) other).backgroundIsSuperset == backgroundIsSuperset;
}
@Override
public int hashCode() {
int result = (includeNegatives ? 1 : 0);
result = 31 * result + (backgroundIsSuperset ? 1 : 0);
return result;
}
protected static class Frequencies {
double N00, N01, N10, N11, N0_, N1_, N_0, N_1, N;
}
protected Frequencies computeNxys(long subsetFreq, long subsetSize, long supersetFreq, long supersetSize, String scoreFunctionName) {
checkFrequencies(subsetFreq, subsetSize, supersetFreq, supersetSize, scoreFunctionName);
Frequencies frequencies = new Frequencies();
if (backgroundIsSuperset) {
//documents not in class and do not contain term
frequencies.N00 = supersetSize - supersetFreq - (subsetSize - subsetFreq);
//documents in class and do not contain term
frequencies.N01 = (subsetSize - subsetFreq);
// documents not in class and do contain term
frequencies.N10 = supersetFreq - subsetFreq;
// documents in class and do contain term
frequencies.N11 = subsetFreq;
//documents that do not contain term
frequencies.N0_ = supersetSize - supersetFreq;
//documents that contain term
frequencies.N1_ = supersetFreq;
//documents that are not in class
frequencies.N_0 = supersetSize - subsetSize;
//documents that are in class
frequencies.N_1 = subsetSize;
//all docs
frequencies.N = supersetSize;
} else {
//documents not in class and do not contain term
frequencies.N00 = supersetSize - supersetFreq;
//documents in class and do not contain term
frequencies.N01 = subsetSize - subsetFreq;
// documents not in class and do contain term
frequencies.N10 = supersetFreq;
// documents in class and do contain term
frequencies.N11 = subsetFreq;
//documents that do not contain term
frequencies.N0_ = supersetSize - supersetFreq + subsetSize - subsetFreq;
//documents that contain term
frequencies.N1_ = supersetFreq + subsetFreq;
//documents that are not in class
frequencies.N_0 = supersetSize;
//documents that are in class
frequencies.N_1 = subsetSize;
//all docs
frequencies.N = supersetSize + subsetSize;
}
return frequencies;
}
protected void checkFrequencies(long subsetFreq, long subsetSize, long supersetFreq, long supersetSize, String scoreFunctionName) {
checkFrequencyValidity(subsetFreq, subsetSize, supersetFreq, supersetSize, scoreFunctionName);
if (backgroundIsSuperset) {
if (subsetFreq > supersetFreq) {
throw new IllegalArgumentException("subsetFreq > supersetFreq" + SCORE_ERROR_MESSAGE);
}
if (subsetSize > supersetSize) {
throw new IllegalArgumentException("subsetSize > supersetSize" + SCORE_ERROR_MESSAGE);
}
if (supersetFreq - subsetFreq > supersetSize - subsetSize) {
throw new IllegalArgumentException("supersetFreq - subsetFreq > supersetSize - subsetSize" + SCORE_ERROR_MESSAGE);
}
}
}
protected void build(XContentBuilder builder) throws IOException {
builder.field(INCLUDE_NEGATIVES_FIELD.getPreferredName(), includeNegatives).field(BACKGROUND_IS_SUPERSET.getPreferredName(),
backgroundIsSuperset);
}
public abstract static class NXYParser implements SignificanceHeuristicParser {
@Override
public SignificanceHeuristic parse(QueryParseContext context)
throws IOException, QueryShardException {
XContentParser parser = context.parser();
String givenName = parser.currentName();
boolean includeNegatives = false;
boolean backgroundIsSuperset = true;
XContentParser.Token token = parser.nextToken();
while (!token.equals(XContentParser.Token.END_OBJECT)) {
if (INCLUDE_NEGATIVES_FIELD.match(parser.currentName())) {
parser.nextToken();
includeNegatives = parser.booleanValue();
} else if (BACKGROUND_IS_SUPERSET.match(parser.currentName())) {
parser.nextToken();
backgroundIsSuperset = parser.booleanValue();
} else {
throw new ElasticsearchParseException("failed to parse [{}] significance heuristic. unknown field [{}]", givenName, parser.currentName());
}
token = parser.nextToken();
}
return newHeuristic(includeNegatives, backgroundIsSuperset);
}
protected abstract SignificanceHeuristic newHeuristic(boolean includeNegatives, boolean backgroundIsSuperset);
}
protected abstract static class NXYBuilder implements SignificanceHeuristicBuilder {
protected boolean includeNegatives = true;
protected boolean backgroundIsSuperset = true;
public NXYBuilder(boolean includeNegatives, boolean backgroundIsSuperset) {
this.includeNegatives = includeNegatives;
this.backgroundIsSuperset = backgroundIsSuperset;
}
protected void build(XContentBuilder builder) throws IOException {
builder.field(INCLUDE_NEGATIVES_FIELD.getPreferredName(), includeNegatives)
.field(BACKGROUND_IS_SUPERSET.getPreferredName(), backgroundIsSuperset);
}
}
}
|
