1 | // Copyright 2012 Georg-August-Universität Göttingen, Germany
|
---|
2 | //
|
---|
3 | // Licensed under the Apache License, Version 2.0 (the "License");
|
---|
4 | // you may not use this file except in compliance with the License.
|
---|
5 | // You may obtain a copy of the License at
|
---|
6 | //
|
---|
7 | // http://www.apache.org/licenses/LICENSE-2.0
|
---|
8 | //
|
---|
9 | // Unless required by applicable law or agreed to in writing, software
|
---|
10 | // distributed under the License is distributed on an "AS IS" BASIS,
|
---|
11 | // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
---|
12 | // See the License for the specific language governing permissions and
|
---|
13 | // limitations under the License.
|
---|
14 |
|
---|
15 | package de.ugoe.cs.autoquest.coverage;
|
---|
16 |
|
---|
17 | import java.util.Collection;
|
---|
18 | import java.util.List;
|
---|
19 | import java.util.Map;
|
---|
20 |
|
---|
21 | import de.ugoe.cs.autoquest.eventcore.Event;
|
---|
22 | import de.ugoe.cs.autoquest.usageprofiles.IStochasticProcess;
|
---|
23 |
|
---|
24 | /**
|
---|
25 | * <p>
|
---|
26 | * This class calculates various types of sequence coverage in relation to a stochastic process.
|
---|
27 | * </p>
|
---|
28 | *
|
---|
29 | * @author Steffen Herbold
|
---|
30 | * @version 1.0
|
---|
31 | */
|
---|
32 | public class CoverageCalculatorProcess {
|
---|
33 |
|
---|
34 | /**
|
---|
35 | * <p>
|
---|
36 | * Stochastic process that is the foundation for probabilistic coverages and coverages with
|
---|
37 | * reference to all possible sequences.
|
---|
38 | * </p>
|
---|
39 | */
|
---|
40 | private final IStochasticProcess process;
|
---|
41 |
|
---|
42 | /**
|
---|
43 | * <p>
|
---|
44 | * Sequences for which the coverage is calculated.
|
---|
45 | * </p>
|
---|
46 | */
|
---|
47 | private Collection<List<Event>> sequences;
|
---|
48 |
|
---|
49 | /**
|
---|
50 | * <p>
|
---|
51 | * Length of the subsequences in relation to which the coverage is calculated.
|
---|
52 | * </p>
|
---|
53 | */
|
---|
54 | private final int length;
|
---|
55 |
|
---|
56 | /**
|
---|
57 | * <p>
|
---|
58 | * All subsequences of {@link #length} of {@link #sequences}.
|
---|
59 | * </p>
|
---|
60 | */
|
---|
61 | private Collection<List<Event>> containedSubSeqs = null;
|
---|
62 |
|
---|
63 | /**
|
---|
64 | * <p>
|
---|
65 | * All subsequences of {@link #length} that can be generated by {@link #process}.
|
---|
66 | * </p>
|
---|
67 | */
|
---|
68 | private Collection<List<Event>> allPossibleSubSeqs = null;
|
---|
69 |
|
---|
70 | /**
|
---|
71 | * <p>
|
---|
72 | * The probabilities of all subsequences of {@link #length} according to {@link #process}.
|
---|
73 | * </p>
|
---|
74 | */
|
---|
75 | private Map<List<Event>, Double> subSeqWeights = null;
|
---|
76 |
|
---|
77 | /**
|
---|
78 | * <p>
|
---|
79 | * Constructor. Creates a new CoverageCalculatorProcess for a given stochastic process and
|
---|
80 | * generated sequences.
|
---|
81 | * </p>
|
---|
82 | *
|
---|
83 | * @param process
|
---|
84 | * stochastic process used for coverage calculations; must not be null
|
---|
85 | * @param sequences
|
---|
86 | * sequences for which the coverage is calculated; must not be null
|
---|
87 | * @param length
|
---|
88 | * length of the subsequences for which the coverage is analyzed; must be >0
|
---|
89 | * @throws IllegalArgumentException
|
---|
90 | * thrown if process or sequences is null or length less than or equal to 0
|
---|
91 | */
|
---|
92 | public CoverageCalculatorProcess(IStochasticProcess process,
|
---|
93 | Collection<List<Event>> sequences,
|
---|
94 | int length)
|
---|
95 | {
|
---|
96 | if (process == null) {
|
---|
97 | throw new IllegalArgumentException("process must not be null");
|
---|
98 | }
|
---|
99 | if (sequences == null) {
|
---|
100 | throw new IllegalArgumentException("sequences must not be null");
|
---|
101 | }
|
---|
102 | if (length <= 0) {
|
---|
103 | throw new IllegalArgumentException("length must be >0; actual value: " + length);
|
---|
104 | }
|
---|
105 | this.process = process;
|
---|
106 | this.sequences = sequences;
|
---|
107 | this.length = length;
|
---|
108 | }
|
---|
109 |
|
---|
110 | /**
|
---|
111 | * <p>
|
---|
112 | * Calculates the percentage of subsequences of length k that occur, including those that cannot
|
---|
113 | * be generated by {@link #process}.
|
---|
114 | * </p>
|
---|
115 | *
|
---|
116 | * @return coverage percentage
|
---|
117 | */
|
---|
118 | public double getCoverageAllNoWeight() {
|
---|
119 | if (containedSubSeqs == null) {
|
---|
120 | containedSubSeqs = SequenceTools.containedSubSequences(sequences, length);
|
---|
121 | }
|
---|
122 | return ((double) containedSubSeqs.size()) / SequenceTools.numSequences(process, length);
|
---|
123 | }
|
---|
124 |
|
---|
125 | /**
|
---|
126 | * <p>
|
---|
127 | * Calculates the percentage of subsequences of length k that occur and can generated by
|
---|
128 | * {@link #process}.
|
---|
129 | * </p>
|
---|
130 | *
|
---|
131 | * @return coverage percentage
|
---|
132 | */
|
---|
133 | public double getCoveragePossibleNoWeight() {
|
---|
134 | if (containedSubSeqs == null) {
|
---|
135 | containedSubSeqs = SequenceTools.containedSubSequences(sequences, length);
|
---|
136 | }
|
---|
137 | if (allPossibleSubSeqs == null) {
|
---|
138 | allPossibleSubSeqs = process.generateSequences(length);
|
---|
139 | }
|
---|
140 | return ((double) containedSubSeqs.size()) / allPossibleSubSeqs.size();
|
---|
141 | }
|
---|
142 |
|
---|
143 | /**
|
---|
144 | * <p>
|
---|
145 | * Calculates the weight of the subsequences that occur with relation to {@link #process}, i.e.,
|
---|
146 | * the mass of the subsequence probability covered by the subsequences.
|
---|
147 | * </p>
|
---|
148 | *
|
---|
149 | * @return coverage weight
|
---|
150 | */
|
---|
151 | public double getCoveragePossibleWeight() {
|
---|
152 | if (containedSubSeqs == null) {
|
---|
153 | containedSubSeqs = SequenceTools.containedSubSequences(sequences, length);
|
---|
154 | }
|
---|
155 | if (allPossibleSubSeqs == null) {
|
---|
156 | allPossibleSubSeqs = process.generateSequences(length);
|
---|
157 | }
|
---|
158 | if (subSeqWeights == null) {
|
---|
159 | subSeqWeights = SequenceTools.generateWeights(process, allPossibleSubSeqs);
|
---|
160 | }
|
---|
161 | double weight = 0.0;
|
---|
162 | for (List<Event> subSeq : containedSubSeqs) {
|
---|
163 | Double curWeight = subSeqWeights.get(subSeq);
|
---|
164 | if (curWeight != null) {
|
---|
165 | weight += curWeight;
|
---|
166 | }
|
---|
167 | }
|
---|
168 | return weight;
|
---|
169 | }
|
---|
170 |
|
---|
171 | /**
|
---|
172 | * <p>
|
---|
173 | * Returns the number of covered subsequences of length k.
|
---|
174 | * </p>
|
---|
175 | *
|
---|
176 | * @return number of covered subsequences
|
---|
177 | */
|
---|
178 | public int getNumCovered() {
|
---|
179 | if (containedSubSeqs == null) {
|
---|
180 | containedSubSeqs = SequenceTools.containedSubSequences(sequences, length);
|
---|
181 | }
|
---|
182 | return containedSubSeqs.size();
|
---|
183 | }
|
---|
184 |
|
---|
185 | /**
|
---|
186 | * <p>
|
---|
187 | * Returns the number of possible subsequences of length k according to the stochastic process.
|
---|
188 | * </p>
|
---|
189 | *
|
---|
190 | * @return number of possible subsequences
|
---|
191 | */
|
---|
192 | public int getNumPossible() {
|
---|
193 | if (allPossibleSubSeqs == null) {
|
---|
194 | allPossibleSubSeqs = process.generateSequences(length);
|
---|
195 | }
|
---|
196 | return allPossibleSubSeqs.size();
|
---|
197 | }
|
---|
198 |
|
---|
199 | /**
|
---|
200 | * <p>
|
---|
201 | * Sets a new collection of sequences for which the coverage is analyzed.
|
---|
202 | * </p>
|
---|
203 | *
|
---|
204 | * @param newSequences
|
---|
205 | * new collection of sequences
|
---|
206 | * @throws IllegalArgumentException
|
---|
207 | * thrown is newSequences is null
|
---|
208 | */
|
---|
209 | public void setSequences(Collection<List<Event>> newSequences) {
|
---|
210 | if (newSequences == null) {
|
---|
211 | throw new IllegalArgumentException("sequences must not be null");
|
---|
212 | }
|
---|
213 | this.sequences = newSequences;
|
---|
214 | containedSubSeqs = null;
|
---|
215 | }
|
---|
216 |
|
---|
217 | }
|
---|