[1256] | 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.tasktrees.temporalrelation; |
---|
| 16 | |
---|
| 17 | import java.util.HashMap; |
---|
| 18 | import java.util.LinkedList; |
---|
| 19 | import java.util.List; |
---|
| 20 | import java.util.Map; |
---|
| 21 | |
---|
| 22 | import de.ugoe.cs.autoquest.tasktrees.treeifc.ITaskInstance; |
---|
| 23 | import de.ugoe.cs.autoquest.tasktrees.treeifc.IUserSession; |
---|
| 24 | import de.ugoe.cs.autoquest.usageprofiles.SymbolMap; |
---|
| 25 | import de.ugoe.cs.autoquest.usageprofiles.Trie; |
---|
| 26 | import de.ugoe.cs.autoquest.usageprofiles.TrieProcessor; |
---|
| 27 | |
---|
| 28 | /** |
---|
| 29 | * <p> |
---|
| 30 | * TODO comment |
---|
| 31 | * </p> |
---|
| 32 | * |
---|
| 33 | * @author Patrick Harms |
---|
| 34 | */ |
---|
| 35 | public class TaskInstanceTrie extends Trie<ITaskInstance> { |
---|
| 36 | |
---|
| 37 | /** */ |
---|
| 38 | private static final long serialVersionUID = 1L; |
---|
| 39 | |
---|
| 40 | /** |
---|
| 41 | * <p> |
---|
| 42 | * the task comparator to be used for comparing tasks |
---|
| 43 | * </p> |
---|
| 44 | */ |
---|
| 45 | private TaskComparator comparator; |
---|
| 46 | |
---|
| 47 | /** |
---|
| 48 | * <p> |
---|
| 49 | * TODO: comment |
---|
| 50 | * </p> |
---|
| 51 | * |
---|
| 52 | * @param taskComparator2 |
---|
| 53 | */ |
---|
| 54 | public TaskInstanceTrie(TaskComparator taskComparator) { |
---|
| 55 | super(taskComparator); |
---|
| 56 | this.comparator = taskComparator; |
---|
| 57 | } |
---|
| 58 | |
---|
| 59 | /** |
---|
| 60 | * |
---|
| 61 | */ |
---|
| 62 | public void trainSessions(List<IUserSession> userSessions, int maxOrder) { |
---|
| 63 | if (maxOrder < 1) { |
---|
| 64 | return; |
---|
| 65 | } |
---|
| 66 | |
---|
| 67 | SymbolMap<ITaskInstance, List<ITaskInstance>> equalTaskInstancesMap = |
---|
| 68 | new SymbolMap<ITaskInstance, List<ITaskInstance>>(comparator); |
---|
| 69 | |
---|
| 70 | Map<ITaskInstance, List<ITaskInstance>> taskInstancesMap = |
---|
| 71 | new HashMap<ITaskInstance, List<ITaskInstance>>(); |
---|
| 72 | |
---|
| 73 | System.out.println("preparing training"); |
---|
| 74 | for (IUserSession session : userSessions) { |
---|
| 75 | for (ITaskInstance taskInstance : session) { |
---|
| 76 | List<ITaskInstance> equalTaskInstances = |
---|
| 77 | equalTaskInstancesMap.getValue(taskInstance); |
---|
| 78 | |
---|
| 79 | if (equalTaskInstances == null) { |
---|
| 80 | equalTaskInstances = new LinkedList<ITaskInstance>(); |
---|
| 81 | equalTaskInstancesMap.addSymbol(taskInstance, equalTaskInstances); |
---|
| 82 | } |
---|
| 83 | |
---|
| 84 | equalTaskInstances.add(taskInstance); |
---|
| 85 | taskInstancesMap.put(taskInstance, equalTaskInstances); |
---|
| 86 | } |
---|
| 87 | } |
---|
| 88 | |
---|
| 89 | System.out.println("performing training"); |
---|
| 90 | for (IUserSession session : userSessions) { |
---|
| 91 | train(session, maxOrder, taskInstancesMap); |
---|
| 92 | } |
---|
| 93 | |
---|
| 94 | updateKnownSymbols(); |
---|
| 95 | } |
---|
| 96 | |
---|
[1267] | 97 | /* (non-Javadoc) |
---|
| 98 | * @see de.ugoe.cs.autoquest.usageprofiles.Trie#equals(java.lang.Object) |
---|
| 99 | */ |
---|
| 100 | @Override |
---|
| 101 | public boolean equals(Object other) { |
---|
| 102 | if (this == other) { |
---|
| 103 | return true; |
---|
| 104 | } |
---|
| 105 | else if (other instanceof TaskInstanceTrie) { |
---|
| 106 | return super.equals(other); |
---|
| 107 | } |
---|
| 108 | else { |
---|
| 109 | return false; |
---|
| 110 | } |
---|
| 111 | } |
---|
| 112 | |
---|
[1273] | 113 | /* (non-Javadoc) |
---|
| 114 | * @see de.ugoe.cs.autoquest.usageprofiles.Trie#hashCode() |
---|
| 115 | */ |
---|
| 116 | @Override |
---|
| 117 | public int hashCode() { |
---|
| 118 | return super.hashCode(); |
---|
| 119 | } |
---|
| 120 | |
---|
[1256] | 121 | /** |
---|
| 122 | * |
---|
| 123 | */ |
---|
| 124 | private void train(IUserSession userSession, |
---|
| 125 | int maxOrder, |
---|
| 126 | Map<ITaskInstance, List<ITaskInstance>> taskInstancesMap) |
---|
| 127 | { |
---|
| 128 | List<ITaskInstance> executedTasks = userSession.getExecutedTasks(); |
---|
| 129 | |
---|
| 130 | List<ITaskInstance> subsequence = new LinkedList<ITaskInstance>(); |
---|
| 131 | |
---|
| 132 | int numberOfTrainedSubsequences = 0; |
---|
| 133 | int sequenceMaxCount = 0; |
---|
| 134 | |
---|
| 135 | for (ITaskInstance currentTaskInstance : executedTasks) { |
---|
| 136 | int occurrenceCount = taskInstancesMap.get(currentTaskInstance).size(); |
---|
| 137 | |
---|
| 138 | if ((numberOfTrainedSubsequences % 10) == 0) { |
---|
| 139 | sequenceMaxCount = getCurrentSequenceMaxCount(); |
---|
| 140 | } |
---|
| 141 | |
---|
| 142 | if (occurrenceCount < sequenceMaxCount) { |
---|
| 143 | // this task instance does not need to be considered, as it occurs not often enough |
---|
| 144 | // to be part of a sequence, that occurs most often. Therefore train all remaining |
---|
| 145 | // sequences so far and go on, until the next useful sequence is found. |
---|
| 146 | |
---|
| 147 | while (subsequence.size() > 1) { |
---|
| 148 | add(subsequence); |
---|
| 149 | subsequence.remove(0); |
---|
| 150 | } |
---|
| 151 | |
---|
| 152 | subsequence.clear(); |
---|
| 153 | } |
---|
| 154 | else { |
---|
| 155 | subsequence.add(currentTaskInstance); |
---|
| 156 | |
---|
| 157 | if (subsequence.size() == maxOrder) { |
---|
| 158 | add(subsequence); |
---|
| 159 | subsequence.remove(0); |
---|
| 160 | |
---|
| 161 | numberOfTrainedSubsequences++; |
---|
| 162 | } |
---|
| 163 | } |
---|
| 164 | } |
---|
| 165 | |
---|
| 166 | // add shorter remaining subsequences, if any |
---|
| 167 | while (subsequence.size() > 1) { |
---|
| 168 | add(subsequence); |
---|
| 169 | subsequence.remove(0); |
---|
| 170 | } |
---|
| 171 | } |
---|
| 172 | |
---|
| 173 | /** |
---|
| 174 | * <p> |
---|
| 175 | * TODO: comment |
---|
| 176 | * </p> |
---|
| 177 | * |
---|
| 178 | * @return |
---|
| 179 | */ |
---|
| 180 | private int getCurrentSequenceMaxCount() { |
---|
| 181 | MaxSequenceCountFinder processor = new MaxSequenceCountFinder(); |
---|
| 182 | super.process(processor); |
---|
| 183 | return processor.getMaxCount(); |
---|
| 184 | } |
---|
| 185 | |
---|
| 186 | /** |
---|
| 187 | * @author Patrick Harms |
---|
| 188 | */ |
---|
[1267] | 189 | private static class MaxSequenceCountFinder implements TrieProcessor<ITaskInstance> { |
---|
[1256] | 190 | |
---|
| 191 | /** |
---|
| 192 | * |
---|
| 193 | */ |
---|
| 194 | private int currentCount = 0; |
---|
| 195 | |
---|
| 196 | /* (non-Javadoc) |
---|
| 197 | * @see de.ugoe.cs.autoquest.usageprofiles.TrieProcessor#process(java.util.List, int) |
---|
| 198 | */ |
---|
| 199 | @Override |
---|
| 200 | public TrieProcessor.Result process(List<ITaskInstance> foundTask, int count) { |
---|
| 201 | if (foundTask.size() == 2) { |
---|
| 202 | this.currentCount = Math.max(this.currentCount, count); |
---|
| 203 | |
---|
| 204 | // do not consider children |
---|
| 205 | return TrieProcessor.Result.SKIP_NODE; |
---|
| 206 | } |
---|
| 207 | else { |
---|
| 208 | return TrieProcessor.Result.CONTINUE; |
---|
| 209 | } |
---|
| 210 | } |
---|
| 211 | |
---|
| 212 | /** |
---|
| 213 | * |
---|
| 214 | */ |
---|
| 215 | private int getMaxCount() { |
---|
| 216 | return currentCount; |
---|
| 217 | } |
---|
| 218 | |
---|
| 219 | } |
---|
| 220 | |
---|
| 221 | } |
---|