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.usageprofiles;
|
---|
16 |
|
---|
17 | import java.util.Collection;
|
---|
18 | import java.util.LinkedList;
|
---|
19 | import java.util.List;
|
---|
20 | import java.util.Random;
|
---|
21 |
|
---|
22 | import de.ugoe.cs.autoquest.eventcore.Event;
|
---|
23 |
|
---|
24 | /**
|
---|
25 | * <p>
|
---|
26 | * Implements an inverted high-order Markov models. The idea behind this is that the most likely events, become the most unlikely events and the most unlikely events become the most likely events.
|
---|
27 | * </p>
|
---|
28 | * <p>
|
---|
29 | * The empirical probability of an observation is N-n/(k-1)N with N the total number of observations, n the number a specific observation was seen and k the number of different observations.
|
---|
30 | *
|
---|
31 | * @author Steffen Herbold
|
---|
32 | * @version 1.0
|
---|
33 | */
|
---|
34 | public class InvertedHighOrderMarkovModel extends TrieBasedModel {
|
---|
35 |
|
---|
36 | /**
|
---|
37 | * <p>
|
---|
38 | * Id for object serialization.
|
---|
39 | * </p>
|
---|
40 | */
|
---|
41 | private static final long serialVersionUID = 1L;
|
---|
42 |
|
---|
43 | /**
|
---|
44 | * <p>
|
---|
45 | * Constructor. Creates a new HighOrderMarkovModel with a defined Markov order.
|
---|
46 | * </p>
|
---|
47 | *
|
---|
48 | * @param maxOrder
|
---|
49 | * Markov order of the model
|
---|
50 | * @param r
|
---|
51 | * random number generator used by probabilistic methods of the class
|
---|
52 | */
|
---|
53 | public InvertedHighOrderMarkovModel(int maxOrder, Random r) {
|
---|
54 | super(maxOrder, r);
|
---|
55 | }
|
---|
56 |
|
---|
57 | /**
|
---|
58 | * <p>
|
---|
59 | * Calculates the probability of the next Event being symbol based on the order of the Markov
|
---|
60 | * model. The order is defined in the constructor {@link #InvertedHighOrderMarkovModel(int, Random)}.
|
---|
61 | * </p>
|
---|
62 | *
|
---|
63 | * @see de.ugoe.cs.autoquest.usageprofiles.IStochasticProcess#getProbability(java.util.List,
|
---|
64 | * de.ugoe.cs.autoquest.eventcore.Event)
|
---|
65 | */
|
---|
66 | @Override
|
---|
67 | public double getProbability(List<Event> context, Event symbol) {
|
---|
68 | if (context == null) {
|
---|
69 | throw new IllegalArgumentException("context must not be null");
|
---|
70 | }
|
---|
71 | if (symbol == null) {
|
---|
72 | throw new IllegalArgumentException("symbol must not be null");
|
---|
73 | }
|
---|
74 | double result = 0.0d;
|
---|
75 |
|
---|
76 | List<Event> contextCopy;
|
---|
77 | if (context.size() >= trieOrder) {
|
---|
78 | contextCopy =
|
---|
79 | new LinkedList<Event>(context.subList(context.size() - trieOrder + 1,
|
---|
80 | context.size()));
|
---|
81 | }
|
---|
82 | else {
|
---|
83 | contextCopy = new LinkedList<Event>(context);
|
---|
84 | }
|
---|
85 |
|
---|
86 | Collection<Event> followers = trie.getFollowingSymbols(contextCopy);
|
---|
87 | int sumCountFollowers = 0; // N(s\sigma')
|
---|
88 | for (Event follower : followers) {
|
---|
89 | sumCountFollowers += trie.getCount(contextCopy, follower);
|
---|
90 | }
|
---|
91 |
|
---|
92 | int countSymbol = trie.getCount(contextCopy, symbol);
|
---|
93 | if (sumCountFollowers != 0) {
|
---|
94 | result = ((double) (sumCountFollowers-countSymbol)) / ((followers.size()-1)*sumCountFollowers);
|
---|
95 | }
|
---|
96 |
|
---|
97 | return result;
|
---|
98 | }
|
---|
99 |
|
---|
100 | }
|
---|