1 | /* |
---|
2 | * This program is free software; you can redistribute it and/or modify |
---|
3 | * it under the terms of the GNU General Public License as published by |
---|
4 | * the Free Software Foundation; either version 2 of the License, or |
---|
5 | * (at your option) any later version. |
---|
6 | * |
---|
7 | * This program is distributed in the hope that it will be useful, |
---|
8 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
9 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
---|
10 | * GNU General Public License for more details. |
---|
11 | * |
---|
12 | * You should have received a copy of the GNU General Public License |
---|
13 | * along with this program; if not, write to the Free Software |
---|
14 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. |
---|
15 | */ |
---|
16 | |
---|
17 | /* |
---|
18 | * NaiveBayesSimple.java |
---|
19 | * Copyright (C) 1999 University of Waikato, Hamilton, New Zealand |
---|
20 | * |
---|
21 | */ |
---|
22 | |
---|
23 | package weka.classifiers.bayes; |
---|
24 | |
---|
25 | import weka.classifiers.Classifier; |
---|
26 | import weka.classifiers.AbstractClassifier; |
---|
27 | import weka.core.Attribute; |
---|
28 | import weka.core.Capabilities; |
---|
29 | import weka.core.Instance; |
---|
30 | import weka.core.Instances; |
---|
31 | import weka.core.RevisionUtils; |
---|
32 | import weka.core.TechnicalInformation; |
---|
33 | import weka.core.TechnicalInformationHandler; |
---|
34 | import weka.core.Utils; |
---|
35 | import weka.core.Capabilities.Capability; |
---|
36 | import weka.core.TechnicalInformation.Field; |
---|
37 | import weka.core.TechnicalInformation.Type; |
---|
38 | |
---|
39 | import java.util.Enumeration; |
---|
40 | |
---|
41 | /** |
---|
42 | <!-- globalinfo-start --> |
---|
43 | * Class for building and using a simple Naive Bayes classifier.Numeric attributes are modelled by a normal distribution.<br/> |
---|
44 | * <br/> |
---|
45 | * For more information, see<br/> |
---|
46 | * <br/> |
---|
47 | * Richard Duda, Peter Hart (1973). Pattern Classification and Scene Analysis. Wiley, New York. |
---|
48 | * <p/> |
---|
49 | <!-- globalinfo-end --> |
---|
50 | * |
---|
51 | <!-- technical-bibtex-start --> |
---|
52 | * BibTeX: |
---|
53 | * <pre> |
---|
54 | * @book{Duda1973, |
---|
55 | * address = {New York}, |
---|
56 | * author = {Richard Duda and Peter Hart}, |
---|
57 | * publisher = {Wiley}, |
---|
58 | * title = {Pattern Classification and Scene Analysis}, |
---|
59 | * year = {1973} |
---|
60 | * } |
---|
61 | * </pre> |
---|
62 | * <p/> |
---|
63 | <!-- technical-bibtex-end --> |
---|
64 | * |
---|
65 | <!-- options-start --> |
---|
66 | * Valid options are: <p/> |
---|
67 | * |
---|
68 | * <pre> -D |
---|
69 | * If set, classifier is run in debug mode and |
---|
70 | * may output additional info to the console</pre> |
---|
71 | * |
---|
72 | <!-- options-end --> |
---|
73 | * |
---|
74 | * @author Eibe Frank (eibe@cs.waikato.ac.nz) |
---|
75 | * @version $Revision: 5928 $ |
---|
76 | */ |
---|
77 | public class NaiveBayesSimple |
---|
78 | extends AbstractClassifier |
---|
79 | implements TechnicalInformationHandler { |
---|
80 | |
---|
81 | /** for serialization */ |
---|
82 | static final long serialVersionUID = -1478242251770381214L; |
---|
83 | |
---|
84 | /** All the counts for nominal attributes. */ |
---|
85 | protected double [][][] m_Counts; |
---|
86 | |
---|
87 | /** The means for numeric attributes. */ |
---|
88 | protected double [][] m_Means; |
---|
89 | |
---|
90 | /** The standard deviations for numeric attributes. */ |
---|
91 | protected double [][] m_Devs; |
---|
92 | |
---|
93 | /** The prior probabilities of the classes. */ |
---|
94 | protected double [] m_Priors; |
---|
95 | |
---|
96 | /** The instances used for training. */ |
---|
97 | protected Instances m_Instances; |
---|
98 | |
---|
99 | /** Constant for normal distribution. */ |
---|
100 | protected static double NORM_CONST = Math.sqrt(2 * Math.PI); |
---|
101 | |
---|
102 | /** |
---|
103 | * Returns a string describing this classifier |
---|
104 | * @return a description of the classifier suitable for |
---|
105 | * displaying in the explorer/experimenter gui |
---|
106 | */ |
---|
107 | public String globalInfo() { |
---|
108 | return |
---|
109 | "Class for building and using a simple Naive Bayes classifier." |
---|
110 | + "Numeric attributes are modelled by a normal distribution.\n\n" |
---|
111 | + "For more information, see\n\n" |
---|
112 | + getTechnicalInformation().toString(); |
---|
113 | } |
---|
114 | |
---|
115 | /** |
---|
116 | * Returns an instance of a TechnicalInformation object, containing |
---|
117 | * detailed information about the technical background of this class, |
---|
118 | * e.g., paper reference or book this class is based on. |
---|
119 | * |
---|
120 | * @return the technical information about this class |
---|
121 | */ |
---|
122 | public TechnicalInformation getTechnicalInformation() { |
---|
123 | TechnicalInformation result; |
---|
124 | |
---|
125 | result = new TechnicalInformation(Type.BOOK); |
---|
126 | result.setValue(Field.AUTHOR, "Richard Duda and Peter Hart"); |
---|
127 | result.setValue(Field.YEAR, "1973"); |
---|
128 | result.setValue(Field.TITLE, "Pattern Classification and Scene Analysis"); |
---|
129 | result.setValue(Field.PUBLISHER, "Wiley"); |
---|
130 | result.setValue(Field.ADDRESS, "New York"); |
---|
131 | |
---|
132 | return result; |
---|
133 | } |
---|
134 | |
---|
135 | /** |
---|
136 | * Returns default capabilities of the classifier. |
---|
137 | * |
---|
138 | * @return the capabilities of this classifier |
---|
139 | */ |
---|
140 | public Capabilities getCapabilities() { |
---|
141 | Capabilities result = super.getCapabilities(); |
---|
142 | result.disableAll(); |
---|
143 | |
---|
144 | // attributes |
---|
145 | result.enable(Capability.NOMINAL_ATTRIBUTES); |
---|
146 | result.enable(Capability.NUMERIC_ATTRIBUTES); |
---|
147 | result.enable(Capability.DATE_ATTRIBUTES); |
---|
148 | result.enable(Capability.MISSING_VALUES); |
---|
149 | |
---|
150 | // class |
---|
151 | result.enable(Capability.NOMINAL_CLASS); |
---|
152 | result.enable(Capability.MISSING_CLASS_VALUES); |
---|
153 | |
---|
154 | return result; |
---|
155 | } |
---|
156 | |
---|
157 | /** |
---|
158 | * Generates the classifier. |
---|
159 | * |
---|
160 | * @param instances set of instances serving as training data |
---|
161 | * @exception Exception if the classifier has not been generated successfully |
---|
162 | */ |
---|
163 | public void buildClassifier(Instances instances) throws Exception { |
---|
164 | |
---|
165 | int attIndex = 0; |
---|
166 | double sum; |
---|
167 | |
---|
168 | // can classifier handle the data? |
---|
169 | getCapabilities().testWithFail(instances); |
---|
170 | |
---|
171 | // remove instances with missing class |
---|
172 | instances = new Instances(instances); |
---|
173 | instances.deleteWithMissingClass(); |
---|
174 | |
---|
175 | m_Instances = new Instances(instances, 0); |
---|
176 | |
---|
177 | // Reserve space |
---|
178 | m_Counts = new double[instances.numClasses()] |
---|
179 | [instances.numAttributes() - 1][0]; |
---|
180 | m_Means = new double[instances.numClasses()] |
---|
181 | [instances.numAttributes() - 1]; |
---|
182 | m_Devs = new double[instances.numClasses()] |
---|
183 | [instances.numAttributes() - 1]; |
---|
184 | m_Priors = new double[instances.numClasses()]; |
---|
185 | Enumeration enu = instances.enumerateAttributes(); |
---|
186 | while (enu.hasMoreElements()) { |
---|
187 | Attribute attribute = (Attribute) enu.nextElement(); |
---|
188 | if (attribute.isNominal()) { |
---|
189 | for (int j = 0; j < instances.numClasses(); j++) { |
---|
190 | m_Counts[j][attIndex] = new double[attribute.numValues()]; |
---|
191 | } |
---|
192 | } else { |
---|
193 | for (int j = 0; j < instances.numClasses(); j++) { |
---|
194 | m_Counts[j][attIndex] = new double[1]; |
---|
195 | } |
---|
196 | } |
---|
197 | attIndex++; |
---|
198 | } |
---|
199 | |
---|
200 | // Compute counts and sums |
---|
201 | Enumeration enumInsts = instances.enumerateInstances(); |
---|
202 | while (enumInsts.hasMoreElements()) { |
---|
203 | Instance instance = (Instance) enumInsts.nextElement(); |
---|
204 | if (!instance.classIsMissing()) { |
---|
205 | Enumeration enumAtts = instances.enumerateAttributes(); |
---|
206 | attIndex = 0; |
---|
207 | while (enumAtts.hasMoreElements()) { |
---|
208 | Attribute attribute = (Attribute) enumAtts.nextElement(); |
---|
209 | if (!instance.isMissing(attribute)) { |
---|
210 | if (attribute.isNominal()) { |
---|
211 | m_Counts[(int)instance.classValue()][attIndex] |
---|
212 | [(int)instance.value(attribute)]++; |
---|
213 | } else { |
---|
214 | m_Means[(int)instance.classValue()][attIndex] += |
---|
215 | instance.value(attribute); |
---|
216 | m_Counts[(int)instance.classValue()][attIndex][0]++; |
---|
217 | } |
---|
218 | } |
---|
219 | attIndex++; |
---|
220 | } |
---|
221 | m_Priors[(int)instance.classValue()]++; |
---|
222 | } |
---|
223 | } |
---|
224 | |
---|
225 | // Compute means |
---|
226 | Enumeration enumAtts = instances.enumerateAttributes(); |
---|
227 | attIndex = 0; |
---|
228 | while (enumAtts.hasMoreElements()) { |
---|
229 | Attribute attribute = (Attribute) enumAtts.nextElement(); |
---|
230 | if (attribute.isNumeric()) { |
---|
231 | for (int j = 0; j < instances.numClasses(); j++) { |
---|
232 | if (m_Counts[j][attIndex][0] < 2) { |
---|
233 | throw new Exception("attribute " + attribute.name() + |
---|
234 | ": less than two values for class " + |
---|
235 | instances.classAttribute().value(j)); |
---|
236 | } |
---|
237 | m_Means[j][attIndex] /= m_Counts[j][attIndex][0]; |
---|
238 | } |
---|
239 | } |
---|
240 | attIndex++; |
---|
241 | } |
---|
242 | |
---|
243 | // Compute standard deviations |
---|
244 | enumInsts = instances.enumerateInstances(); |
---|
245 | while (enumInsts.hasMoreElements()) { |
---|
246 | Instance instance = |
---|
247 | (Instance) enumInsts.nextElement(); |
---|
248 | if (!instance.classIsMissing()) { |
---|
249 | enumAtts = instances.enumerateAttributes(); |
---|
250 | attIndex = 0; |
---|
251 | while (enumAtts.hasMoreElements()) { |
---|
252 | Attribute attribute = (Attribute) enumAtts.nextElement(); |
---|
253 | if (!instance.isMissing(attribute)) { |
---|
254 | if (attribute.isNumeric()) { |
---|
255 | m_Devs[(int)instance.classValue()][attIndex] += |
---|
256 | (m_Means[(int)instance.classValue()][attIndex]- |
---|
257 | instance.value(attribute))* |
---|
258 | (m_Means[(int)instance.classValue()][attIndex]- |
---|
259 | instance.value(attribute)); |
---|
260 | } |
---|
261 | } |
---|
262 | attIndex++; |
---|
263 | } |
---|
264 | } |
---|
265 | } |
---|
266 | enumAtts = instances.enumerateAttributes(); |
---|
267 | attIndex = 0; |
---|
268 | while (enumAtts.hasMoreElements()) { |
---|
269 | Attribute attribute = (Attribute) enumAtts.nextElement(); |
---|
270 | if (attribute.isNumeric()) { |
---|
271 | for (int j = 0; j < instances.numClasses(); j++) { |
---|
272 | if (m_Devs[j][attIndex] <= 0) { |
---|
273 | throw new Exception("attribute " + attribute.name() + |
---|
274 | ": standard deviation is 0 for class " + |
---|
275 | instances.classAttribute().value(j)); |
---|
276 | } |
---|
277 | else { |
---|
278 | m_Devs[j][attIndex] /= m_Counts[j][attIndex][0] - 1; |
---|
279 | m_Devs[j][attIndex] = Math.sqrt(m_Devs[j][attIndex]); |
---|
280 | } |
---|
281 | } |
---|
282 | } |
---|
283 | attIndex++; |
---|
284 | } |
---|
285 | |
---|
286 | // Normalize counts |
---|
287 | enumAtts = instances.enumerateAttributes(); |
---|
288 | attIndex = 0; |
---|
289 | while (enumAtts.hasMoreElements()) { |
---|
290 | Attribute attribute = (Attribute) enumAtts.nextElement(); |
---|
291 | if (attribute.isNominal()) { |
---|
292 | for (int j = 0; j < instances.numClasses(); j++) { |
---|
293 | sum = Utils.sum(m_Counts[j][attIndex]); |
---|
294 | for (int i = 0; i < attribute.numValues(); i++) { |
---|
295 | m_Counts[j][attIndex][i] = |
---|
296 | (m_Counts[j][attIndex][i] + 1) |
---|
297 | / (sum + (double)attribute.numValues()); |
---|
298 | } |
---|
299 | } |
---|
300 | } |
---|
301 | attIndex++; |
---|
302 | } |
---|
303 | |
---|
304 | // Normalize priors |
---|
305 | sum = Utils.sum(m_Priors); |
---|
306 | for (int j = 0; j < instances.numClasses(); j++) |
---|
307 | m_Priors[j] = (m_Priors[j] + 1) |
---|
308 | / (sum + (double)instances.numClasses()); |
---|
309 | } |
---|
310 | |
---|
311 | /** |
---|
312 | * Calculates the class membership probabilities for the given test instance. |
---|
313 | * |
---|
314 | * @param instance the instance to be classified |
---|
315 | * @return predicted class probability distribution |
---|
316 | * @exception Exception if distribution can't be computed |
---|
317 | */ |
---|
318 | public double[] distributionForInstance(Instance instance) throws Exception { |
---|
319 | |
---|
320 | double [] probs = new double[instance.numClasses()]; |
---|
321 | int attIndex; |
---|
322 | |
---|
323 | for (int j = 0; j < instance.numClasses(); j++) { |
---|
324 | probs[j] = 1; |
---|
325 | Enumeration enumAtts = instance.enumerateAttributes(); |
---|
326 | attIndex = 0; |
---|
327 | while (enumAtts.hasMoreElements()) { |
---|
328 | Attribute attribute = (Attribute) enumAtts.nextElement(); |
---|
329 | if (!instance.isMissing(attribute)) { |
---|
330 | if (attribute.isNominal()) { |
---|
331 | probs[j] *= m_Counts[j][attIndex][(int)instance.value(attribute)]; |
---|
332 | } else { |
---|
333 | probs[j] *= normalDens(instance.value(attribute), |
---|
334 | m_Means[j][attIndex], |
---|
335 | m_Devs[j][attIndex]);} |
---|
336 | } |
---|
337 | attIndex++; |
---|
338 | } |
---|
339 | probs[j] *= m_Priors[j]; |
---|
340 | } |
---|
341 | |
---|
342 | // Normalize probabilities |
---|
343 | Utils.normalize(probs); |
---|
344 | |
---|
345 | return probs; |
---|
346 | } |
---|
347 | |
---|
348 | /** |
---|
349 | * Returns a description of the classifier. |
---|
350 | * |
---|
351 | * @return a description of the classifier as a string. |
---|
352 | */ |
---|
353 | public String toString() { |
---|
354 | |
---|
355 | if (m_Instances == null) { |
---|
356 | return "Naive Bayes (simple): No model built yet."; |
---|
357 | } |
---|
358 | try { |
---|
359 | StringBuffer text = new StringBuffer("Naive Bayes (simple)"); |
---|
360 | int attIndex; |
---|
361 | |
---|
362 | for (int i = 0; i < m_Instances.numClasses(); i++) { |
---|
363 | text.append("\n\nClass " + m_Instances.classAttribute().value(i) |
---|
364 | + ": P(C) = " |
---|
365 | + Utils.doubleToString(m_Priors[i], 10, 8) |
---|
366 | + "\n\n"); |
---|
367 | Enumeration enumAtts = m_Instances.enumerateAttributes(); |
---|
368 | attIndex = 0; |
---|
369 | while (enumAtts.hasMoreElements()) { |
---|
370 | Attribute attribute = (Attribute) enumAtts.nextElement(); |
---|
371 | text.append("Attribute " + attribute.name() + "\n"); |
---|
372 | if (attribute.isNominal()) { |
---|
373 | for (int j = 0; j < attribute.numValues(); j++) { |
---|
374 | text.append(attribute.value(j) + "\t"); |
---|
375 | } |
---|
376 | text.append("\n"); |
---|
377 | for (int j = 0; j < attribute.numValues(); j++) |
---|
378 | text.append(Utils. |
---|
379 | doubleToString(m_Counts[i][attIndex][j], 10, 8) |
---|
380 | + "\t"); |
---|
381 | } else { |
---|
382 | text.append("Mean: " + Utils. |
---|
383 | doubleToString(m_Means[i][attIndex], 10, 8) + "\t"); |
---|
384 | text.append("Standard Deviation: " |
---|
385 | + Utils.doubleToString(m_Devs[i][attIndex], 10, 8)); |
---|
386 | } |
---|
387 | text.append("\n\n"); |
---|
388 | attIndex++; |
---|
389 | } |
---|
390 | } |
---|
391 | |
---|
392 | return text.toString(); |
---|
393 | } catch (Exception e) { |
---|
394 | return "Can't print Naive Bayes classifier!"; |
---|
395 | } |
---|
396 | } |
---|
397 | |
---|
398 | /** |
---|
399 | * Density function of normal distribution. |
---|
400 | * |
---|
401 | * @param x the value to get the density for |
---|
402 | * @param mean the mean |
---|
403 | * @param stdDev the standard deviation |
---|
404 | * @return the density |
---|
405 | */ |
---|
406 | protected double normalDens(double x, double mean, double stdDev) { |
---|
407 | |
---|
408 | double diff = x - mean; |
---|
409 | |
---|
410 | return (1 / (NORM_CONST * stdDev)) |
---|
411 | * Math.exp(-(diff * diff / (2 * stdDev * stdDev))); |
---|
412 | } |
---|
413 | |
---|
414 | /** |
---|
415 | * Returns the revision string. |
---|
416 | * |
---|
417 | * @return the revision |
---|
418 | */ |
---|
419 | public String getRevision() { |
---|
420 | return RevisionUtils.extract("$Revision: 5928 $"); |
---|
421 | } |
---|
422 | |
---|
423 | /** |
---|
424 | * Main method for testing this class. |
---|
425 | * |
---|
426 | * @param argv the options |
---|
427 | */ |
---|
428 | public static void main(String [] argv) { |
---|
429 | runClassifier(new NaiveBayesSimple(), argv); |
---|
430 | } |
---|
431 | } |
---|