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 | * PART.java |
---|
19 | * Copyright (C) 1999 University of Waikato, Hamilton, New Zealand |
---|
20 | * |
---|
21 | */ |
---|
22 | |
---|
23 | package weka.classifiers.rules; |
---|
24 | |
---|
25 | import weka.classifiers.Classifier; |
---|
26 | import weka.classifiers.AbstractClassifier; |
---|
27 | import weka.classifiers.rules.part.MakeDecList; |
---|
28 | import weka.classifiers.trees.j48.BinC45ModelSelection; |
---|
29 | import weka.classifiers.trees.j48.C45ModelSelection; |
---|
30 | import weka.classifiers.trees.j48.ModelSelection; |
---|
31 | import weka.core.AdditionalMeasureProducer; |
---|
32 | import weka.core.Capabilities; |
---|
33 | import weka.core.Instance; |
---|
34 | import weka.core.Instances; |
---|
35 | import weka.core.Option; |
---|
36 | import weka.core.OptionHandler; |
---|
37 | import weka.core.RevisionUtils; |
---|
38 | import weka.core.Summarizable; |
---|
39 | import weka.core.TechnicalInformation; |
---|
40 | import weka.core.TechnicalInformationHandler; |
---|
41 | import weka.core.Utils; |
---|
42 | import weka.core.WeightedInstancesHandler; |
---|
43 | import weka.core.TechnicalInformation.Field; |
---|
44 | import weka.core.TechnicalInformation.Type; |
---|
45 | |
---|
46 | import java.util.Enumeration; |
---|
47 | import java.util.Vector; |
---|
48 | |
---|
49 | /** |
---|
50 | <!-- globalinfo-start --> |
---|
51 | * Class for generating a PART decision list. Uses separate-and-conquer. Builds a partial C4.5 decision tree in each iteration and makes the "best" leaf into a rule.<br/> |
---|
52 | * <br/> |
---|
53 | * For more information, see:<br/> |
---|
54 | * <br/> |
---|
55 | * Eibe Frank, Ian H. Witten: Generating Accurate Rule Sets Without Global Optimization. In: Fifteenth International Conference on Machine Learning, 144-151, 1998. |
---|
56 | * <p/> |
---|
57 | <!-- globalinfo-end --> |
---|
58 | * |
---|
59 | <!-- technical-bibtex-start --> |
---|
60 | * BibTeX: |
---|
61 | * <pre> |
---|
62 | * @inproceedings{Frank1998, |
---|
63 | * author = {Eibe Frank and Ian H. Witten}, |
---|
64 | * booktitle = {Fifteenth International Conference on Machine Learning}, |
---|
65 | * editor = {J. Shavlik}, |
---|
66 | * pages = {144-151}, |
---|
67 | * publisher = {Morgan Kaufmann}, |
---|
68 | * title = {Generating Accurate Rule Sets Without Global Optimization}, |
---|
69 | * year = {1998}, |
---|
70 | * PS = {http://www.cs.waikato.ac.nz/\~eibe/pubs/ML98-57.ps.gz} |
---|
71 | * } |
---|
72 | * </pre> |
---|
73 | * <p/> |
---|
74 | <!-- technical-bibtex-end --> |
---|
75 | * |
---|
76 | <!-- options-start --> |
---|
77 | * Valid options are: <p/> |
---|
78 | * |
---|
79 | * <pre> -C <pruning confidence> |
---|
80 | * Set confidence threshold for pruning. |
---|
81 | * (default 0.25)</pre> |
---|
82 | * |
---|
83 | * <pre> -M <minimum number of objects> |
---|
84 | * Set minimum number of objects per leaf. |
---|
85 | * (default 2)</pre> |
---|
86 | * |
---|
87 | * <pre> -R |
---|
88 | * Use reduced error pruning.</pre> |
---|
89 | * |
---|
90 | * <pre> -N <number of folds> |
---|
91 | * Set number of folds for reduced error |
---|
92 | * pruning. One fold is used as pruning set. |
---|
93 | * (default 3)</pre> |
---|
94 | * |
---|
95 | * <pre> -B |
---|
96 | * Use binary splits only.</pre> |
---|
97 | * |
---|
98 | * <pre> -U |
---|
99 | * Generate unpruned decision list.</pre> |
---|
100 | * |
---|
101 | * <pre> -J |
---|
102 | * Do not use MDL correction for info gain on numeric attributes.</pre> |
---|
103 | * |
---|
104 | * <pre> -Q <seed> |
---|
105 | * Seed for random data shuffling (default 1).</pre> |
---|
106 | * |
---|
107 | <!-- options-end --> |
---|
108 | * |
---|
109 | * @author Eibe Frank (eibe@cs.waikato.ac.nz) |
---|
110 | * @version $Revision: 6089 $ |
---|
111 | */ |
---|
112 | public class PART |
---|
113 | extends AbstractClassifier |
---|
114 | implements OptionHandler, WeightedInstancesHandler, Summarizable, |
---|
115 | AdditionalMeasureProducer, TechnicalInformationHandler { |
---|
116 | |
---|
117 | /** for serialization */ |
---|
118 | static final long serialVersionUID = 8121455039782598361L; |
---|
119 | |
---|
120 | /** The decision list */ |
---|
121 | private MakeDecList m_root; |
---|
122 | |
---|
123 | /** Confidence level */ |
---|
124 | private float m_CF = 0.25f; |
---|
125 | |
---|
126 | /** Minimum number of objects */ |
---|
127 | private int m_minNumObj = 2; |
---|
128 | |
---|
129 | /** Use MDL correction? */ |
---|
130 | private boolean m_useMDLcorrection = true; |
---|
131 | |
---|
132 | /** Use reduced error pruning? */ |
---|
133 | private boolean m_reducedErrorPruning = false; |
---|
134 | |
---|
135 | /** Number of folds for reduced error pruning. */ |
---|
136 | private int m_numFolds = 3; |
---|
137 | |
---|
138 | /** Binary splits on nominal attributes? */ |
---|
139 | private boolean m_binarySplits = false; |
---|
140 | |
---|
141 | /** Generate unpruned list? */ |
---|
142 | private boolean m_unpruned = false; |
---|
143 | |
---|
144 | /** The seed for random number generation. */ |
---|
145 | private int m_Seed = 1; |
---|
146 | |
---|
147 | /** |
---|
148 | * Returns a string describing classifier |
---|
149 | * @return a description suitable for |
---|
150 | * displaying in the explorer/experimenter gui |
---|
151 | */ |
---|
152 | public String globalInfo() { |
---|
153 | |
---|
154 | return "Class for generating a PART decision list. Uses " |
---|
155 | + "separate-and-conquer. Builds a partial C4.5 decision tree " |
---|
156 | + "in each iteration and makes the \"best\" leaf into a rule.\n\n" |
---|
157 | + "For more information, see:\n\n" |
---|
158 | + getTechnicalInformation().toString(); |
---|
159 | } |
---|
160 | |
---|
161 | /** |
---|
162 | * Returns an instance of a TechnicalInformation object, containing |
---|
163 | * detailed information about the technical background of this class, |
---|
164 | * e.g., paper reference or book this class is based on. |
---|
165 | * |
---|
166 | * @return the technical information about this class |
---|
167 | */ |
---|
168 | public TechnicalInformation getTechnicalInformation() { |
---|
169 | TechnicalInformation result; |
---|
170 | |
---|
171 | result = new TechnicalInformation(Type.INPROCEEDINGS); |
---|
172 | result.setValue(Field.AUTHOR, "Eibe Frank and Ian H. Witten"); |
---|
173 | result.setValue(Field.TITLE, "Generating Accurate Rule Sets Without Global Optimization"); |
---|
174 | result.setValue(Field.BOOKTITLE, "Fifteenth International Conference on Machine Learning"); |
---|
175 | result.setValue(Field.EDITOR, "J. Shavlik"); |
---|
176 | result.setValue(Field.YEAR, "1998"); |
---|
177 | result.setValue(Field.PAGES, "144-151"); |
---|
178 | result.setValue(Field.PUBLISHER, "Morgan Kaufmann"); |
---|
179 | result.setValue(Field.PS, "http://www.cs.waikato.ac.nz/~eibe/pubs/ML98-57.ps.gz"); |
---|
180 | |
---|
181 | return result; |
---|
182 | } |
---|
183 | |
---|
184 | /** |
---|
185 | * Returns default capabilities of the classifier. |
---|
186 | * |
---|
187 | * @return the capabilities of this classifier |
---|
188 | */ |
---|
189 | public Capabilities getCapabilities() { |
---|
190 | Capabilities result; |
---|
191 | |
---|
192 | if (m_unpruned) |
---|
193 | result = new MakeDecList(null, m_minNumObj).getCapabilities(); |
---|
194 | else if (m_reducedErrorPruning) |
---|
195 | result = new MakeDecList(null, m_numFolds, m_minNumObj, m_Seed).getCapabilities(); |
---|
196 | else |
---|
197 | result = new MakeDecList(null, m_CF, m_minNumObj).getCapabilities(); |
---|
198 | |
---|
199 | return result; |
---|
200 | } |
---|
201 | |
---|
202 | /** |
---|
203 | * Generates the classifier. |
---|
204 | * |
---|
205 | * @param instances the data to train with |
---|
206 | * @throws Exception if classifier can't be built successfully |
---|
207 | */ |
---|
208 | public void buildClassifier(Instances instances) |
---|
209 | throws Exception { |
---|
210 | |
---|
211 | // can classifier handle the data? |
---|
212 | getCapabilities().testWithFail(instances); |
---|
213 | |
---|
214 | // remove instances with missing class |
---|
215 | instances = new Instances(instances); |
---|
216 | instances.deleteWithMissingClass(); |
---|
217 | |
---|
218 | ModelSelection modSelection; |
---|
219 | |
---|
220 | if (m_binarySplits) |
---|
221 | modSelection = new BinC45ModelSelection(m_minNumObj, instances, m_useMDLcorrection); |
---|
222 | else |
---|
223 | modSelection = new C45ModelSelection(m_minNumObj, instances, m_useMDLcorrection); |
---|
224 | if (m_unpruned) |
---|
225 | m_root = new MakeDecList(modSelection, m_minNumObj); |
---|
226 | else if (m_reducedErrorPruning) |
---|
227 | m_root = new MakeDecList(modSelection, m_numFolds, m_minNumObj, m_Seed); |
---|
228 | else |
---|
229 | m_root = new MakeDecList(modSelection, m_CF, m_minNumObj); |
---|
230 | m_root.buildClassifier(instances); |
---|
231 | if (m_binarySplits) { |
---|
232 | ((BinC45ModelSelection)modSelection).cleanup(); |
---|
233 | } else { |
---|
234 | ((C45ModelSelection)modSelection).cleanup(); |
---|
235 | } |
---|
236 | } |
---|
237 | |
---|
238 | /** |
---|
239 | * Classifies an instance. |
---|
240 | * |
---|
241 | * @param instance the instance to classify |
---|
242 | * @return the classification |
---|
243 | * @throws Exception if instance can't be classified successfully |
---|
244 | */ |
---|
245 | public double classifyInstance(Instance instance) |
---|
246 | throws Exception { |
---|
247 | |
---|
248 | return m_root.classifyInstance(instance); |
---|
249 | } |
---|
250 | |
---|
251 | /** |
---|
252 | * Returns class probabilities for an instance. |
---|
253 | * |
---|
254 | * @param instance the instance to get the distribution for |
---|
255 | * @return the class probabilities |
---|
256 | * @throws Exception if the distribution can't be computed successfully |
---|
257 | */ |
---|
258 | public final double [] distributionForInstance(Instance instance) |
---|
259 | throws Exception { |
---|
260 | |
---|
261 | return m_root.distributionForInstance(instance); |
---|
262 | } |
---|
263 | |
---|
264 | /** |
---|
265 | * Returns an enumeration describing the available options. |
---|
266 | * |
---|
267 | * Valid options are: <p> |
---|
268 | * |
---|
269 | * -C confidence <br> |
---|
270 | * Set confidence threshold for pruning. (Default: 0.25) <p> |
---|
271 | * |
---|
272 | * -M number <br> |
---|
273 | * Set minimum number of instances per leaf. (Default: 2) <p> |
---|
274 | * |
---|
275 | * -R <br> |
---|
276 | * Use reduced error pruning. <p> |
---|
277 | * |
---|
278 | * -N number <br> |
---|
279 | * Set number of folds for reduced error pruning. One fold is |
---|
280 | * used as the pruning set. (Default: 3) <p> |
---|
281 | * |
---|
282 | * -B <br> |
---|
283 | * Use binary splits for nominal attributes. <p> |
---|
284 | * |
---|
285 | * -U <br> |
---|
286 | * Generate unpruned decision list. <p> |
---|
287 | * |
---|
288 | * -Q <br> |
---|
289 | * The seed for reduced-error pruning. <p> |
---|
290 | * |
---|
291 | * @return an enumeration of all the available options. |
---|
292 | */ |
---|
293 | public Enumeration listOptions() { |
---|
294 | |
---|
295 | Vector newVector = new Vector(8); |
---|
296 | |
---|
297 | newVector. |
---|
298 | addElement(new Option("\tSet confidence threshold for pruning.\n" + |
---|
299 | "\t(default 0.25)", |
---|
300 | "C", 1, "-C <pruning confidence>")); |
---|
301 | newVector. |
---|
302 | addElement(new Option("\tSet minimum number of objects per leaf.\n" + |
---|
303 | "\t(default 2)", |
---|
304 | "M", 1, "-M <minimum number of objects>")); |
---|
305 | newVector. |
---|
306 | addElement(new Option("\tUse reduced error pruning.", |
---|
307 | "R", 0, "-R")); |
---|
308 | newVector. |
---|
309 | addElement(new Option("\tSet number of folds for reduced error\n" + |
---|
310 | "\tpruning. One fold is used as pruning set.\n" + |
---|
311 | "\t(default 3)", |
---|
312 | "N", 1, "-N <number of folds>")); |
---|
313 | newVector. |
---|
314 | addElement(new Option("\tUse binary splits only.", |
---|
315 | "B", 0, "-B")); |
---|
316 | newVector. |
---|
317 | addElement(new Option("\tGenerate unpruned decision list.", |
---|
318 | "U", 0, "-U")); |
---|
319 | newVector. |
---|
320 | addElement(new Option("\tDo not use MDL correction for info gain on numeric attributes.", |
---|
321 | "J", 0, "-J")); |
---|
322 | newVector. |
---|
323 | addElement(new Option("\tSeed for random data shuffling (default 1).", |
---|
324 | "Q", 1, "-Q <seed>")); |
---|
325 | |
---|
326 | return newVector.elements(); |
---|
327 | } |
---|
328 | |
---|
329 | /** |
---|
330 | * Parses a given list of options. <p/> |
---|
331 | * |
---|
332 | <!-- options-start --> |
---|
333 | * Valid options are: <p/> |
---|
334 | * |
---|
335 | * <pre> -C <pruning confidence> |
---|
336 | * Set confidence threshold for pruning. |
---|
337 | * (default 0.25)</pre> |
---|
338 | * |
---|
339 | * <pre> -M <minimum number of objects> |
---|
340 | * Set minimum number of objects per leaf. |
---|
341 | * (default 2)</pre> |
---|
342 | * |
---|
343 | * <pre> -R |
---|
344 | * Use reduced error pruning.</pre> |
---|
345 | * |
---|
346 | * <pre> -N <number of folds> |
---|
347 | * Set number of folds for reduced error |
---|
348 | * pruning. One fold is used as pruning set. |
---|
349 | * (default 3)</pre> |
---|
350 | * |
---|
351 | * <pre> -B |
---|
352 | * Use binary splits only.</pre> |
---|
353 | * |
---|
354 | * <pre> -U |
---|
355 | * Generate unpruned decision list.</pre> |
---|
356 | * |
---|
357 | * <pre> -J |
---|
358 | * Do not use MDL correction for info gain on numeric attributes.</pre> |
---|
359 | * |
---|
360 | * <pre> -Q <seed> |
---|
361 | * Seed for random data shuffling (default 1).</pre> |
---|
362 | * |
---|
363 | <!-- options-end --> |
---|
364 | * |
---|
365 | * @param options the list of options as an array of strings |
---|
366 | * @throws Exception if an option is not supported |
---|
367 | */ |
---|
368 | public void setOptions(String[] options) throws Exception { |
---|
369 | |
---|
370 | // Pruning options |
---|
371 | m_unpruned = Utils.getFlag('U', options); |
---|
372 | m_reducedErrorPruning = Utils.getFlag('R', options); |
---|
373 | m_binarySplits = Utils.getFlag('B', options); |
---|
374 | m_useMDLcorrection = !Utils.getFlag('J', options); |
---|
375 | String confidenceString = Utils.getOption('C', options); |
---|
376 | if (confidenceString.length() != 0) { |
---|
377 | if (m_reducedErrorPruning) { |
---|
378 | throw new Exception("Setting CF doesn't make sense " + |
---|
379 | "for reduced error pruning."); |
---|
380 | } else { |
---|
381 | m_CF = (new Float(confidenceString)).floatValue(); |
---|
382 | if ((m_CF <= 0) || (m_CF >= 1)) { |
---|
383 | throw new Exception("CF has to be greater than zero and smaller than one!"); |
---|
384 | } |
---|
385 | } |
---|
386 | } else { |
---|
387 | m_CF = 0.25f; |
---|
388 | } |
---|
389 | String numFoldsString = Utils.getOption('N', options); |
---|
390 | if (numFoldsString.length() != 0) { |
---|
391 | if (!m_reducedErrorPruning) { |
---|
392 | throw new Exception("Setting the number of folds" + |
---|
393 | " does only make sense for" + |
---|
394 | " reduced error pruning."); |
---|
395 | } else { |
---|
396 | m_numFolds = Integer.parseInt(numFoldsString); |
---|
397 | } |
---|
398 | } else { |
---|
399 | m_numFolds = 3; |
---|
400 | } |
---|
401 | |
---|
402 | // Other options |
---|
403 | String minNumString = Utils.getOption('M', options); |
---|
404 | if (minNumString.length() != 0) { |
---|
405 | m_minNumObj = Integer.parseInt(minNumString); |
---|
406 | } else { |
---|
407 | m_minNumObj = 2; |
---|
408 | } |
---|
409 | String seedString = Utils.getOption('Q', options); |
---|
410 | if (seedString.length() != 0) { |
---|
411 | m_Seed = Integer.parseInt(seedString); |
---|
412 | } else { |
---|
413 | m_Seed = 1; |
---|
414 | } |
---|
415 | } |
---|
416 | |
---|
417 | /** |
---|
418 | * Gets the current settings of the Classifier. |
---|
419 | * |
---|
420 | * @return an array of strings suitable for passing to setOptions |
---|
421 | */ |
---|
422 | public String [] getOptions() { |
---|
423 | |
---|
424 | String [] options = new String [12]; |
---|
425 | int current = 0; |
---|
426 | |
---|
427 | if (m_unpruned) { |
---|
428 | options[current++] = "-U"; |
---|
429 | } |
---|
430 | if (m_reducedErrorPruning) { |
---|
431 | options[current++] = "-R"; |
---|
432 | } |
---|
433 | if (m_binarySplits) { |
---|
434 | options[current++] = "-B"; |
---|
435 | } |
---|
436 | options[current++] = "-M"; options[current++] = "" + m_minNumObj; |
---|
437 | if (!m_reducedErrorPruning) { |
---|
438 | options[current++] = "-C"; options[current++] = "" + m_CF; |
---|
439 | } |
---|
440 | if (m_reducedErrorPruning) { |
---|
441 | options[current++] = "-N"; options[current++] = "" + m_numFolds; |
---|
442 | } |
---|
443 | options[current++] = "-Q"; options[current++] = "" + m_Seed; |
---|
444 | if (!m_useMDLcorrection) { |
---|
445 | options[current++] = "-J"; |
---|
446 | } |
---|
447 | |
---|
448 | while (current < options.length) { |
---|
449 | options[current++] = ""; |
---|
450 | } |
---|
451 | return options; |
---|
452 | } |
---|
453 | |
---|
454 | /** |
---|
455 | * Returns a description of the classifier |
---|
456 | * |
---|
457 | * @return a string representation of the classifier |
---|
458 | */ |
---|
459 | public String toString() { |
---|
460 | |
---|
461 | if (m_root == null) { |
---|
462 | return "No classifier built"; |
---|
463 | } |
---|
464 | return "PART decision list\n------------------\n\n" + m_root.toString(); |
---|
465 | } |
---|
466 | |
---|
467 | /** |
---|
468 | * Returns a superconcise version of the model |
---|
469 | * |
---|
470 | * @return a concise version of the model |
---|
471 | */ |
---|
472 | public String toSummaryString() { |
---|
473 | |
---|
474 | return "Number of rules: " + m_root.numRules() + "\n"; |
---|
475 | } |
---|
476 | |
---|
477 | /** |
---|
478 | * Return the number of rules. |
---|
479 | * @return the number of rules |
---|
480 | */ |
---|
481 | public double measureNumRules() { |
---|
482 | return m_root.numRules(); |
---|
483 | } |
---|
484 | |
---|
485 | /** |
---|
486 | * Returns an enumeration of the additional measure names |
---|
487 | * @return an enumeration of the measure names |
---|
488 | */ |
---|
489 | public Enumeration enumerateMeasures() { |
---|
490 | Vector newVector = new Vector(1); |
---|
491 | newVector.addElement("measureNumRules"); |
---|
492 | return newVector.elements(); |
---|
493 | } |
---|
494 | |
---|
495 | /** |
---|
496 | * Returns the value of the named measure |
---|
497 | * @param additionalMeasureName the name of the measure to query for its value |
---|
498 | * @return the value of the named measure |
---|
499 | * @throws IllegalArgumentException if the named measure is not supported |
---|
500 | */ |
---|
501 | public double getMeasure(String additionalMeasureName) { |
---|
502 | if (additionalMeasureName.compareToIgnoreCase("measureNumRules") == 0) { |
---|
503 | return measureNumRules(); |
---|
504 | } else { |
---|
505 | throw new IllegalArgumentException(additionalMeasureName |
---|
506 | + " not supported (PART)"); |
---|
507 | } |
---|
508 | } |
---|
509 | |
---|
510 | /** |
---|
511 | * Returns the tip text for this property |
---|
512 | * @return tip text for this property suitable for |
---|
513 | * displaying in the explorer/experimenter gui |
---|
514 | */ |
---|
515 | public String confidenceFactorTipText() { |
---|
516 | return "The confidence factor used for pruning (smaller values incur " |
---|
517 | + "more pruning)."; |
---|
518 | } |
---|
519 | |
---|
520 | /** |
---|
521 | * Get the value of CF. |
---|
522 | * |
---|
523 | * @return Value of CF. |
---|
524 | */ |
---|
525 | public float getConfidenceFactor() { |
---|
526 | |
---|
527 | return m_CF; |
---|
528 | } |
---|
529 | |
---|
530 | /** |
---|
531 | * Set the value of CF. |
---|
532 | * |
---|
533 | * @param v Value to assign to CF. |
---|
534 | */ |
---|
535 | public void setConfidenceFactor(float v) { |
---|
536 | |
---|
537 | m_CF = v; |
---|
538 | } |
---|
539 | |
---|
540 | /** |
---|
541 | * Returns the tip text for this property |
---|
542 | * @return tip text for this property suitable for |
---|
543 | * displaying in the explorer/experimenter gui |
---|
544 | */ |
---|
545 | public String minNumObjTipText() { |
---|
546 | return "The minimum number of instances per rule."; |
---|
547 | } |
---|
548 | |
---|
549 | /** |
---|
550 | * Get the value of minNumObj. |
---|
551 | * |
---|
552 | * @return Value of minNumObj. |
---|
553 | */ |
---|
554 | public int getMinNumObj() { |
---|
555 | |
---|
556 | return m_minNumObj; |
---|
557 | } |
---|
558 | |
---|
559 | /** |
---|
560 | * Set the value of minNumObj. |
---|
561 | * |
---|
562 | * @param v Value to assign to minNumObj. |
---|
563 | */ |
---|
564 | public void setMinNumObj(int v) { |
---|
565 | |
---|
566 | m_minNumObj = v; |
---|
567 | } |
---|
568 | |
---|
569 | /** |
---|
570 | * Returns the tip text for this property |
---|
571 | * @return tip text for this property suitable for |
---|
572 | * displaying in the explorer/experimenter gui |
---|
573 | */ |
---|
574 | public String reducedErrorPruningTipText() { |
---|
575 | return "Whether reduced-error pruning is used instead of C.4.5 pruning."; |
---|
576 | } |
---|
577 | |
---|
578 | /** |
---|
579 | * Get the value of reducedErrorPruning. |
---|
580 | * |
---|
581 | * @return Value of reducedErrorPruning. |
---|
582 | */ |
---|
583 | public boolean getReducedErrorPruning() { |
---|
584 | |
---|
585 | return m_reducedErrorPruning; |
---|
586 | } |
---|
587 | |
---|
588 | /** |
---|
589 | * Set the value of reducedErrorPruning. |
---|
590 | * |
---|
591 | * @param v Value to assign to reducedErrorPruning. |
---|
592 | */ |
---|
593 | public void setReducedErrorPruning(boolean v) { |
---|
594 | |
---|
595 | m_reducedErrorPruning = v; |
---|
596 | } |
---|
597 | |
---|
598 | /** |
---|
599 | * Returns the tip text for this property |
---|
600 | * @return tip text for this property suitable for |
---|
601 | * displaying in the explorer/experimenter gui |
---|
602 | */ |
---|
603 | public String unprunedTipText() { |
---|
604 | return "Whether pruning is performed."; |
---|
605 | } |
---|
606 | |
---|
607 | /** |
---|
608 | * Get the value of unpruned. |
---|
609 | * |
---|
610 | * @return Value of unpruned. |
---|
611 | */ |
---|
612 | public boolean getUnpruned() { |
---|
613 | |
---|
614 | return m_unpruned; |
---|
615 | } |
---|
616 | |
---|
617 | /** |
---|
618 | * Set the value of unpruned. |
---|
619 | * |
---|
620 | * @param newunpruned Value to assign to unpruned. |
---|
621 | */ |
---|
622 | public void setUnpruned(boolean newunpruned) { |
---|
623 | |
---|
624 | m_unpruned = newunpruned; |
---|
625 | } |
---|
626 | |
---|
627 | /** |
---|
628 | * Returns the tip text for this property |
---|
629 | * @return tip text for this property suitable for |
---|
630 | * displaying in the explorer/experimenter gui |
---|
631 | */ |
---|
632 | public String useMDLcorrectionTipText() { |
---|
633 | return "Whether MDL correction is used when finding splits on numeric attributes."; |
---|
634 | } |
---|
635 | |
---|
636 | /** |
---|
637 | * Get the value of useMDLcorrection. |
---|
638 | * |
---|
639 | * @return Value of useMDLcorrection. |
---|
640 | */ |
---|
641 | public boolean getUseMDLcorrection() { |
---|
642 | |
---|
643 | return m_useMDLcorrection; |
---|
644 | } |
---|
645 | |
---|
646 | /** |
---|
647 | * Set the value of useMDLcorrection. |
---|
648 | * |
---|
649 | * @param newuseMDLcorrection Value to assign to useMDLcorrection. |
---|
650 | */ |
---|
651 | public void setUseMDLcorrection(boolean newuseMDLcorrection) { |
---|
652 | |
---|
653 | m_useMDLcorrection = newuseMDLcorrection; |
---|
654 | } |
---|
655 | |
---|
656 | /** |
---|
657 | * Returns the tip text for this property |
---|
658 | * @return tip text for this property suitable for |
---|
659 | * displaying in the explorer/experimenter gui |
---|
660 | */ |
---|
661 | public String numFoldsTipText() { |
---|
662 | return "Determines the amount of data used for reduced-error pruning. " |
---|
663 | + " One fold is used for pruning, the rest for growing the rules."; |
---|
664 | } |
---|
665 | |
---|
666 | /** |
---|
667 | * Get the value of numFolds. |
---|
668 | * |
---|
669 | * @return Value of numFolds. |
---|
670 | */ |
---|
671 | public int getNumFolds() { |
---|
672 | |
---|
673 | return m_numFolds; |
---|
674 | } |
---|
675 | |
---|
676 | /** |
---|
677 | * Set the value of numFolds. |
---|
678 | * |
---|
679 | * @param v Value to assign to numFolds. |
---|
680 | */ |
---|
681 | public void setNumFolds(int v) { |
---|
682 | |
---|
683 | m_numFolds = v; |
---|
684 | } |
---|
685 | |
---|
686 | /** |
---|
687 | * Returns the tip text for this property |
---|
688 | * @return tip text for this property suitable for |
---|
689 | * displaying in the explorer/experimenter gui |
---|
690 | */ |
---|
691 | public String seedTipText() { |
---|
692 | return "The seed used for randomizing the data " + |
---|
693 | "when reduced-error pruning is used."; |
---|
694 | } |
---|
695 | |
---|
696 | /** |
---|
697 | * Get the value of Seed. |
---|
698 | * |
---|
699 | * @return Value of Seed. |
---|
700 | */ |
---|
701 | public int getSeed() { |
---|
702 | |
---|
703 | return m_Seed; |
---|
704 | } |
---|
705 | |
---|
706 | /** |
---|
707 | * Set the value of Seed. |
---|
708 | * |
---|
709 | * @param newSeed Value to assign to Seed. |
---|
710 | */ |
---|
711 | public void setSeed(int newSeed) { |
---|
712 | |
---|
713 | m_Seed = newSeed; |
---|
714 | } |
---|
715 | |
---|
716 | /** |
---|
717 | * Returns the tip text for this property |
---|
718 | * @return tip text for this property suitable for |
---|
719 | * displaying in the explorer/experimenter gui |
---|
720 | */ |
---|
721 | public String binarySplitsTipText() { |
---|
722 | return "Whether to use binary splits on nominal attributes when " |
---|
723 | + "building the partial trees."; |
---|
724 | } |
---|
725 | |
---|
726 | /** |
---|
727 | * Get the value of binarySplits. |
---|
728 | * |
---|
729 | * @return Value of binarySplits. |
---|
730 | */ |
---|
731 | public boolean getBinarySplits() { |
---|
732 | |
---|
733 | return m_binarySplits; |
---|
734 | } |
---|
735 | |
---|
736 | /** |
---|
737 | * Set the value of binarySplits. |
---|
738 | * |
---|
739 | * @param v Value to assign to binarySplits. |
---|
740 | */ |
---|
741 | public void setBinarySplits(boolean v) { |
---|
742 | |
---|
743 | m_binarySplits = v; |
---|
744 | } |
---|
745 | |
---|
746 | /** |
---|
747 | * Returns the revision string. |
---|
748 | * |
---|
749 | * @return the revision |
---|
750 | */ |
---|
751 | public String getRevision() { |
---|
752 | return RevisionUtils.extract("$Revision: 6089 $"); |
---|
753 | } |
---|
754 | |
---|
755 | /** |
---|
756 | * Main method for testing this class. |
---|
757 | * |
---|
758 | * @param argv command line options |
---|
759 | */ |
---|
760 | public static void main(String [] argv){ |
---|
761 | runClassifier(new PART(), argv); |
---|
762 | } |
---|
763 | } |
---|
764 | |
---|