SCM

SCM Repository

[tm] Annotation of /trunk/tm/ChangeLog
ViewVC logotype

Annotation of /trunk/tm/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 748 - (view) (download)

1 : feinerer 748 2007-05-04 Ingo Feinerer <h0125130@wu-wien.ac.at>
2 :    
3 :     * R/termdocmatrix.R (findFreqTerms): Perform efficient computation
4 :     on sparse matrices as proposed by Martin Maechler.
5 :    
6 : feinerer 747 2007-04-27 Ingo Feinerer <h0125130@wu-wien.ac.at>
7 :    
8 :     * R/textdoccol.R: Removed \code{dbDisconnect} calls since last
9 :     \pkg{filehash} version makes them deprecated.
10 :    
11 : feinerer 744 2007-04-22 Ingo Feinerer <h0125130@wu-wien.ac.at>
12 :    
13 :     * R/termdocmatrix.R (textvector): Stemming is now performed before
14 :     erasing stopwords.
15 :     (weightMatrix): Adapted to handle sparse matrices.
16 :     (TermDocMatrix): Sparse matrix is now efficiently built by
17 :     direct stepwise insertion of row values into it.
18 :    
19 : feinerer 741 2007-04-21 Ingo Feinerer <h0125130@wu-wien.ac.at>
20 :    
21 :     * DESCRIPTION: Replaced \pkg{filehashSQLite} with \pkg{filehash}
22 :     due to ongoing problems. For our purposes the latter is as useful
23 :     as the replaced package.
24 :    
25 : feinerer 738 2007-04-20 Ingo Feinerer <h0125130@wu-wien.ac.at>
26 :    
27 :     * man/TextDocCol.Rd: Replaced \code{readPlain} with \code{object@DefaultReader}.
28 :    
29 :     * man/TermDocMatrix.Rd: Remove deprecated \code{language} argument.
30 :    
31 : feinerer 737 2007-04-15 Ingo Feinerer <h0125130@wu-wien.ac.at>
32 :    
33 :     * R/resolve.R (resolveISOCode): Added ISO 639-1 codes for
34 :     languages with available stopwords.
35 :    
36 : feinerer 736 2007-04-14 Ingo Feinerer <h0125130@wu-wien.ac.at>
37 :    
38 :     * inst/doc/tm.Rnw: Minor corrections in the vignette.
39 :    
40 : feinerer 731 2007-04-11 Ingo Feinerer <h0125130@wu-wien.ac.at>
41 :    
42 : feinerer 732 * DESCRIPTION: Update to version 0.2, since a lot of new features
43 :     have been integrated.
44 : feinerer 731
45 : feinerer 732 * inst/stopwords: Updated existing stopwords and added stopwords
46 :     for various other languages.
47 :    
48 : feinerer 729 2007-04-10 Ingo Feinerer <h0125130@wu-wien.ac.at>
49 :    
50 : feinerer 730 * man/: Updated documentation.
51 :    
52 : feinerer 729 * Work/testDb.R: Script to test database stuff.
53 :    
54 :     * R/: Fixed various database related bugs. Seems to be rather
55 :     useable now, i.e., consider as alpha status for now.
56 :    
57 : feinerer 727 2007-04-08 Ingo Feinerer <h0125130@wu-wien.ac.at>
58 :    
59 :     * R/: Fixed some bugs related to database support.
60 :    
61 : feinerer 726 2007-04-07 Ingo Feinerer <h0125130@wu-wien.ac.at>
62 :    
63 :     * man/: Added a lot of examples to the manuals.
64 :    
65 : feinerer 725 2007-04-05 Ingo Feinerer <h0125130@wu-wien.ac.at>
66 :    
67 :     * man/: Updated parts of the documentation.
68 :    
69 :     * R/textdoccol.R (asPlain): Added conversion from newsgroup
70 :     documents to plain text documents.
71 :    
72 : feinerer 722 2007-04-01 Ingo Feinerer <h0125130@wu-wien.ac.at>
73 :    
74 : feinerer 724 * R/textdoccol.R: Finished experimental database support. Not yet
75 :     intensively tested.
76 :    
77 : feinerer 723 * R/source.R: Now each source has a default reader.
78 :    
79 : feinerer 722 * R/reader.R: \code{FunctionGenerator} is now an attribute, not a
80 :     class anymore.
81 :    
82 :     * R/plaintextdoc.R: Custom show method for plain text documents.
83 :    
84 :     * R/aobjects.R: Added a class for structured text documents.
85 :    
86 :     * R/reader.R: Replaced remaining \code{parser} occurrences with
87 :     \code{reader}.
88 :    
89 :     * R/textdoccol.R (summary): Indent tags.
90 :    
91 : feinerer 725 * R/textdoccol.R (removePunctuation): Transform method to remove
92 : feinerer 722 punctuation marks.
93 :    
94 : feinerer 721 2007-03-21 Ingo Feinerer <h0125130@wu-wien.ac.at>
95 :    
96 :     * R/textdoccol.R (sFilter): Simplified sFilter significantly by
97 :     using prescindMeta().
98 :    
99 : feinerer 719 2007-03-18 Ingo Feinerer <h0125130@wu-wien.ac.at>
100 :    
101 :     * R/textdoccol.R: Improved database support.
102 :    
103 : feinerer 717 2007-03-16 Ingo Feinerer <h0125130@wu-wien.ac.at>
104 :    
105 : feinerer 718 * R/termdocmatrix.R (TermDocMatrix): Uses sparse matrices.
106 :    
107 : feinerer 717 * R/resolve.R (resolveISOcode): Extracts the language from a ISO
108 :     language code.
109 :    
110 :     * R/textdoccol.R (TextDocCol): Refactored several parser arguments
111 :     into parserControl argument.
112 :    
113 :     * R/aobjects.R (TextDocument): Introduced the "Language" slot.
114 :    
115 : feinerer 713 2007-03-14 Ingo Feinerer <h0125130@wu-wien.ac.at>
116 :    
117 : feinerer 715 * Work/tmDataSetup.R: The datasets acq and crude can now be
118 :     created on the fly.
119 :    
120 : feinerer 713 * R/stopwords.R: Introduced a function returning the stopwords for
121 :     a given language (English, German and French at the moment)
122 :    
123 :     * R/textdoccol.R (stemDoc): Stemming uses Rstem if available,
124 :     otherwise falls back to Snowball package.
125 :    
126 : feinerer 711 2007-01-30 Ingo Feinerer <h0125130@wu-wien.ac.at>
127 :    
128 :     * man/dissimilarity-methods.Rd: Make clear that any method offered
129 :     by "dists" from package "cba" can be used.
130 :    
131 : feinerer 708 2007-01-22 Ingo Feinerer <h0125130@wu-wien.ac.at>
132 :    
133 : feinerer 710 * inst/doc/tm.Rnw: Fixed quotes-appearing-as-boxes-bug according
134 :     to Kurt's latex suggestion. Removed points and underscores in
135 :     variable names for consistent naming.
136 :    
137 : feinerer 708 * DESCRIPTION: Update to version 0.1-2.
138 :    
139 :     * man/TextRepository.Rd: Fixed bug in documentation.
140 :    
141 : feinerer 704 2007-01-12 Ingo Feinerer <h0125130@wu-wien.ac.at>
142 :    
143 :     * DESCRIPTION: Update to version 0.1-1.
144 :    
145 : feinerer 702 2007-01-09 Ingo Feinerer <h0125130@wu-wien.ac.at>
146 :    
147 :     * R/textdoccol.R (stemDoc): Use Rstem::wordStem instead of
148 :     wordStem.
149 :    
150 : feinerer 698 2007-01-06 Ingo Feinerer <h0125130@wu-wien.ac.at>
151 :    
152 :     * R/: Changes due to Kurt's review.
153 :    
154 : feinerer 694 2006-12-31 Ingo Feinerer <h0125130@wu-wien.ac.at>
155 :    
156 :     * R/: Implemented improvements based upon comments by David
157 :     Meyer.
158 :    
159 : feinerer 691 2006-12-17 Ingo Feinerer <h0125130@wu-wien.ac.at>
160 :    
161 :     * inst/doc/: Rewrote vignette.
162 :    
163 :     * man/: Improved documentation.
164 :    
165 : feinerer 690 2006-12-16 Ingo Feinerer <h0125130@wu-wien.ac.at>
166 :    
167 :     * man/: Updated documentation.
168 :    
169 :     * DESCRIPTION: Changed package name to "tm". Updated version to
170 :     0.1 for first CRAN release.
171 :    
172 :     * inst/texts/gmane.comp.lang.r.general.mbox: mbox Gmane R mailing
173 :     list archive example.
174 :    
175 :     * inst/texts/gmane.comp.lang.r.gr.rdf: RSS Gmane R mailing list
176 :     archive example.
177 :    
178 :     * R/preprocess.R (convert_mbox_eml): A simple e-mail converter
179 :     from (several mails per box) mbox format to (single mail per file)
180 :     eml format.
181 :    
182 : feinerer 689 2006-12-08 Ingo Feinerer <h0125130@wu-wien.ac.at>
183 :    
184 :     * data/crude.rda: Rebuilt.
185 :    
186 :     * data/acq.rda: Rebuilt.
187 :    
188 :     * R/reader.R: Factored out reader and parser methods from
189 :     textdoccol.R.
190 :    
191 :     * R/source.R: Factored out Source methods from aobjects.R and
192 :     textdoccol.R.
193 :     (GmaneRSource): Encapsulates Gmane R mailing list archive RSS
194 :     feeds.
195 :    
196 :     * R/textdoccol.R (DirSource): Added support for recursive
197 :     traversal of directories.
198 :    
199 :     2006-12-07 Ingo Feinerer <h0125130@wu-wien.ac.at>
200 :    
201 :     * R/textdoccol.R ([[): Loads the document corpus automatically
202 :     into memory upon access.
203 :     (tm_transform, tm_filter): Removed several checks whether the
204 :     document is already loaded ([[ ensures this now).
205 :     (gmane_r_reader): Reader for RSS feeds as provided by the Gmane R
206 :     mailing list archive.
207 :    
208 :     2006-12-06 Ingo Feinerer <h0125130@wu-wien.ac.at>
209 :    
210 :     * R/aobjects.R (TextDocument): Is now a virtual class.
211 :     (Source): Is now a virtual class.
212 :    
213 :     2006-12-05 Ingo Feinerer <h0125130@wu-wien.ac.at>
214 :    
215 :     * R/textdoccol.R (c): Support for an arbitrary number of document
216 :     collections.
217 :    
218 : feinerer 77 2006-11-26 Ingo Feinerer <h0125130@wu-wien.ac.at>
219 :    
220 :     * R/textrepo.R: Updated TextRepository (constructor), append_elem,
221 :     append_meta and remove_meta.
222 :    
223 :     * R/textdoccol.R: Removed modify_metadata method.
224 :    
225 :     * R/textrepo.R: Removed modify_metadata method.
226 :    
227 :     * R/textdoccol.R (remove_meta): Supports removal of document
228 :     collection metadata and document (= in data frame) metadata.
229 :    
230 : feinerer 76 2006-11-23 Ingo Feinerer <h0125130@wu-wien.ac.at>
231 :    
232 :     * R/textdoccol.R (append_doc): Bug fix for handling empty metadata.
233 :    
234 :     * data/crude.rda: Rebuilt.
235 :    
236 :     * data/acq.rda: Rebuilt.
237 :    
238 :     * inst/doc/textmin.Rnw: Updated vignette to reflect code changes.
239 :    
240 :     * R/textdoccol.R ([): Bug fix for subsetting a document
241 :     collection's data frame.
242 :    
243 : feinerer 75 2006-11-22 Ingo Feinerer <h0125130@wu-wien.ac.at>
244 :    
245 :     * R/textdoccol.R: Bug fixes in s_filter. Added full query support
246 :     to s_filter.
247 :    
248 :     * R/textdoccol.R: Local text documents' metadata can now be copied
249 :     to a document collection's data frame with prescind_meta.
250 :    
251 : feinerer 73 2006-11-21 Ingo Feinerer <h0125130@wu-wien.ac.at>
252 :    
253 : feinerer 74 * R/: Text documents' slot metadata is now accessible in s_filter.
254 :    
255 : feinerer 73 * R/: Rewrote s_filter function (has still some restrictions).
256 :    
257 : feinerer 72 2006-11-20 Ingo Feinerer <h0125130@wu-wien.ac.at>
258 :    
259 :     * R/: Various fixes in handling metadata.
260 :    
261 :     * R/: Added update mechanism for text document collections.
262 :    
263 : feinerer 71 2006-11-19 Ingo Feinerer <h0125130@wu-wien.ac.at>
264 :    
265 :     * R/: Merging of document collections now creates a binary tree
266 :     for reconstructing merged document collections.
267 :    
268 :     * R/: Redesign of metadata for document collections.
269 :    
270 : feinerer 70 2006-11-07 Ingo Feinerer <h0125130@wu-wien.ac.at>
271 :    
272 :     * R/: Messages now use \code{ngettext}.
273 :    
274 : feinerer 69 2006-11-03 Ingo Feinerer <h0125130@wu-wien.ac.at>
275 :    
276 :     * R/: Added functions for modifying and removing metadata.
277 :    
278 : feinerer 67 2006-11-01 Ingo Feinerer <h0125130@wu-wien.ac.at>
279 :    
280 :     * man/: Updated some documentation.
281 :    
282 :     * R/: Corrected some connection issues.
283 :    
284 :     * inst/doc: Worked on the vignette.
285 :    
286 : feinerer 65 2006-10-31 Ingo Feinerer <h0125130@wu-wien.ac.at>
287 :    
288 : feinerer 66 * inst/: Added texts and started vignette.
289 :    
290 : feinerer 65 * R/: Final changes based upon David's comments.
291 :    
292 : feinerer 64 2006-10-29 Ingo Feinerer <h0125130@wu-wien.ac.at>
293 :    
294 :     * NAMESPACE: Corrected exports (generic methods need exportMethods
295 :     directives!).
296 :    
297 : feinerer 63 2006-10-26 Ingo Feinerer <h0125130@wu-wien.ac.at>
298 :    
299 :     * R/: Modified the TextDocCol constructur and various parsers. It
300 :     is now modular and supports various file formats via plugins (see
301 :     the new "Source" class).
302 :    
303 : feinerer 62 2006-10-24 Ingo Feinerer <h0125130@wu-wien.ac.at>
304 :    
305 :     * man/: Revised documentation after previous code changes.
306 :    
307 : feinerer 61 2006-10-23 Ingo Feinerer <h0125130@wu-wien.ac.at>
308 :    
309 :     * R/: Remaining changes as discussed with David.
310 :    
311 : feinerer 60 2006-10-22 Ingo Feinerer <h0125130@wu-wien.ac.at>
312 :    
313 :     * R/: Some changes as suggested by David. The rest will follow
314 :     within the next days.
315 :    
316 : feinerer 59 2006-09-26 Ingo Feinerer <h0125130@wu-wien.ac.at>
317 :    
318 :     * man/: Finished documentation.
319 :    
320 : feinerer 58 2006-09-25 Ingo Feinerer <h0125130@wu-wien.ac.at>
321 :    
322 :     * man/: Wrote some documentation.
323 :    
324 : feinerer 56 2006-09-24 Ingo Feinerer <h0125130@wu-wien.ac.at>
325 :    
326 :     * R/: Further syntactic sugar in form of additional assignment and
327 :     accessor methods.
328 :    
329 : feinerer 54 2006-09-13 Ingo Feinerer <h0125130@wu-wien.ac.at>
330 :    
331 :     * R/: Syntactic sugar in form of "length", "show" and "summary"
332 : feinerer 56 operators.
333 : feinerer 54
334 : feinerer 53 2006-08-24 Ingo Feinerer <h0125130@wu-wien.ac.at>
335 :    
336 :     * R/: Diverse updates. Mainly on default operators ("[" or "c")
337 :     and dissimilarities.
338 :    
339 : feinerer 52 2006-08-12 Ingo Feinerer <h0125130@wu-wien.ac.at>
340 :    
341 :     * R/: Added similarity functions.
342 :    
343 :     * data/: Added english stopwords.
344 :    
345 : feinerer 50 2006-08-07 Ingo Feinerer <h0125130@wu-wien.ac.at>
346 :    
347 : feinerer 51 * data/: Examples compiled for new features
348 :    
349 :     * R/: Changes due to new structure.
350 :    
351 : feinerer 50 * NAMESPACE: Corrected namespace to reflect new structure.
352 :    
353 :     * R/termdocmatrix.R: Adapted for new naming scheme.
354 :    
355 : feinerer 49 2006-08-06 Ingo Feinerer <h0125130@wu-wien.ac.at>
356 :    
357 :     * R/textdoccol.R: Adapted code for new class structure. Wrote
358 :     several transform and filter functions operating on text document
359 :     collections (alias text document databases).
360 :    
361 :     * R/aobjects.R: Adapted class structure with inheritance,
362 :     repositories and additional meta data. Loading files on demand is
363 :     now possible.
364 :    
365 : feinerer 48 2006-07-13 Ingo Feinerer <h0125130@wu-wien.ac.at>
366 :    
367 :     * R/: Some cosmetic cleanups.
368 :    
369 :     * inst/: Removed vignette on clustering. That and much more is now
370 :     described in the JSS paper on text mining. Based upon that
371 :     article an elaborated vignette will be incorporated in the future.
372 :    
373 : feinerer 42 2006-07-01 Ingo Feinerer <h0125130@wu-wien.ac.at>
374 :    
375 :     * R/: Updated generic S4 methods to comply with signature changes
376 :     in newer versions of R (> 2.3)
377 :    
378 : feinerer 41 2006-03-12 Ingo Feinerer <h0125130@wu-wien.ac.at>
379 :    
380 :     * ext/R/importRIS.R: Automatic RIS import is now possible.
381 :    
382 : feinerer 40 2006-02-14 Ingo Feinerer <h0125130@wu-wien.ac.at>
383 :    
384 :     * R/textdoccol.R: Added RIS HTML input format.
385 :    
386 : feinerer 39 2006-01-21 Ingo Feinerer <h0125130@wu-wien.ac.at>
387 :    
388 :     * R/textdoccol.R: Removed bug that caused invalid text document
389 :     collections when handling many input files.
390 :    
391 : feinerer 36 2006-01-11 Ingo Feinerer <h0125130@wu-wien.ac.at>
392 :    
393 : feinerer 37 * R/textdoccol.R: Restructured and extended file import
394 :     mechanism.
395 :    
396 : feinerer 36 * inst/doc/clustering.Rnw: Adapted vignette for use with
397 :     ReutNews.rda
398 :    
399 :     * man/ReutNews.Rd: Documentation for ReutNews.rda
400 :    
401 :     * data/ReutNews.rda: A tiny Reuters21578 example data set.
402 :    
403 : feinerer 34 2005-12-22 Ingo Feinerer <h0125130@wu-wien.ac.at>
404 :    
405 :     * inst/doc/clustering.Rnw: Wrote a small vignette to present the
406 :     clustering facilities of this package.
407 :    
408 : feinerer 33 2005-12-15 Ingo Feinerer <h0125130@wu-wien.ac.at>
409 :    
410 :     * R/aobjects.R: Changed package document structure to avoid class
411 :     dependency problems.
412 :    
413 : feinerer 28 2005-12-06 Ingo Feinerer <h0125130@wu-wien.ac.at>
414 :    
415 : feinerer 29 * Wrote a script for the ModLewis Split for the Reuters-21578 XML
416 :     data set.
417 :    
418 : feinerer 28 * Finished documentation and reordered directory structure. Now "R
419 :     CMD check textmin" works without errors.
420 :    
421 : feinerer 27 2005-12-04 Ingo Feinerer <h0125130@wu-wien.ac.at>
422 :    
423 :     * src/: Various splits can now be easily created for the
424 :     Reuters21578 data set.
425 :    
426 : feinerer 26 2005-12-03 Ingo Feinerer <h0125130@wu-wien.ac.at>
427 :    
428 :     * Updated documentation
429 :    
430 : feinerer 25 2005-11-30 Ingo Feinerer <h0125130@wu-wien.ac.at>
431 :    
432 :     * Wrote R documentation for some classes and methods.
433 :    
434 : feinerer 21 2005-11-19 Ingo Feinerer <h0125130@wu-wien.ac.at>
435 :    
436 : feinerer 22 * R/textdoccol.R: Constructor of textdoccol allows import of CSV
437 :     files. See the questionnaire data/Umfrage.csv for such an example.
438 : feinerer 23 We are now able to import files in Reuters-21578 XML format.
439 : feinerer 22
440 : feinerer 21 * Changed class interfaces in various files. Weighting of the text
441 :     matrix is now possible.
442 :    
443 : feinerer 20 2005-11-08 Ingo Feinerer <h0125130@wu-wien.ac.at>
444 :    
445 :     * R/textdoccol.R: One can build term-document matrices if
446 :     nessecary (with buildTDM(...)) and fill the field tdm from a text
447 :     document collection with it.
448 :    
449 :     * R/textmatrix.R: Wrote S4 class for term-document matrices.
450 :    
451 : feinerer 19 2005-11-06 Ingo Feinerer <h0125130@wu-wien.ac.at>
452 :    
453 :     * R/textdoccol.R: We now can read in a whole XML file with several
454 :     news items.
455 :    
456 : feinerer 17 2005-11-05 Ingo Feinerer <h0125130@wu-wien.ac.at>
457 :    
458 :     * R/textdoccol.R: Set up an S4 class for a collection of text
459 :     documents. A first attempt to read in XML input (like the RCV1
460 :     set) was made.
461 :    
462 :     * R/textdocument.R: Set up an S4 class for text documents. Wrote
463 :     some accessor functions.
464 :    
465 :     * data/newsitem.xml: Added this XML file for testing purposes. It
466 :     contains a single news item from the Reuters Corpus Volume 1
467 :     (RCV1) XML set.
468 :    
469 : feinerer 16 2005-10-07 Ingo Feinerer <h0125130@wu-wien.ac.at>
470 :    
471 :     * R/textmatrix.R (textmatrix): Removed the transpose of the original
472 :     textmatrix as k-means clustering provided by R (kmeans) now works on
473 :     this textmatrix. The result is a k-means text clustering with a
474 :     similarity measure based upon word frequences.
475 :    
476 :     2005-10-05 Ingo Feinerer <h0125130@wu-wien.ac.at>
477 :    
478 :     * R/textmatrix.R: Adapted the preprocessing code from the R
479 :     package "lsa" written by Fridolin Wild to build a document text matrix.
480 :    
481 :     2005-10-02 Ingo Feinerer <h0125130@wu-wien.ac.at>
482 :    
483 :     * Set up the R Text Mining Package infrastructure.

root@r-forge.r-project.org
ViewVC Help
Powered by ViewVC 1.0.0  
Thanks to:
Vienna University of Economics and Business Powered By FusionForge