| Version Number |
Release Date |
Description of Changes |
| Extractor 7.2 |
July 23, 2003 |
- improvements to keyphrases and
highlights in English, French, German, and Spanish
- increased error checking, to warn of
incorrect usage of the API
- increased support for special
punctuation characters
- improvements to wrappers for calling
Extractor from Java, Perl, and Python, in Windows, Linux, and
Solaris
|
| Extractor 7.1 |
June 1, 2001 |
- new API function for deactivating
the plain text filter (useful for writing custom filters)
- changes to the source code to
facilitate the use of customized memory routines
- improved handling of hyphens
|
| Extractor 7.0 |
November 27, 2000 |
- handles English, French, Japanese,
German, Spanish, and Korean text
- choice of three Korean character
encodings: EUC-KR (KS C 5601-1987), Johap (KS X 1001:1992 alternate
encoding), and Unicode UCS-2 (double-byte character code, using
native byte ordering)
- new go phrase feature allows
user to specify important words and phrases
- improvements to highlights in all
languages
- improvements to keyphrases in German
|
| Extractor 6.1 |
September 7, 2000 |
- improvements to Japanese highlights
|
| Extractor 6.0 |
July 17, 2000 |
- handles English, French, Japanese,
German, and Spanish text
- new API function for finding how
many words were read
|
| Extractor 5.1 |
January 21, 2000 |
- extracts key sentences
(highlights) in addition to keyphrases
- important phrases inside highlights
can be automatically marked bold
- unimportant words inside highlights
can be automatically marked grey
- improved filtering of e-mail;
attachments processed according to MIME type
- improved filtering of HTML
|
| Extractor 5.0 |
July 6, 1999 |
- handles English, French, Japanese,
and German text
- improved filtering of HTML
|
| Extractor 4.1 |
May 6, 1999 |
|
| Extractor 4.0 |
March 22, 1999 |
- handles English, French, and Japanese
text
- choice of four Japanese character
encodings: JIS, Shift-JIS, EUC-JP, and Unicode UCS-2 (double-byte
character code, using native byte ordering)
|
| Extractor 3.3 |
February 1, 1999 |
- quality of the keyphrases has been
further improved, especially for French text
- some improvements to handling of
HTML
|
| Extractor 3.2 |
December 14, 1998 |
- user may now request from 3 to 30
keyphrases (previously 5 to 15)
- quality of the keyphrases has been
further improved
|
| Extractor 3.1 |
September 18, 1998 |
- fully reentrant, to allow multithreading
without the use of Win32 services such as semaphores and the
EnterCriticalSection and LeaveCriticalSection functions
- added arguments to ExtrAddStopWord
and ExtrAddStopPhrase, to specify character code
- added support of Unicode UCS2
double-byte character codes, using native byte ordering
|
| Extractor 3.0 |
April 30, 1998 |
- handles both French and
English
- new API functions for French /
English language options
- better API method for specifying
desired number of keyphrases
- e-mail filter handles MIME
quoted-printable accents
- HTML filter handles HTML escape
sequences for accents and ISO Latin-1 HTML character entities
- handles both ISO Latin-1 and MS-DOS
Code Page 437 character codes
|
| Extractor 2.0 |
January 27, 1998 |
- new API function for controlling
number of phrases
|
| Extractor 1.7 |
December 19, 1997 |
- new API function for finding
numerical score of keyphrase
|
| Extractor 1.6 |
November 26, 1997 |
|
| Extractor 1.5 |
September 11, 1997 |
- improved keyphrases
- better filtering of HTML
- better filtering of e-mail
|
| Extractor 1.4 |
July 16, 1997 |
- first version with API
and DLL
- can be embedded in other software
|
| Extractor 1.3 |
June 11, 1997 |
- improved keyphrases
- better filtering of HTML
- better filtering of e-mail
|
| Extractor 1.2 |
April 16, 1997 |
- improved keyphrases
- more stop words
|
| Extractor 1.1 |
January 17, 1997 |
- improved interface
- simplified output
|
| Extractor 1.0 |
January 9, 1997 |
- first release of Extractor
- English only
|
