| ;; File et94_3.glm, Updated 950117 |
| ;; Global rules for mapping one lexical equivalent to another, |
| ;; for November 1994 CSR Hub and Spoke Evaluation Test Material |
| ;; (Applies to .lsn, not .dot) |
| * name "et94_3.glm" |
| * desc "November 1994 Eval Test Global Orthographic Map" |
| * copy_no_hit = 'T' |
| ;; |
| ;; Verbal Punctuation equivalences |
| !EXCLAMATION-POINT => EXCLAMATION POINT |
| "BEGIN-QUOTE => BEGIN QUOTE |
| "CLOSE-QUOTE => CLOSE QUOTE |
| "CLOSE-QUOTES => CLOSE QUOTES |
| "DOUBLE-QUOTE => DOUBLE QUOTE |
| "END-QUOTE => END QUOTE |
| "OPEN-QUOTE => OPEN QUOTE |
| "QUOTE => QUOTE |
| "UNQUOTE => UNQUOTE |
| #SHARP-SIGN => SHARP SIGN |
| %PERCENT => PERCENT |
| &ERSAND => AMPERSAND |
| 'SINGLE-QUOTE => SINGLE QUOTE |
| (LEFT-PAREN => LEFT PAREN |
| (OPEN-PARENTHESIS => OPEN PARENTHESIS |
| (PARENTHESES => PARENTHESES |
| )CLOSE-PARENTHESIS => CLOSE PARENTHESIS |
| )RIGHT-PAREN => RIGHT PAREN |
| ,COMMA => COMMA |
| --DASH => DASH |
| -DASH => DASH |
| -HYPHEN => HYPHEN |
| .DECIMAL => DECIMAL |
| .PERIOD => PERIOD |
| .POINT => POINT |
| /SLASH => SLASH |
| :COLON => COLON |
| ;SEMI-COLON => SEMICOLON |
| ?QUESTION-MARK => QUESTION MARK |
| END-QUOTE => END QUOTE |
| {LEFT-BRACE => LEFT BRACE |
| }RIGHT-BRACE => RIGHT BRACE |
| ;; |
| ;; Both ways in Training .lsn's, with no syntactic/semantic difference: |
| AD WEEK => ADWEEK |
| AIR LINES => AIRLINES |
| ANTI DRUG => ANTIDRUG |
| AUTO MAKERS => AUTOMAKERS |
| BRIEF CASES => BRIEFCASES |
| BUY BACK => BUYBACK |
| BUY OUT => BUYOUT |
| CEASEFIRE => CEASE FIRE |
| COUNTER PRODUCTIVE => COUNTERPRODUCTIVE |
| FREE WHEELING => FREEWHEELING |
| HEALTHCARE => HEALTH CARE |
| HOTDOG => HOT DOG |
| LAP TOP => LAPTOP |
| LAP TOPS => LAPTOPS |
| LIGHT WEIGHT => LIGHTWEIGHT |
| MID AFTERNOON => MIDAFTERNOON |
| MID YEAR => MIDYEAR |
| MOUSE TRAP => MOUSETRAP |
| MULTI MILLION => MULTIMILLION |
| OUT STRIPPED => OUTSTRIPPED |
| PAPERWORK => PAPERWORK |
| POWER BOOK => POWERBOOK |
| POWER BOOKS => POWERBOOKS |
| PRE TAX => PRETAX |
| RE ELECTION => REELECTION |
| SEVEN FOLD => SEVENFOLD |
| SHORTSELLERS => SHORT SELLERS |
| STING RAY => STINGRAY |
| STORY BOARD => STORYBOARD |
| STORY BOARDS => STORYBOARDS |
| TRACK BALL => TRACKBALL |
| TRACK BALLS => TRACKBALLS |
| TWENTY SOMETHING => TWENTYSOMETHING |
| UNDER FUNDED => UNDERFUNDED |
| UNDER FUNDING => UNDERFUNDING |
| UNDER WAY => UNDERWAY |
| VIDEO TAPE => VIDEOTAPE |
| WAL MART => WALMART |
| WORK FORCE => WORKFORCE |
| WORLD WIDE => WORLDWIDE |
| ;; |
| ;; Both ways in test data with no syntactic/semantic difference: |
| BESICORP. => BESICORP |
| ;; |
| ;; Denise's tilde-words that appear in test, treated as verbalized punctuation: |
| ~EARNINGS => EARNINGS |
| ~END-OF-TOPIC => END OF TOPIC |
| ~GRAPH => GRAPH |
| ~NEW-PARAGRAPH => NEW PARAGRAPH |
| ~NEW-SENTENCE => NEW SENTENCE |
| ~NEW-TOPIC => NEW TOPIC |
| ~OPENING => OPENING |
| ~PARAGRAPH => PARAGRAPH |
| ;; |
| ;; NO SEMANTIC OR ACOUSTIC DIFFERENCE *in this corpus* : |
| CO AUTHOR => COAUTHOR |
| DE FACTO => DEFACTO |
| FEATHER WEIGHT => FEATHERWEIGHT |
| FLY WHEEL => FLYWHEEL |
| GREENE => GREEN |
| MIS INFORMATION => MISINFORMATION |
| NON GOVERNMENT => NONGOVERNMENT |
| OVER ALLOTMENT => OVERALLOTMENT |
| PRE SPLIT => PRESPLIT |
| RE DIRECTED => REDIRECTED |
| RE ELECTION => REELECTION |
| RE INVENTED => REINVENTED |
| ST. => SAINT |
| TELE COMMUTING => TELECOMMUTING |
| ;; |
| ;; BOTH WAYS IN COMMON DICTIONARY |
| RAQUET => RACKET |
| RAQUETS => RACKETS |
| ROM'S => ROMS |
| ;; |
| ;; Test cases |
| " OK " => " OKAY " |
| SHE'S => SHE IS |
| [S.'S] => [{S.'S / S. IS / S. HAS}] / [ ] __ [ ] |
| [RECURSIVE] => [{RECURSIVE/RECUR {SIVE / SI VE}}] / [ ] __A [ ] |
|
|