nyancrimew
/
LatinIME
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Change the formula of the missing character. 

- Bug: 4271049
- Due to the result of the recent user study, a word with a missing character needs to be promoted a bit.

so I changed the formula from:
- freq * 70 * (n - 2) / (n - 1)
to:
- freq * 90 * (10n - 12) / (10n - 2)

Change-Id: Ibff72cbdb0f2d7b91460a06a0fd39a9f5749aa46
main
satok 2011-04-11 16:14:45 +09:00
parent a8c163f262
commit dc5301e590
2 changed files with 8 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
native/src/defines.h
View File

@ -138,7 +138,8 @@ static void prof_out(void) {
#define SUGGEST_WORDS_WITH_SPACE_PROXIMITY true #define SUGGEST_WORDS_WITH_SPACE_PROXIMITY true
// The following "rate"s are used as a multiplier before dividing by 100, so they are in percent. // The following "rate"s are used as a multiplier before dividing by 100, so they are in percent.
#define WORDS_WITH_MISSING_CHARACTER_DEMOTION_RATE 70 #define WORDS_WITH_MISSING_CHARACTER_DEMOTION_RATE 90
#define WORDS_WITH_MISSING_CHARACTER_DEMOTION_START_POS_10X 12
#define WORDS_WITH_MISSING_SPACE_CHARACTER_DEMOTION_RATE 80 #define WORDS_WITH_MISSING_SPACE_CHARACTER_DEMOTION_RATE 80
#define WORDS_WITH_EXCESSIVE_CHARACTER_DEMOTION_RATE 75 #define WORDS_WITH_EXCESSIVE_CHARACTER_DEMOTION_RATE 75
#define WORDS_WITH_EXCESSIVE_CHARACTER_OUT_OF_PROXIMITY_DEMOTION_RATE 75 #define WORDS_WITH_EXCESSIVE_CHARACTER_OUT_OF_PROXIMITY_DEMOTION_RATE 75

9
native/src/unigram_dictionary.cpp
View File

@ -518,9 +518,12 @@ inline int UnigramDictionary::calculateFinalFreq(const int inputIndex, const int
// TODO: Demote by edit distance // TODO: Demote by edit distance
int finalFreq = freq * matchWeight; int finalFreq = freq * matchWeight;
if (skipPos >= 0) { if (skipPos >= 0) {
if (mInputLength >= 3) { if (mInputLength >= 2) {
multiplyRate(WORDS_WITH_MISSING_CHARACTER_DEMOTION_RATE * const int demotionRate = WORDS_WITH_MISSING_CHARACTER_DEMOTION_RATE
(mInputLength - 2) / (mInputLength - 1), &finalFreq); * (10 * mInputLength - WORDS_WITH_MISSING_CHARACTER_DEMOTION_START_POS_10X)
/ (10 * mInputLength
- WORDS_WITH_MISSING_CHARACTER_DEMOTION_START_POS_10X + 10);
multiplyRate(demotionRate, &finalFreq);
} else { } else {
finalFreq = 0; finalFreq = 0;
} }