beenull
/
mwmbl
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Experiment with score variations (best is simple weighted domain score)

This commit is contained in:
Daoud Clarke 2022-02-27 21:24:16 +00:00
parent 6fb310c363
commit ee5ca6bcf6
1 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
mwmbl/tinysearchengine/rank.py
View File

@ -14,7 +14,7 @@ from mwmbl.tinysearchengine.indexer import TinyIndex, Document
logger = getLogger(__name__)
SCORE_THRESHOLD = 0.25
SCORE_THRESHOLD = 0.0
def _get_query_regex(terms, is_complete):
@ -49,9 +49,9 @@ def _score_result(terms, result: Document, is_complete: bool, max_score: float):
total_possible_match_length = sum(len(x) for x in terms)
match_score = (match_length + 1. / last_match_char) / (total_possible_match_length + 1)
# score = 0.1 * domain_score + 0.9
# score = (0.01 + 0.99*match_score) * (0.01 + 0.99*(result.score / max_score))
score = 0.1 * match_score + 0.9 * (result.score / max_score)
score = 0.01 * domain_score + 0.99 * match_score
# score = (0.1 + 0.9*match_score) * (0.1 + 0.9*(result.score / max_score))
# score = 0.01 * match_score + 0.99 * (result.score / max_score)
return score