diff options
| author | mpearson@chromium.org <mpearson@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2013-05-03 19:36:28 +0000 |
|---|---|---|
| committer | mpearson@chromium.org <mpearson@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2013-05-03 19:36:28 +0000 |
| commit | a2fed5a4286f31ff2d6a144106023b3d94ed7534 (patch) | |
| tree | 9f3dde408e79c3e60a2f21197d2cfdb22ae7e1ef /device | |
| parent | 60578e9aee65f59eb7b2898251cc19615f715a86 (diff) | |
| download | chromium_src-a2fed5a4286f31ff2d6a144106023b3d94ed7534.zip chromium_src-a2fed5a4286f31ff2d6a144106023b3d94ed7534.tar.gz chromium_src-a2fed5a4286f31ff2d6a144106023b3d94ed7534.tar.bz2 | |
Omnibox: Add Frecency Scoring Algorithm to HistoryQuick Provider
This uses the HistoryQuick provider "new scoring" code path, replacing
the current scoring in there with code based on the frecency algorith.
Notes:
* The "new scoring" code path is not enabled for anyone. It's guarded by
a field trial that's currently receiving no traffic.
* For this code to behavior usefully, it should be run at the same time as
the HUP-like mode in HQP. This is because if the top result from HQP
is not inlineable, all HQP's results get demoted to 1199 or below. If,
however, HQP does the normal HUP-like scoring, HQP will be more likely to
have a high-scoring inlineable result and thus there's a much greater
opportunity for frecency-scored results to keep their high score.
TBR=pkasting@chromium.org
BUG=175751
Review URL: https://codereview.chromium.org/14161012
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@198170 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'device')
0 files changed, 0 insertions, 0 deletions