Omnibiox: Tweak HQP New Scoring and Re-Enable Field Trial

Omnibiox: Tweak HQP New Scoring and Re-Enable Field Trial

Tweaks HQP "new scoring" by
(1) By a smooth squash, prevent HQP from scoring above 1400.  This makes sure inlineable, high-quality, prefix matches (whether scored by HUP or by the HUP-like mode in HQP), which has scores in the 1400-1410 range, are not trumped.
(2) Remove any credit for mid-word matches in titles.  I decided spacing in titles should always make sense and allowing mid-word matches here seems wrong.

Also, I changed the name of the "new scoring" field trial as well so we can distinguish the new behavior trial from the old behavior trial.

This trial is set up independent of the "old scoring" ignore-all-mid-word-matches field trial.  I prefer it this way because then I can easily enable or disable one without having to change much code.  Will analyze them in a dependent/correlated fashion--that is, I will only look at this trial when in the control group of the ignore-mid-word-matches trial and will only analyze the mid-word-matches trial when in the control group of this trial.

BUG=161911
TEST=


Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=170869

[Mark P, 2012-11-30 19:07:33]

Hi Peter,

When it rains field trials, it pours.

I thought more about the results from the last new scoring trial and think this
will help enough to make worth trying again.  The first part of the change is
the more important one; my analysis of the earlier trial showed that most of the
losses came from highly visited results and we saw a lost of usage of the inline
suggestion.  This meant the HQP was trumping the inlineable-HUP scores.  Even
when the HUP was making a good judgment about what inlineable results to put in
what order, sometimes HQP was overruling them in a worse manner.  This change
makes new scoring not be able to beat HUP/HUP-like scoring for inlineable
results.

Please take a look.

thanks,
mark

[Peter Kasting, 2012-12-01 02:45:30]

LGTM

https://codereview.chromium.org/11416285/diff/1/chrome/browser/history/scored...
File chrome/browser/history/scored_history_match.h (right):

https://codereview.chromium.org/11416285/diff/1/chrome/browser/history/scored...
chrome/browser/history/scored_history_match.h:104: float topicality_score, float
recency_score, float popularity_score);
Nit: One arg per line (local code style)

[Mark P, 2012-12-01 22:27:06]

https://codereview.chromium.org/11416285/diff/1/chrome/browser/history/scored...
File chrome/browser/history/scored_history_match.h (right):

https://codereview.chromium.org/11416285/diff/1/chrome/browser/history/scored...
chrome/browser/history/scored_history_match.h:104: float topicality_score, float
recency_score, float popularity_score);
On 2012/12/01 02:45:30, Peter Kasting wrote:
> Nit: One arg per line (local code style)

Done.

[Mark P, 2012-12-01 22:27:54]

Scott,

Can you get your OWNERS stamp on this changelist (as well as the other one I
assigned to you)?

thanks,
mark

[sky, 2012-12-03 15:46:42]

LGTM

[commit-bot: I haz the power, 2012-12-03 17:14:57]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/mpearson@chromium.org/11416285/5001

[commit-bot: I haz the power, 2012-12-03 18:52:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/mpearson@chromium.org/11416285/12002

[commit-bot: I haz the power, 2012-12-03 22:08:20]

Retried try job too often on win_rel for step(s) browser_tests

[commit-bot: I haz the power, 2012-12-03 22:28:41]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/mpearson@chromium.org/11416285/7008

[commit-bot: I haz the power, 2012-12-03 22:38:00]

Sorry for I got bad news for ya.
Compile failed with a clobber build on ios_dbg_simulator.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=ios_dbg_si...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2012-12-03 22:48:31]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/mpearson@chromium.org/11416285/7008

[commit-bot: I haz the power, 2012-12-04 02:57:03]

Change committed as 170869