You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@mahout.apache.org by Steven Bourke <sb...@gmail.com> on 2010/11/17 01:18:55 UTC
Standard way of getting precision for specific users?
Hi - I'd like to get the precision for a group of users in my dataset. Is
there a standard way of doing this ? I've looked at the IRStatsEvaluator and
cannot see anything.
Re: Standard way of getting precision for specific users?
Posted by Lance Norskog <go...@gmail.com>.
You mean, do you want to get how confident the recommender is about the
quality of its recommendations?
Or an external analysis of the quality?
I've wrestled with the problem, but don't have any good suggestions.
Steven Bourke wrote:
> Just a little more information :-) When I ask about getting precision, I
> mean is there a function I can call in mahout to calculate the score for an
> individual user without implementing my own code.
>
> On Wed, Nov 17, 2010 at 12:18 AM, Steven Bourke<sb...@gmail.com> wrote:
>
>
>> Hi - I'd like to get the precision for a group of users in my dataset. Is
>> there a standard way of doing this ? I've looked at the IRStatsEvaluator and
>> cannot see anything.
>>
>
Re: Standard way of getting precision for specific users?
Posted by Steven Bourke <sb...@gmail.com>.
Just a little more information :-) When I ask about getting precision, I
mean is there a function I can call in mahout to calculate the score for an
individual user without implementing my own code.
On Wed, Nov 17, 2010 at 12:18 AM, Steven Bourke <sb...@gmail.com> wrote:
> Hi - I'd like to get the precision for a group of users in my dataset. Is
> there a standard way of doing this ? I've looked at the IRStatsEvaluator and
> cannot see anything.
Re: Standard way of getting precision for specific users?
Posted by Sean Owen <sr...@gmail.com>.
No, there's no way to force the test data that is held out in these tests,
but, it's pretty simple to modify the code to hold out whatever test data
you want. In this way you could conduct a test over just particular users.
On Wed, Nov 17, 2010 at 12:18 AM, Steven Bourke <sb...@gmail.com> wrote:
> Hi - I'd like to get the precision for a group of users in my dataset. Is
> there a standard way of doing this ? I've looked at the IRStatsEvaluator
> and
> cannot see anything.
>