mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Lee Carroll <lee.a.carr...@googlemail.com>
Subject Re: Add on to itemsimilarity
Date Mon, 30 Jan 2012 13:30:22 GMT
Hi Anatoliy

We have separated different forms of suggestions to users. Mainly to
better explain why the user is being presented with these products.

section of CF recommendations (with explain info)
section of previous search (including previous views)
section of content based recommendations (this is a straight
implementation of mlt out of the box and has the best click through
:-)

Now I'm not saying a combined list would not work but the danger is it
looses its focus and confuses or at least is not understandable by the
user.

Is the need for a combined list a UI problem?



On 30 January 2012 10:26, Anatoliy Kats <a.kats@rambler-co.ru> wrote:
> It stands to reason that if you click on an Amazon book description, you
> should be offered to buy it next time.  Amazon certainly does.  I see Sean's
> point, in a pure form a recommender should only recommend unknown items.
>  Certainly that's the behavior you need in a theoretical test framework.  I
> think where we differ is that some people here build systems where ratings
> are computed from user behavior, and therefore decoupled from the set of
> candidate items.  I understand this is not Mahout's original purpose, and
> that it's difficult to build this support into Mahout in a principled way.
>  But it would be helpful to some of us if Mahout had that capability.  I
> accomplished it by wrapping my own recommender class around Mahout's
> delegate, overriding estimatePreference(), and using a CandidateSimilarity
> that allows previously rated items.  Perhaps a more acceptable solution is
> adding a post-recommendation processing step that combines the
> recommendation result with a set of userChoosableRecommendedItems, and
> returns the top N items from that combined list.  This design intrudes a lot
> less into Mahout's internals.
>
> Would anyone else benefit from this addition?
>
>
> On 01/29/2012 12:33 AM, Ted Dunning wrote:
>>
>> Also, Lee, I think you have it backwards.  It is true that clicks are not
>> the same thing as preferences, but I don't think that I give a fig about
>> preferences since they are the internal mental state of the visitor.  I
>> know that some of the visitors are mental, but I don't really care since
>> that is their own business.  What I care about is encouraging certain
>> behaviors.
>>
>> So I find that mental state estimations are the indirect way to model and
>> predict behaviors while directly modeling behaviors based on observed
>> behaviors is, well, more direct.
>>
>> This is compounded by the fact that asking people to rate things invokes a
>> really complicated social interaction which does not directly sample the
>> internal mental state (i.e. the real preference) but instead measures
>> behavior that is a very complicated outcome of social pressures,
>> expectations and the internal mental state.  So using ratings boils down
>> to
>> using one kind of behavior to estimate mental state that then is
>> hypothesized to result in the desired second kind of behavior.
>>
>>
>>
>> On Sat, Jan 28, 2012 at 10:51 AM, Sean Owen<srowen@gmail.com>  wrote:
>>
>>> It means *something* that a user clicked on one item and not 10,000
>>> others.
>>> You will learn things like that Star Wars and Star Trek are somehow
>>> related
>>> from this data. I don't think that clicks are a bad input per se.
>>>
>>> I agree that it's not obvious how to meaningfully translate user actions
>>> into a linear scale. "1" per click and "10" for purchase or something is
>>> a
>>> guess. I do think you will learn something from the data this way.
>>>
>>> There is nothing conceptually wrong with mixing real data and estimated
>>> data. If the results aren't looking right, it is not a problem with the
>>> concept, but the mapping of action onto some rating scale. I think it's
>>> hard to get that right, but is not impossible to get it "good".
>>>
>>> On Sat, Jan 28, 2012 at 10:15 AM, Lee Carroll
>>> <lee.a.carroll@googlemail.com>wrote:
>>>
>>>>> I would argue, though, that .recommend() is aimed at the latter task:
>>>>
>>>> No . I think the mismatch here is you are using at best a wild guess
>>>> at a preference for the convenience of using a recommender and then in
>>>> the same breath expecting the recommender to "understand" that you are
>>>> not using preferences at all and actually have no idea what the user
>>>> preference is. You cant have it both ways :-)
>>>>
>>>> A click through on an item is not a measure of user preference for
>>>> that item. I know its not what you want to hear (or better what your
>>>> business users want to here) but there it is.
>>>>
>>>> We can pretend, or maybe even build a convincing narrative that a
>>>> click is some sort of item association and use that as a proxy
>>>> preference and we might even get some mileage out of it, but we should
>>>> not change the behaviour of the .recommend() to hide its short
>>>> comings.
>>>>
>

Mime
View raw message