mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sean Owen <sro...@gmail.com>
Subject Re: ItemBasedRecommender
Date Fri, 11 Dec 2009 15:41:14 GMT
On Fri, Dec 11, 2009 at 2:47 PM, F.Ozgur Catak <f.ozgur.catak@gmail.com> wrote:
> and Converted using MemoryIDMigrator to long values like
>
> 1903325046098094985,5192157078505275458,-3162216497309240828

I don't understand what the third value is? I think this is your
problem -- should be a preference value.

> When i used Euclidean distance there is no recommendation, but Log
> likelihood Based Item Similarity gives me results which seems very good.
> So, If I use string based input data for recommendation, do I have to use
> "Log likelihood Based Item Similarity"?

You would have to tell me more about your data (see question above)
and what code you are using. Something tells me you have accidentally
transformed your *preference values* or something and are using
"-3162216497309240828" as a preference value! This would mean all
Euclidean distance similarities round to 0, so nothing can be
recommended.

Log-likelihood ignores preference value so isn't affected by the issue above.

Mime
View raw message