Discrete Rating Scale

2014-07-14 Thread Floris Devriendt
Hey all, When using a discrete rating scale (e.g. likes / dislikes), what are the things that I should consider when using Mahout for Collaborative Filtering? If I'm not mistaking I've read a mail a week or two ago from this mailing list stating that one should avoid using 0 (disl

Re: Rating scale

2013-02-04 Thread Sean Owen
to have a fix rating scale while running > recommendations or it can be dynamic based on the users' data ? > > Many Thanks

Re: Rating scale

2013-02-06 Thread 万代豊
> > On Mon, Feb 4, 2013 at 3:56 PM, Zia mel wrote: > > Hi , is there a necessity to have a fix rating scale while running > > recommendations or it can be dynamic based on the users' data ? > > > > Many Thanks >

Re: Rating scale

2013-02-06 Thread Sean Owen
ave users rate on a 1-5 scale one day, and 1-100 > > tomorrow (unless you go back and normalize the old data to 1-100). > > > > On Mon, Feb 4, 2013 at 3:56 PM, Zia mel wrote: > > > Hi , is there a necessity to have a fix rating scale while running > > > recommendations or it can be dynamic based on the users' data ? > > > > > > Many Thanks > > >

Re: Rating scale

2013-02-06 Thread 万代豊
> > It wouldn't work to have users rate on a 1-5 scale one day, and 1-100 > > > tomorrow (unless you go back and normalize the old data to 1-100). > > > > > > On Mon, Feb 4, 2013 at 3:56 PM, Zia mel > wrote: > > > > Hi , is there a necessity to have a fix rating scale while running > > > > recommendations or it can be dynamic based on the users' data ? > > > > > > > > Many Thanks > > > > > >

Re: Rating scale

2013-02-06 Thread Zia mel
> Yutaka >> > >> > >> > >> > 2013/2/5 Sean Owen >> > >> > > You don't have to fix a scale. But your data needs to be consistent. >> > > It wouldn't work to have users rate on a 1-5 scale one day, and 1-100 >> > > tomorrow (unless you go back and normalize the old data to 1-100). >> > > >> > > On Mon, Feb 4, 2013 at 3:56 PM, Zia mel >> wrote: >> > > > Hi , is there a necessity to have a fix rating scale while running >> > > > recommendations or it can be dynamic based on the users' data ? >> > > > >> > > > Many Thanks >> > > >> > >>

Re: Rating scale

2013-02-06 Thread Sean Owen
SVD recommender is more suitable for this purpose. > >> > Apprecaite your advise. > >> > Yutaka > >> > > >> > > >> > > >> > 2013/2/5 Sean Owen > >> > > >> > > You don't have to fix a scale. But your data needs to be consistent. > >> > > It wouldn't work to have users rate on a 1-5 scale one day, and > 1-100 > >> > > tomorrow (unless you go back and normalize the old data to 1-100). > >> > > > >> > > On Mon, Feb 4, 2013 at 3:56 PM, Zia mel > >> wrote: > >> > > > Hi , is there a necessity to have a fix rating scale while running > >> > > > recommendations or it can be dynamic based on the users' data ? > >> > > > > >> > > > Many Thanks > >> > > > >> > > >> >

Re: Rating scale

2013-02-06 Thread Ted Dunning
7;t have to fix a scale. But your data needs to be consistent. > > > It wouldn't work to have users rate on a 1-5 scale one day, and 1-100 > > > tomorrow (unless you go back and normalize the old data to 1-100). > > > > > > On Mon, Feb 4, 2013 at 3:56 PM, Zia mel > wrote: > > > > Hi , is there a necessity to have a fix rating scale while running > > > > recommendations or it can be dynamic based on the users' data ? > > > > > > > > Many Thanks > > > > > >

Re: Discrete Rating Scale

2014-07-14 Thread mario . alemi
4:21 PM, Floris Devriendt wrote: > Hey all, > > When using a discrete rating scale (e.g. likes / dislikes), what are the > things that I should consider when using Mahout for Collaborative > Filtering? > > If I'm not mistaking I've read a mail a week or two ago fr

Re: Discrete Rating Scale

2014-07-14 Thread Ted Dunning
:21 AM, Floris Devriendt wrote: > Hey all, > > When using a discrete rating scale (e.g. likes / dislikes), what are the > things that I should consider when using Mahout for Collaborative > Filtering? > > If I'm not mistaking I've read a mail a week or two ago from t

Re: Discrete Rating Scale

2014-07-14 Thread Floris Devriendt
kelihood are True/False, but False means "not > interacted". Having "dislike = False" would be extremely misleading. > > For all the other algorithms, I'd say one should make similar > considerations. > > Cheers > Mario > > > On Mon, Jul 14, 2

Re: Discrete Rating Scale

2014-07-14 Thread Floris Devriendt
ng that into the DSL as we speak. > > > > On Mon, Jul 14, 2014 at 7:21 AM, Floris Devriendt < > florisdevrie...@gmail.com > > wrote: > > > Hey all, > > > > When using a discrete rating scale (e.g. likes / dislikes), what are the > > things t

Re: Discrete Rating Scale

2014-07-14 Thread Ted Dunning
commendation. Pat is working > > on bringing that into the DSL as we speak. > > > > > > > > On Mon, Jul 14, 2014 at 7:21 AM, Floris Devriendt < > > florisdevrie...@gmail.com > > > wrote: > > > > > Hey all, > > > > > > Wh