Contextualized Attention Metadata (CAM) vs. Attention Profiling Markup Language (APML)

[Will]

I have been following APML for the past year or so and I have begun to
officially start my research in attention profiling. I have read
about Contextualized Attention Metadata (CAM) scheme developed by
Wolpers, Najjar, and such. It takes user's context into account,
which I think is a big thing missing in APML. On the other hand, the
folks at APML have gotten alot of web 2.0-type services to be APML-
enabled. From my understanding, much research in CAM is done in the
learning communities, while APML is more for the regular day-to-day
users. Does anybody see these 2 efforts coming together?

[Gérard Dupont]

Hi Will,

Thank you for the links. Indeed, I see a lot of relation between the two models (APML and CAM). In my understanding, the CAM initiative is more related to a fine analysis of user interaction with information and APML could be then seen as a compilation of this into a more general model. IMHO those models are complementary and links should be established.

Do you personally work on that model ? You mentioned the "learning communities", what do you mean ?

br,

gdupont

--
Gérard Dupont
Information Processing Competence Center (IPCC) - EADS DS
http://weblab-project.org

Perception & Machine Learning team - LITIS Laboratory

[Ash Angell]

Yes I completely agree that these are complementary. 

Remember, that there might be many ways to store and interact with users' attention data and that APML is a high level snapshot of that data - so you can exchange and interact with attention data from other sources very easily.

Ash

[William Yip]

Hi Ash,
Do you see the schema of APML as a little oversimplified? I
personally think that one of the biggest selling points of APML is
portability. But when moving from one application to another, how do
you make sure that the 2 applications are interpreting the same data
the same way. For example, the weight value of each concept in an
APML file, if a user has a high value for a particular concept, it
could mean that he has shown an interest for a long period of time, or
it could mean that he is extremely interested in this concept for a
short duration. How do application developers make that distinction?
In addition, I see context is a big issue. If there is a concept
called "tennis" in my profile, how does it interpret it? That I like
to read tennis articles? Shop for tennis apparels? Play tennis
myself?

Will

On Oct 19, 10:43 pm, "Ash Angell" <ash...@faradaymedia.com> wrote:
> Yes I completely agree that these are complementary.
>
> Remember, that there might be many ways to store and interact with users'
> attention data and that APML is a high level snapshot of that data - so you
> can exchange and interact with attention data from other sources very
> easily.
>
> Ash
>

> On Mon, Oct 20, 2008 at 6:14 PM, Gérard Dupont <ger.dup...@gmail.com> wrote:
> > Hi Will,
>
> > Thank you for the links. Indeed, I see a lot of relation between the two
> > models (APML and CAM). In my understanding, the CAM initiative is more
> > related to a fine analysis of user interaction with information and APML
> > could be then seen as a compilation of this into a more general model. IMHO
> > those models are complementary and links should be established.
>
> > Do you personally work on that model ? You mentioned the "learning
> > communities", what do you mean ?
>
> > br,
>
> > gdupont
>

[William Yip]

Hi gdupont,
No, I do not work on those model myself. I have been reading up
on the CAM schema for the past week or so. When I said learning
communities, I meant that CAM is heavily researched on its modeling
the behavior of teachers and students in a learning environment (e.g.
Moodle, WebCT). It seems like much academic efforts are spent in CAM,
but not in APML. Have you come across any academic researches in
APML?

Will

On Oct 19, 10:14 pm, "Gérard Dupont" <ger.dup...@gmail.com> wrote:
> Hi Will,
>
> Thank you for the links. Indeed, I see a lot of relation between the two
> models (APML and CAM). In my understanding, the CAM initiative is more
> related to a fine analysis of user interaction with information and APML
> could be then seen as a compilation of this into a more general model. IMHO
> those models are complementary and links should be established.
>
> Do you personally work on that model ? You mentioned the "learning
> communities", what do you mean ?
>
> br,
>
> gdupont
>

> On Mon, Oct 20, 2008 at 09:03, Will <william.w....@gmail.com> wrote:
>
> > I have been following APML for the past year or so and I have begun to
> > officially start my research in attention profiling.  I have read
> > about Contextualized Attention Metadata (CAM) scheme developed by
> > Wolpers, Najjar, and such.  It takes user's context into account,
> > which I think is a big thing missing in APML.  On the other hand, the
> > folks at APML have gotten alot of web 2.0-type services to be APML-
> > enabled.  From my understanding, much research in CAM is done in the
> > learning communities, while APML is more for the regular day-to-day
> > users.  Does anybody see these 2 efforts coming together?
>
> --
> Gérard Dupont

> Information Processing Competence Center (IPCC) - EADS DShttp://weblab-project.org

[David P. Novakovic]

William,

APML specifically tries not to fill all the shoes required by academic
endevours... it's primary purpose is to bring an understanding and
more importantly, trust of attention data to the masses. In this case
the best model is the one that reduces the barriers to "getting it" to
the smallest factor possible. This is exactly why APML has had big
wins in the social space, it's easy for users to understand, not
because it is so easily portable (thought portability is very
important).

In the future there will be requirements for more advanced modeling of
context, time, language and other more specific elements. APML is a
format that is growing itself but also seeking to interoperate with
other formats as much as possible without sacrificing the simplicity
that has helped it garner so much attention.

Hope that clarifies things for you.

David

[Ash Angell]

yeah, what he said ;)  lol

Ash