[Accessforall] [Profiles] "User preference key" instead of "User ID"

Gregg Vanderheiden gv at trace.wisc.edu
Fri Jul 6 10:22:12 UTC 2012


here Is more evidence why we don't want to use PROFILE  or  USER  PROFILE

The term is very tightly bound to the idea of describing the USER rather than their needs of the user or what they need of the outside environment . 

                   
Gregg
--------------------------------------------------------
Gregg Vanderheiden Ph.D.
Director Trace R&D Center
Professor Industrial & Systems Engineering
and Biomedical Engineering
University of Wisconsin-Madison

Co-Director, Raising the Floor - International
and the Global Public Inclusive Infrastructure Project
http://Raisingthefloor.org   ---   http://GPII.net




1.1.1    Review of user profile models and standards

1.1.1.1     Formal User Profile Models

Semantic modelling in the domain of eInclusion and accessibility has been elaborated in various EU projects. Projects such as AEGIS  (AEGIS, n.d.), ACCESSIBLE  (ACCESSIBLE, n.d.), ASK-IT  (ASK-IT, n.d.), GUIDE  (GUIDE, n.d.), myUI  (myUI, n.d.), VERITAS  (Veritas, n.d.), VICON  (VICON, n.d.) and VUMS  (VUMS, n.d.), have all proposed and developed semantic models for a variety of application domains and scopes, e.g., related with user interaction, functional limitations, AT frameworks, accessibility assessment, services, devices, etc. From the above list, GUIDE, myUI, VICON, VERITAS and their cluster VUMS are the most relevant ones with user profile modelling.

More specifically, GUIDE is developing a user model that reflects impairments and preferences of elderly people. The model represents knowledge about the user’s impairments, his cognition, perception and motor capabilities and individual preferences. Basic data for this model has been collected in user trials.

myUI aims at providing support for user interfaces that dynamically adapt to the needs of the user and the characteristics of the user’s environment, i.e., the context of a user. In this regard, its major emphasis is in the structured representation of the context, often labelled as “user model” and “environment model”. For this representation, myUI defines a formalism comprising of key features and their possible values, following an ontology-based approach. In general, a user profile in myUI is a collection of information about an end-user of the underlying system, including personal information like email address, first name, last name, etc. as well as information about his/her user capabilities and characteristics as far as they are relevant to determine the human computer interaction (HCI) abilities of the user. myUI concentrates on a task-oriented user profile, where information captures disabilities and impairments in relation to HCI.

VERITAS employs ontologies to provide an interoperable and extensible description of its Abstract User Models, i.e. high-level descriptions of potential user models. The user models are developed with respect to several specific disabilities and are broken down according to the disability category, i.e. cognitive user models, physical user models, behavioral & psychological user models. An Abstract User Model stored in the ontology includes the type of user disability, user capabilities according to the ICF functional abilities framework, user needs, characteristics from cognitive user models, physical user models, behavioral and psychological user models, guidelines and standards.

VICON introduces the notion of Virtual User Model comprising (among other) of a User Model, where all information about the potential users of a product is stored. Focus is upon exemplary users with mild to moderate physical impairments (specifically, hearing, sight and dexterity impairments). The respective user models are divided into several subgroups (profiles), which are divided into the different levels of impairments.

VUMS, being the cluster of the abovementioned four EU projects, aims to lay the foundation for an improved interoperability of the projects' solutions and software. VUMS defines a “user profile” as an instantiation of a “user model”, i.e. as a set of user characteristics - represented as variables - that are required to describe the user of a product. The user model is established by the declaration of the variables and the profile is formally described in a machine-readable and human-readable format. User models can be considered as explicit representations of the properties of an individual user and can be used to reason about the needs, preferences or future behaviour of that user. To this end, VUMS defines a taxonomy of variables comprising of the following categories: Anthropometrics, Motor parameters, Strength parameters, Dexterity/control parameters, Affective parameters, Interaction related states, Hearing parameters, Visual parameters, Cognitive parameters, Equilibrium, Others.

Beyond EU projects, several approaches targeting the formal modelling of user profiles have been presented in the scientific literature, for a variety of application domains. For example, Heckmann et al. proposed an architecture for decentralized user modelling having its basis on the user model markup language (UserML) and the general user model ontology (GUMO)  (Heckmann, Schwartz, Brandherm, & Kröner, 2005). UserML is an RDF-based exchange language for user modelling between decentralized systems, while GUMO (General User Model Ontology) is an ontology expressed in OWL. The study aimed to address uniform interpretation of decentralized user models, and the integration of ubiquitous applications with a user model service.

Sutterer et al. proposed a user profile ontology that is dedicated to describe situation-dependent sub-profiles, aiming to support context-aware adaptive service platforms for mobile communication and information services, so as to automatically trigger the situation-dependent personalization of services (Sutterer, Droegehorn, & David, 2008). The design took into consideration recommendations from the human factors engineering perspective, enabling the specification of situational conditions and situation-dependent user sub-profiles.

Yurchyshyna et al. elaborated on the adaptation of domain ontologies to different contexts and user profiles (Yurchyshyna, Faron-Zucker, Le Thanh, & Zarli, 2010). The authors highlighted the need to reduce the gap between the expert knowledge primary captured in a domain ontology and the end-user knowledge and, consequently, proposed a method to contextualize the ontological knowledge initially acquired and adapt it to different user profiles.

Panagiotopoulos et al. presented an ontology for modelling user profiles capable of encompassing and representing user needs and preferences regarding every “activity sphere” the user participates in, in a way that supports sphere adaptation to the user’s changing context (Panagiotopoulos, Seremeti, & Kameas, 2011). Such a profile represents important permanent user trails like personal information, interests, capabilities, etc., as well as the user role(s) in each sphere and the user’s preferences associated with the sphere tasks. The approach is explicitly targeting Ambient Intelligence (AmI) environments and applications.

Overall, ontology-based user profile modelling involves the definition of concepts that correspond to either static user information, or dynamic features pertaining to temporal conditional preferences and interests, according to specific situations. Interestingly, a study aiming to explicitly assess the reusability of user profile models has been conducted by Hella et al.  (Hella & Krogstie, 2010) The study was based on a systematic review of existing (publicly available) ontologies from a reusability perspective, having the SEQUAL quality framework as the basis for the evaluation. The study concluded that none of the evaluated ontologies satisfied the requirements as models that can be reused or built upon. This particular study indicates the need to build user profile models that are generic and not tightly coupled with the application of discourse, as it has been typically elaborated.

In addition, in practical terms, the need for formal modelling of user profiles is illustrated by Kobsa (Kobsa, 2007), in a review devoted to generic user modelling systems. It describes the purposes of such systems, their services within user-adaptive systems, and the different design requirements for research prototypes and commercial deployments.

Cloud4all approaches the notion of “user profile” as a set of user needs and preferences. Thus, Cloud4all concentrates explicitly its user profile modelling activities on user preferences and needs, rather than proposing a generic user model as elaborated in other efforts presented above. In the scope of A101.2, we elaborated on ontology-based user profile modelling by concentrating on the representation of user needs and preferences per se, and in general terms, i.e. without focusing on a particular, application-specific context, as the majority of the above-mentioned approaches have done. Nevertheless, the targeted applications and (preliminary) scenarios of use considered in the project have been taken into account as the basis for gathering diverse input (from vendors, experts and users) so as to broaden our design towards a generic approach rather than eliminating our work on an application-specific scope. As a consequent step, Cloud4all aims to provide the means for “translating” the generic profile into application-specific contexts, through semantic technologies.




On Jul 6, 2012, at 12:05 PM, Gregg Vanderheiden wrote:

> We seem to have converged on using KEY.   Seems to have many nice aspects -- including the feeling of security it engenders. 
> 
> 
> Jutta   -  have you seen the discussion about using  "SET" instead of 'STATEMENT" ?
> 
> Your thoughts there?
> 
> 
> 
> 
> On Jul 6, 2012, at 1:27 AM, Treviranus, Jutta (Academic) wrote:
> 
>> I like the notion of a key. You can have several keys, they don't need to have your personal information attached to them.  They unlock something or start something. Each key could be attached to a specific PNP. There could be a master key as well. 
>> 
>> Given the sensitivity regarding the term profile and profiling, I'm wondering whether we can avoid using the term. The PNP referred to Personal Needs and Preferences. We have been using the term "statement" rather than "profile", or PNP statement as a less contentious term. Also it was pointed out to us that people are profiled by others but they state what they need and prefer. 
>> 
>> Jutta
>> 
>> 
>> On 2012-07-05, at 7:00 PM, Liddy Nevile wrote:
>> 
>>> I think we simply need to say that they are identifying the needs profile (PNP) they want to use -
>>> 
>>> Liddy
>>> 
>>> On 06/07/2012, at 8:52 AM, Andy Heath wrote:
>>> 
>>>> What is the relationship between the keys and sets of preferences ?
>>>> Is it that one key can be associated with more than one set of preferences ?  Is one set of preferences associated with more than one key ?
>>>> 
>>>> andy
>>>>> 
>>>>> instead of saying "user is identified"
>>>>> 
>>>>> how about  we say "user’spreferencekeyis read (or obtained)"
>>>>> 
>>>>> 
>>>>> In general
>>>>> 
>>>>> We talk about the user "identifying themselves"  and that info being
>>>>> used to fetch their preferences all over the place as the first step
>>>>> 
>>>>> In fact they don't have to identify themselves - they just need to
>>>>> provide a key that can be used to look up a set of preferences.  They
>>>>> don't even have to be their preferences. they could be a generic preference.
>>>>> 
>>>>> This has privacy implications and implies that users identify themselves
>>>>> in order to use GPII when they don't.
>>>>> 
>>>>> 
>>>>> OK to change our language?
>>>>> 
>>>>> Someone have a better phrase to use than     "User preference key"  ?
>>>>> 
>>>>> /Gregg/
>>>>> --------------------------------------------------------
>>>>> Gregg Vanderheiden Ph.D.
>>>>> Director Trace R&D Center
>>>>> Professor Industrial & Systems Engineering
>>>>> and Biomedical Engineering
>>>>> University of Wisconsin-Madison
>>>>> 
>>>>> Co-Director, Raising the Floor - International
>>>>> and the Global Public Inclusive Infrastructure Project
>>>>> http://Raisingthefloor.org   --- http://GPII.net
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> _______________________________________________
>>>>> Accessforall mailing list
>>>>> Accessforall at fluidproject.org
>>>>> http://lists.idrc.ocad.ca/cgi-bin/mailman/listinfo/accessforall
>>>>> 
>>>> 
>>>> 
>>>> 
>>>> Cheers
>>>> 
>>>> andy
>>>> -- 
>>>> __________________
>>>> Andy Heath
>>>> http://axelafa.com
>>>> 
>>>> 
>>>> 
>>>> _______________________________________________
>>>> Accessforall mailing list
>>>> Accessforall at fluidproject.org
>>>> http://lists.idrc.ocad.ca/cgi-bin/mailman/listinfo/accessforall
>>> 
>>> _______________________________________________
>>> Accessforall mailing list
>>> Accessforall at fluidproject.org
>>> http://lists.idrc.ocad.ca/cgi-bin/mailman/listinfo/accessforall
>> 
>> _______________________________________________
>> Profiles mailing list
>> Profiles at lists.gpii.net
>> http://lists.gpii.net/cgi-bin/mailman/listinfo/profiles
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.idrc.ocad.ca/pipermail/accessforall/attachments/20120706/13c16a79/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 7318 bytes
Desc: not available
URL: <http://lists.idrc.ocad.ca/pipermail/accessforall/attachments/20120706/13c16a79/attachment-0001.bin>


More information about the Accessforall mailing list