actor_concept_instance_model

Tripartite models, simple ontologies & emergent semantics
870 words 1160707511 'single quotes' “double quotes”

Peter Mika (2005) explores a vision for incorporating an extra 'social element' into the common Semantic Web model of formal knowledge representation. He argues ontologies are “inseparable from the [community] context … in which they are created and used,” (sect. 5, conclusion.) To Mika the creation of a 'community-based ontology' is a process of meaning-negotiation; a process that is both incremental and social. Mika observes, based on his own analysis of content on the World Wide Web, "that RSS and FOAF (and various extensions) constitute the overwhelming majority of metadata, originating mostly from the technologically-aware blog community," (Mika, 2005, sect. 2). Mika speculates that "ordinary web users" find it difficult to adopt "even simple knowledge representation paradigms such as taxonomies," (Sect. 2). Finally, he argues that there needs to be some way to understand how simple acts by ordinary individual users of online classifications systems can be understood as a social knowledge sharing practice.

The problem
The dominant view of meaning creation within the Semantic Web research community of semantics largely based around a bipartite 'concept-instance' model, which Mika sees as a conceptual abstraction that overlooks author’ contribution to creation concept-instances that make up ontologies' ontological metadata. To Mika any potential for Semantic Web technologies to support social processes of meaning creation is constrained by this traditional view.

Mika argues that the dominant model of Semantic Web technologies ignores authors’ social context, be it online or face to face (F2F). He suggests that to use such a model as the basis on which to build ‘community’ web portals, for example, would appear misguided. He adds, that when it comes to implementing systems that combine a vision of community and formal semantics/ontologies, there is “a large conceptual gap in the literature between the vision” (intro. sect.) and the reality, especially when it comes to systems aimed at populations of “ordinary web users,” (sect. 2).

Mika's examined the role content classification by users of applications like CiteULike, Flickr™, and Del.icio.us™ (e.g. assigning a particular keyword to a specific book in CiteULike, image in Flickr™, or URL in Del.icio.us™).

Mika acknowledges the (potential) community vision and the role of and collaborative categorisations for the development of folksonomies by users of systems such as Del.icio.us™, Flickr™, and CiteULike.

Mika's argument highlights how social tagging systems enable the collective activities of a “set of actors (users),” create of a “set of concepts (tags, keywords)” (Mika, 2005, sect. 2) through their independent actions.

The vision
Mika's a-c-i (actor-concept-instance) model

Mika proposes a tripartite "actor-concept-instance" model and calls for online applications to be developed to support the incremental emergence of "community-based ontologies"(Abstract section).

The A-C-I model privileges social-acts of meaning creation over the creation of formal representations of community knowledge through formal ontological, or semantic relations (see discussion on RDF above). Mika argues that when it comes to implementing systems that combine a vision of community and formal semantics/ontologies, there is "a large conceptual gap in the literature between the vision" (intro. section) and the reality, especially when it comes to systems aimed at populations of “ordinary web users," (intro. section?).

Mika (2005) investigates advantages of incorporating social context into the knowledge representation, and he speculating that ontologies are "inseparable from the context of the community in which they are created and used," (Mika, 2005, sect. 5, conclusion.)

Rashmi Sinha concurs, except she speculates that to average users the task of creating or choosing a tag carries less 'cognitive load' than choosing a classification from within taxonomy (Sinha, 2006). What is needed is some way to understand how simple acts by ordinary web users of online classifications systems.

To extend the bipartite model (mentioned above) Mika adds the actor as an additional element, thus he suggest an extra social dimension in addition to the concept and its instance.

Any potential for semantic web technologies to support social processes of meaning creation with is constrained by what Mika sees as a view of semantics based largely around the traditional bipartite 'concept-instance', model a conceptual abstraction that overlooks authors' contribution in creating concept instances. It is also a model that ignores the authors' social context, be it online or face to face (F2F); to use such a model as the basis on which to build 'community' web portals for example would appear misguided.

Mika acknowledges the (potential) community vision and the role of and collaborative categorisations for the development of folksonomies by users of systems such as Del.icio.us™, Flickr™, and CiteULike. In Mika's vision, the creation of meaning in online spaces results from the intertwined activities of a community of agents:

> The vision is a community of self-organizing, autonomous, networked and localized agents co-operating in dynamic, open environments, each organizing knowledge (e.g. document instances) according to a self-established ontology, establishing connections and negotiating meaning only when it becomes necessary for co-operation. (Intro. sect.)

This vision privileges content classification as an act of meaning creation that should not be conceptualised as being separated from its social context. In other words the creation of formal representations of community knowledge through formal ontological, or semantic relations (see discussion on RDF) needs to consider the social context in which acts of knowledge representation occur.

//duplication?// Mika proposes a tripartite “actor-concept-instance” model and calls for the online applications that developed in a way that supports the incremental emergence of “community-based ontologies”  (Abstract section). Mika speculates that “ ordinary web users”   find it difficult to adopt “  even simple knowledge representation paradigms such as taxonomies”   (Sect. 2).