Debbie Journal Presentation

From Semantic Portal Wiki

Presentation given at CSCI 6966 Advanced Semantic Web (Fall 2008)#Lesson 13

• Speaker: Debbie Heisler
• Title: Ontologies are us: A unified model of social networks and semantics
• Authors: Peter Mika
• Conference: Journal of Web Semantics 5 (1), 2007
• URL: http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B758F-4MYF67P-1&_user=10&_coverDate=03%2F31%2F2007&_rdoc=3&_fmt=high&_orig=browse&_srch=doc-info(%23toc%2312925%232007%23999949998%23645422%23FLA%23display%23Volume)&_cdi=12925&_sort=d&_docanchor=&_ct=6&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=737d4e9bfc8916ca8c6d47a892c072e0
• Date of Presentation: 2008/12/04

Questions

ID	Question	Name	Answer
Debbie Journal Presentation GTW 1	Section 2 claims "it is important to note that in terms of knowledge representation, the set of these keywords cannot even be considered as vocabularies, the simplest possible form of an ontology on the continuous scale of Smith and Welty." Footnote 4 in section 3.1 discusses the crawled del.icio.us data as "the largest ontology annotation data set ever studied" (emphasis added), and section 5 continues "in fact, folksonomies are ontologies." Is the distinction between folksonomies and ontologies important (especially in light of folksonomies' lack of one-to-one mapping of tags and concepts)? If so, how are we to understand this apparent confusion in terminology in discussing del.icio.us tag data?	Gregory Todd Williams
Debbie Journal Presentation Jesse Weaver	In section 5, the author feels the need to discuss the debate between the roles/similarities/etc. of folksonomies and ontologies. The author seems to say (more or less) that even though folksonomies are not as stable, formal, or sharable (i.e., limited sharing scope) as (traditional) ontologies, they benefit from one major advantage: "In particular, it is the first time that the barriers of providing knowledge have been lowered to such an extent that ordinary users are willing to provide metadata on web resources on a large scale." The approach in this paper shows that while "folksonomies are lightweight," "there is more semantics in folksonomies than meets the eye." What do you think about these claims? Are they accurate? Do they convince you that folksonomies are more useful than others may have initially assumed? While folksonomies may enable ordinary users to provide metadata on a large scale, it seems that the limited sharing scope may outweigh this fact. "Tagging systems at the moment represent islands of semantics that do not cross the boundaries of a single website." Can this be remedied? If not, how do folksonomies enable sharing of knowledge between agents as described in the introduction?	Jesse Weaver
Debbie Journal Presentation Joshua Shinavier 1	The ability to infer broader/narrower relationships among tags seems promising, but it's not clear to me how the author achieves this using the presented framework. Some additional discussion of the technique involved would help to understand the significance of the experimental results.	Joshua Shinavier
Debbie Journal Presentation Joshua Taylor 1	The authors are at the point of analysis where they can begin to look at how certain groups of people tag various articles and the like. Rather than evaluating a single produced ontology, could this technology be used to extract multiple ontologies? After all, if a Semantic Web researchers tag articles methodically according to their own terminology, and a group of arachnologists does likewise, it would be good to notice that both groups use the tag "web", but that they use them in different ways. Could the Semantic Web and Arachnology ontologies be extracted as independent entities?	Joshua A. Taylor
Debbie Journal Presentation Joshua Taylor 2	Can the more sophisticated tripartite model capture everything of the earlier bipartite model? (This is a typical test of whether one framework is more general than another — that is, it is more general if the less general model can be captured as a special case.) Also, in the bipartite model, associations between instances and concepts were tracked. In the tripartite model, the same associations are tracked, but with the additional information of who made the association. How is this different than a provenance model which tracks the same thing? Or, if addressing ontology drift is one of the author's concerns, why not use a quadripartite model that also tracks time? Then ontologies could be extracted with temporal information and ontologies from different times compared.	Joshua A. Taylor
Debbie Journal Presentation Question Shangguan	I'm a little skeptical about Section 5: ontologies and folksonomies, in which the author makes several statements: "The methods to extract it (semantics) are not logic based, but borrowed from network analysis and in fact folksonomes may require a shift in the methods applied by the semantic web community." -- What is the *shift* mentioned here? Is it general techniques already widely adopted in the Social Network Analysis (SNA) community? Or is it an improved variant of the techniques in the SW community plus some ideas borrowed from the SNA community? The author also mentions three features of folksonomies: lighweight, dynamic, and limited sharing scope. While I'm ok with the first one, the other two features are kind of debatable, at least to me. Dynamic: "Much like ontologies that stabilized after long, heated debates, these stable parts of folksonomies can in fact be considered as lightweight ontologies that represent a consensus over the meaning of terms". -- What if the stablized part becomes so small that it becomes even *useless*? Limited Sharing Scope: "...the mapping of folksonomies is predictably difficult, if not impossible... Folksonomies, however, will have a significant impact on realizing the original vision of the Semantic Web...". With this *limited* interoperability, can folksonomies really help to implement the original vision of SW. (Original vision of SW: "The Semantic Web provides a common framework that allows data to be shared and reused across application, enterprise, and community boundaries." from http://www.w3.org/2001/sw)	Zhenning Shangguan
Ontologyareus Jiao	In the evaluation section, the answers to question "In terms of the association between the concepts, which ontology of Semantic Web related concepts do you consider more accurate?" are used as metrics to demonstrate community-based ontology extraction method is better than item-based ontology extraction method. It's not surprising that community-based one is more accurate since it incorporates one more element, actor. Probably improving accurateness can be achieved by many ways, for example, provenance. And is "accurateness" enough for the comparison? How about the other issues, like scalability, tool support, etc.?	Jiao Tao
Social Network Semantics Ankesh	I think the paper presents a great piece of work. The idea and results are superb. However, I am skeptical on the use of the word ontology. Does the word Ontology trouble you too? Semantic web is about linking resources and documents on the web with more than just hyperlinks. The paper presents techniques to relate concepts as close (similar) and generic-specifc (subclass). It creates a thesauri and extends it with subclass relationships. To me it can be called more than just a thesauri only if it contains relations other than similarity and subclass between its concepts. A system that supports tagging can use the ontologies of concepts to classify things better (things tagged by similar terms/ concepts can be kept together). Which other semantic web application can use the ontologies of concepts?	Ankesh Khandelwal