Online Information London, December 2007

Folksonomies and Ontologies Two New Players in Indexing and Knowledge Representation

Online InformationLondon, December 2007

Katrin WellerInstitute for Language and Information,

Department of Information Science, Heinrich-Heine-University Düsseldorf

Funded by

05.12.2007 Folksonomies and Ontologies 2

Key Aspects

• What are the characteristics of and the differences between folksonomies, taxonomies, thesauri and ontologies?

• How does indexing work in Web 2.0 environments and how in a Semantic Web?

• How can social approaches be combined with enhanced vocabulary control and how may traditional content suppliers profit from this?

• What are the characteristics of and the differences between folksonomies, taxonomies, thesauri and ontologies?

• How does indexing work in Web 2.0 environments and how in a Semantic Web?

• How can social approaches be combined with enhanced vocabulary control and how may traditional content suppliers profit from this?


Knowledge Representation & Document Indexing

• Classical methods: Classification system, thesaurus, nomenclature…Used in professional environments, e.g. libraries, databases.

• Recently two new trends:– Folksonomies: new social dimension (Web 2.0).– Ontologies: more options for formal vocabulary structuring

(Semantic Web).

Modified from figure by Lassila & McGuinness (2001), see also Gomez-Perez et al. (2004), p.28.


Knowledge Representation & Document Indexing

Modified from figure by Lassila & McGuinness (2001), see also Gomez-Perez et al. (2004), p.28.

Controlled vs. uncontrolled vocabularies

Sometimes: Ontologies as generic term.

Sometimes: lightweight vs. Heavyweight ontologies


Knowledge Representation Methods

Distinguishing methods according to the semantic relations in use


Pro and Cons of Folksonomies

Controlled vs. active vocabularyNo vocabulary control but capturing of user‘s contemporary language.

Content indexing vs. data managementNot only content is described in tags, they are used as document descriptions and for all aspects of document management.

Social vs. personal taggingPeople tag for their own use, not for others. Different viewpoints are included.

Retrieval vs. explorationLess about classical retrieval values but more about new entry points to document collections: serendipity.


Information derived from Folksonomies

• Insight to user behavior, language usage – also as resources for other vocabularies

• Grouping of documents according to assigned tags or users that tagged them – new pathways in document collections

• Finding communities with shared interests – people that tag the same document or use the same tags.

• Generating expert profiles from tagging behavior – cloud of tags, that one user has added

• Information on topic developments – e.g. changes in tags assigned to the same document over time


Ontologies – Semantic Structures for the Web

• Formal conceptualizations of knowledge domains, expressed in means of classes, instances, the relations between them and certain rules.


Ontologies – Example


Ontologies – Example


Advantages and Applications?

• Aim: make meaning of information explicit and unambiguous.• Precise domain models. • Not only for interpersonal communication but also for human-

computer interaction.

Applications in Life Sciences• Large amounts of heterogonous data have to be interrelated

explicitly, e.g. nucleotide sequences, amino acid sequences, molecular 3D structures.

• Formal structuring of highly specialized scientific fields. • Currently done: annotation of literature (e.g. UMLS) and gene data

(e.g. gene ontology).


Usage Example: Semantic Wikipedia

Semantic Media Wiki [Krötzsch et al. 2007].

Lists of certain items can be produced because of additional semantic information typed into the source texts.

Lists of certain items can be produced because of additional semantic information typed into the source texts.


Combining different methods?

Current approaches and trends

• New contexts for folksonomies (side by side with other methods)

• Ontologies supporting folksonomies

• Folksonomies supporting ontologies

• Collaborative ontology engineering

Trend: variable degrees of formalism, ontology maturing, emergent semantics.


1. New Contexts for Folksonomies

Folksonomies are discussed for the use within• intranets or other corporate information systems [e.g.Fichter

2006; Peters 2006]

• for usage in professional databases [e.g. Stock 2007]

• libraries [e.g. Kroski 2006; Spiteri 2007].

• The library of the University of Pennsylvania: PennTags• Elsevier’s Engineering Village pioneer in social tagging

within professional databases

Example: Document Collection in Ontoverse. Combinations of original PubMed terms, free tags, and ontology terms

Example: Document Collection in Ontoverse. Combinations of original PubMed terms, free tags, and ontology terms


1. New Contexts for Folksonomies

„Tagging across various and varied applications“ (Gruber 2005)

Example: Group Me• Different Ressouces (photos, bookmarks, wiki-article) can be

tagged within one platform.• They can then be organized as groups, which again can be

tagged.

[Abel et al. 2007]


2. Ontologies Supporting Folksonomies

• Establish some kind of semantic control in folksonomy based systems?

• Use of controlled vocabularies behind the scenes?

Not promising: provide users with guidelines and vocabularies for indexing documents, e.g.



• Establish some kind of semantic control in folksonomy based systems?

• Use of controlled vocabularies behind the scenes?

Use of ontologies behind the scenes, e.g. suggest terms for indexing based on underlying ontologies



Query expansion in folksonomy based systems with help of underlying ontologies.

Example:

Wordflickr – search in flickr with using WordNet for Query Expansion (Kolbitsch 2007)

In search for shoes also synonyms and hierarchical relations are included.


3. Folksonomies Supporting Ontologies

Folksonomies can be statistically analyzed to build up or enrich controlled vocabularies:

• User based suggestions for new terms.• Furthermore: collection of semantic relations implicit in

folksonomies.

(geographic) Hierarchy

Synonyms

Is_A (Hyponymy)

05.12.2007 Folksonomies and Ontologies 20Photos and tags from www.flickr.com

Equivalent? football - soccer

Plays against

Related colours

Name of Stadium – changes over time

Location - (geographic) hierarchy with region

3. Folksonomies supporting Ontologies


• How can ontology engineering profit from current web collaboration tendencies?

• Systems are needed, that support collaborative ontology planning, developing and maintenance.

Approaches: • Ad hoc & lightweight,

e.g. Soboleo.• Multi-level development

process, e.g. Ontoverse.

4. Collaborative Ontology Engineering

Soboleo/ImWissensnetz. Hierarchies can be constructed while tagging web ressources [Zacharias, Braun, 2007]



Ontoverse Ontology Wiki:

1st step – planning of an ontology, informal knowledge collection



Ontoverse Ontology Wiki:

2nd step – formal ontology editing, incl. community awareness functions


Conclusion

• Folksonomies and ontologies form the two ends of a scale of documentation languages ranging from unstructured to highly formalized systems.

• They are not to be seen as rivals but rather as elements in a toolbox which can be used together.

• The key challenge now is to find the right approaches or combinations of approaches to support concrete applications.


GreetingsGreetings fromfrom DDüüsseldorfsseldorf

Questions, suggestions, comments? Contact me: [email protected]

Katrin WellerInstitute for Language and Information

Dept. Information ScienceHeinrich-Heine-University Düsseldorf

Universitätsstraße 1, Building 23.21.04D-40225 Düsseldorf

www.phil-fak.uni-duesseldorf.de/infowissand www.ontoverse.org

mailto:[email protected]

http://www.phil-fak.uni-duesseldorf.de/infowiss

http://www.ontoverse.org/


References

• Abel F, Frank M, Henze N, Krause D, Plappert D, Siehndel P (2007): Group Me! Where Semantic Web meets Web 2.0. In: Proceedings of the 6th International Semantic Web Conference (ISWC 2007), Busan, Korea, Nov. 2007.

• Alexiev V, Breu M et al. (2005): Information Integration with Ontologies. Experiences from an Industrial Showcase, Chichester: Wiley & Sons.

• Al-Khalifa HS, Davis HC (2007): Towards Better Understanding of Folksonomic Patterns. In: Conference on Hypertext and Hypermedia, Proceedings of the 18th Conference on Hypertext and Hypermedia, Manchester, UK: ACM Press, pp.163- 166.

• Christiaens S (2006): Metadata mechanisms: From ontology to folksonomy … and back. In: Lecture Notes in Computer Science, 4277, 199-207.

• Furnas GW, Landauer TK, Gomez LM, Dumais ST (1987): The Vocabulary Problem in Human-System Communication. An Analysis and a Solution. In: Communications of the ACM, 30, 1987, 964-971.

• Golder SA, Huberman BA (2006): The Structure of Collaborative Tagging Systems. Journal of Information Science, 2006 32(2), pp. 198-208.

• Gomez-Perez A, Fernandez-Lopez M, Corcho O (2004): Ontological Engineering. London: Springer.

• Guy M, Tonkin E (2006): Folksonomies. Tidying up Tags? In: D-Lib Magazine 12(1). Available at: http://www.dlib.org/dlib/january06/guy/01guy.html.

• Kolbitsch J (2007): WordFlickr. A Solution to the Vocabulary Problem in Social Tagging Systems. In: Proceedings of I-MEDIA ’07 and I-SEMANTICS ’07, Graz, Austria, Sept. 2007.

• Lassila O, McGuinness D (2001): The Role of Frame-Based Representation on the Semantic Web. Technical Report KSL-01-02. Knowledge Systems Laboratory, Stanford University. Stanford, California.

• Krötzsch, M., Vrandecic, D. et al. (2007): Semantic Wikipedia. In: Journal of Web Semantics 5: 251--261. September 2007.


References (see proceedings for more)

• Kroski E (2006): The Hive Mind. Folksonomies and User-Based Tagging. Retrieved September 17, 2007 from: http://infotangle.blogsome.com/2005/12/07/the-hive- mind-folksonomies-and-user-based-tagging.

• Macgregor, G., McCulloch, E. (2006). Collaborative tagging as a knowledge organisation and resource discovery tool. In: Library Review, 55(5), 291-300.

• Mathes A (2004): Folksonomies. Cooperative Classification and Communication Through Shared Metadata. Retrieved May 7 2007, from http://www.adammathes.com/ academic/computer-mediated- communication/folksonomies.html.

• Paulsen I, Mainz D, Weller K, Mainz I, Kohl J, von Haeseler A (2007): Ontoverse. Collaborative Knowledge Management in the Life Science Network. In: Proceedings of the German eScience Conference 2007, Max Planck Digital Library, ID 316588.0.

• Peters I, Stock WG (2007): Folksonomy and Information Retrieval. In: Proceedings of the 70th Annual Meeting of the American Society for Information Science and Technology (Vol. 45) (CD-ROM).

• Quintarelli E (2005): Folksonomies: Power to the people. Paper presented at the ISKO Italy UniMIB meeting, Milan, June 24, 2005. Online available: http://www.iskoi.org/doc/folksonomies.htm.).

• Vander Wal, T. (2005). Explaining and Showing Broad and Narrow Folksonomies (Blog Post 2005-02-21). Retrieved September 17, 2007 from: http://www.vanderwal.net/random/category.php?cat=153.

• Zacharias V, Braun S (2007): SOBOLEO. Social Bookmarking and Lightweight Ontology Engineering. In: Workshop on Social and Collaborative Construction of Structured Knowledge (CKC), 16th International World Wide Web Conference (WWW 2007), Banff, Alberata, Canada, May 2007.

• Zhang L, Wu X, Yu Y (2006): Emergent Semantics from Folksonomies: A Quantitative Study. Lecture Notes in Computer Science, 4090, 168-186.


Weblinks

• Bibsonomy: http://www.bibsonomy.org• CiteULike: http://www.citeulike.org• Connotea: http://www.connotea.org• DBin: http://www.dbin.org• del.icio.us: http://del.icio.us• Elsevier’s Engineering Village: http://www.engineeringvillage.org• Flickr: http://www.flickr.com• Library of the University of Pennsylvania, PennTags: http://tags.library.upenn.edu/• Ontoverse: http://www.ontoverse.org• OntoWiki: http://ontowiki.net/Projects/OntoWiki• OWL, Web Ontology Language: http://www.w3.org/TR/owl-ref/• PennTag: http://tags.library.upenn.edu• Protégé: http://protege.stanford.edu/• Semantic Media Wiki (Project): http://ontoworld.org/wiki/Semantic_MediaWiki• Soboleo: http://soboleo.fzi.de:8080/webPortal/• Tagline Generator: http://chir.ag/tech/download/tagline• Technorati: http://technorati.com/• Upcoming: http://upcoming.yahoo.com• WordFlickr (Demo): http://www.kolbitsch.org/cgi-bin/search• YouTube: http://www.youtube.com

http://www.bibsonomy.org/

http://www.citeulike.org/

http://www.connotea.org/

http://www.dbin.org/

http://del.icio.us/

http://www.engineeringvillage.org/

http://www.flickr.com/

http://tags.library.upenn.edu/

http://www.ontoverse.org/

http://ontowiki.net/Projects/OntoWiki

http://www.w3.org/TR/owl-ref/

http://tags.library.upenn.edu/

http://protege.stanford.edu/

http://ontoworld.org/wiki/Semantic_MediaWiki

http://soboleo.fzi.de:8080/webPortal/

http://chir.ag/tech/download/tagline

http://technorati.com/

http://upcoming.yahoo.com/

http://www.kolbitsch.org/cgi-bin/search

http://www.youtube.com/


Acknowledgements

Thanks to my colleagues at the Dept. of Information Science, Heinrich-Heine-University Düsseldorf, particularly Wolfgang G. Stock and Isabella Peters

and to my colleagues in the ONTOVERSE research project, particularly Indra Mainz, Dominic Mainz and Ingo Paulsen!

Documents

Online Information London, December 2007