@inproceedings{benz2010semantics, author = {Benz, Dominik and Hotho, Andreas and Stützer, Stefan and Stumme, Gerd}, title = {Semantics made by you and me: Self-emerging ontologies can capture the diversity of shared knowledge}, booktitle = {Proceedings of the 2nd Web Science Conference (WebSci10)}, address = {Raleigh, NC, USA}, year = {2010}, url = {http://www.kde.cs.uni-kassel.de/pub/pdf/benz2010semantics.pdf}, keywords = {2010, ibm-kde-tagging, itegpub, myown, ol_web2.0, semantics, web_science, websci, websci10} }

%0 = inproceedings %A = Benz, Dominik and Hotho, Andreas and Stützer, Stefan and Stumme, Gerd %B = Proceedings of the 2nd Web Science Conference (WebSci10) %C = Raleigh, NC, USA %D = 2010 %T = Semantics made by you and me: Self-emerging ontologies can capture the diversity of shared knowledge %U = http://www.kde.cs.uni-kassel.de/pub/pdf/benz2010semantics.pdf

Recent research provides evidence for the presence of emergent semantics in collaborative tagging systems. While several methods have been proposed, little is known about the factors that influence the evolution of semantic structures in these systems. A natural hypothesis is that the quality of the emergent semantics depends on the pragmatics of tagging: Users with certain usage patterns might contribute more to the resulting semantics than others. In this work, we propose several measures which enable a pragmatic differentiation of taggers by their degree of contribution to emerging semantic structures. We distinguish between categorizers, who typically use a small set of tags as a replacement for hierarchical classification schemes, and describers, who are annotating resources with a wealth of freely associated, descriptive keywords. To study our hypothesis, we apply semantic similarity measures to 64 different partitions of a real-world and large-scale folksonomy containing different ratios of categorizers and describers. Our results not only show that ‘verbose’ taggers are most useful for the emergence of tag semantics, but also that a subset containing only 40% of the most ‘verbose’ taggers can produce results that match and even outperform the semantic precision obtained from the whole dataset. Moreover, the results suggest that there exists a causal link between the pragmatics of tagging and resulting emergent semantics. This work is relevant for designers and analysts of tagging systems interested (i) in fostering the semantic development of their platforms, (ii) in identifying users introducing “semantic noise??, and (iii) in learning ontologies.

@inproceedings{koerner2010stop, author = {Körner, Christian and Benz, Dominik and Strohmaier, Markus and Hotho, Andreas and Stumme, Gerd}, title = {Stop Thinking, start Tagging - Tag Semantics emerge from Collaborative Verbosity}, booktitle = {Proceedings of the 19th International World Wide Web Conference (WWW 2010)}, publisher = {ACM}, address = {Raleigh, NC, USA}, year = {2010}, url = {http://www.kde.cs.uni-kassel.de/pub/pdf/koerner2010stop.pdf}, keywords = {2010, collaborative_verbosity, emergentsemantics_factors, ibm-kde-tagging, itegpub, myown, ol_web2.0, www, www2010}, abstract = {Recent research provides evidence for the presence of emergent semantics in collaborative tagging systems. While several methods have been proposed, little is known about the factors that influence the evolution of semantic structures in these systems. A natural hypothesis is that the quality of the emergent semantics depends on the pragmatics of tagging: Users with certain usage patterns might contribute more to the resulting semantics than others. In this work, we propose several measures which enable a pragmatic differentiation of taggers by their degree of contribution to emerging semantic structures. We distinguish between categorizers, who typically use a small set of tags as a replacement for hierarchical classification schemes, and describers, who are annotating resources with a wealth of freely associated, descriptive keywords. To study our hypothesis, we apply semantic similarity measures to 64 different partitions of a real-world and large-scale folksonomy containing different ratios of categorizers and describers. Our results not only show that ‘verbose’ taggers are most useful for the emergence of tag semantics, but also that a subset containing only 40% of the most ‘verbose’ taggers can produce results that match and even outperform the semantic precision obtained from the whole dataset. Moreover, the results suggest that there exists a causal link between the pragmatics of tagging and resulting emergent semantics. This work is relevant for designers and analysts of tagging systems interested (i) in fostering the semantic development of their platforms, (ii) in identifying users introducing “semantic noise??, and (iii) in learning ontologies.} }

%0 = inproceedings %A = Körner, Christian and Benz, Dominik and Strohmaier, Markus and Hotho, Andreas and Stumme, Gerd %B = Proceedings of the 19th International World Wide Web Conference (WWW 2010) %C = Raleigh, NC, USA %D = 2010 %I = ACM %T = Stop Thinking, start Tagging - Tag Semantics emerge from Collaborative Verbosity %U = http://www.kde.cs.uni-kassel.de/pub/pdf/koerner2010stop.pdf

@inproceedings{benz2009characterizing, author = {Benz, Dominik and Krause, Beate and Kumar, G. Praveen and Hotho, Andreas and Stumme, Gerd}, title = {Characterizing Semantic Relatedness of Search Query Terms}, booktitle = {Proceedings of the 1st Workshop on Explorative Analytics of Information Networks (EIN2009)}, address = {Bled, Slovenia}, year = {2009}, url = {http://www.kde.cs.uni-kassel.de/pub/pdf/benz2009characterizing.pdf}, keywords = {2009, ecml_pkdd, ein, itegpub, logsonomies, myown, ol_web2.0, similarity_measures, widely_related, workshop} }

%0 = inproceedings %A = Benz, Dominik and Krause, Beate and Kumar, G. Praveen and Hotho, Andreas and Stumme, Gerd %B = Proceedings of the 1st Workshop on Explorative Analytics of Information Networks (EIN2009) %C = Bled, Slovenia %D = 2009 %T = Characterizing Semantic Relatedness of Search Query Terms %U = http://www.kde.cs.uni-kassel.de/pub/pdf/benz2009characterizing.pdf

Social bookmarking systems and their emergent information structures, known as folksonomies, are increasingly important data sources for Semantic Web applications. A key question for harvesting semantics from these systems is how to extend and adapt traditional notions of similarity to folksonomies, and which measures are best suited for applications such as navigation support, semantic search, and ontology learning. Here we build an evaluation framework to compare various general folksonomy-based similarity measures derived from established information-theoretic, statistical, and practical measures. Our framework deals generally and symmetrically with users, tags, and resources. For evaluation purposes we focus on similarity among tags and resources, considering different ways to aggregate annotations across users. After comparing how tag similarity measures predict user-created tag relations, we provide an external grounding by user-validated semantic proxies based on WordNet and the Open Directory. We also investigate the issue of scalability. We ?nd that mutual information with distributional micro-aggregation across users yields the highest accuracy, but is not scalable; per-user projection with collaborative aggregation provides the best scalable approach via incremental computations. The results are consistent across resource and tag similarity.

@inproceedings{markines2009evaluating, author = {Markines, Benjamin and Cattuto, Ciro and Menczer, Filippo and Benz, Dominik and Hotho, Andreas and Stumme, Gerd}, title = {Evaluating Similarity Measures for Emergent Semantics of Social Tagging}, booktitle = {18th International World Wide Web Conference}, year = {2009}, pages = {641--641}, url = {http://www.kde.cs.uni-kassel.de/pub/pdf/markines2009evaluating.pdf}, keywords = {2009, itegpub, methods_concepts, myown, ol_web2.0, semantic_relatedness, social_similarity, tagorapub, www2009}, abstract = {Social bookmarking systems and their emergent information structures, known as folksonomies, are increasingly important data sources for Semantic Web applications. A key question for harvesting semantics from these systems is how to extend and adapt traditional notions of similarity to folksonomies, and which measures are best suited for applications such as navigation support, semantic search, and ontology learning. Here we build an evaluation framework to compare various general folksonomy-based similarity measures derived from established information-theoretic, statistical, and practical measures. Our framework deals generally and symmetrically with users, tags, and resources. For evaluation purposes we focus on similarity among tags and resources, considering different ways to aggregate annotations across users. After comparing how tag similarity measures predict user-created tag relations, we provide an external grounding by user-validated semantic proxies based on WordNet and the Open Directory. We also investigate the issue of scalability. We ?nd that mutual information with distributional micro-aggregation across users yields the highest accuracy, but is not scalable; per-user projection with collaborative aggregation provides the best scalable approach via incremental computations. The results are consistent across resource and tag similarity.} }

%0 = inproceedings %A = Markines, Benjamin and Cattuto, Ciro and Menczer, Filippo and Benz, Dominik and Hotho, Andreas and Stumme, Gerd %B = 18th International World Wide Web Conference %D = 2009 %T = Evaluating Similarity Measures for Emergent Semantics of Social Tagging %U = http://www.kde.cs.uni-kassel.de/pub/pdf/markines2009evaluating.pdf

Several learning tasks comprise hierarchies. Comparison with a "goldstandard" is often performed to evaluate the quality of a learned hierarchy. We assembled various similarity metrics that have been proposed in different disciplines and compared them in a unified interdisciplinary framework for hierarchical evaluation which is based on the distinction of three fundamental dimensions. Identifying deficiencies for measuring structural similarity, we suggest three new measures for this purpose, either extending existing ones or based on new ideas. Experiments with an artificial dataset were performed to compare the different measures. As shown by our results, the measures vary greatly in their properties.

@inproceedings{bade2008evaluation, author = {Bade, Korinna and Benz, Dominik}, title = {Evaluation Strategies for Learning Algorithms of Hierarchical Structures}, booktitle = {Proceedings of the 32nd Annual Conference of the German Classification Society - Advances in Data Analysis, Data Handling and Business Intelligence (GfKl 2008)}, series = {Studies in Classification, Data Analysis, and Knowledge Organization}, publisher = {Springer}, address = {Berlin-Heidelberg}, year = {2008}, note = {in press}, url = {http://www.kde.cs.uni-kassel.de/pub/pdf/bade2008evaluation.pdf}, keywords = {2008, evaluation_strategies, hiearchies, itegpub, myown, ol_web2.0, widely_related}, abstract = {Several learning tasks comprise hierarchies. Comparison with a "goldstandard" is often performed to evaluate the quality of a learned hierarchy. We assembled various similarity metrics that have been proposed in different disciplines and compared them in a unified interdisciplinary framework for hierarchical evaluation which is based on the distinction of three fundamental dimensions. Identifying deficiencies for measuring structural similarity, we suggest three new measures for this purpose, either extending existing ones or based on new ideas. Experiments with an artificial dataset were performed to compare the different measures. As shown by our results, the measures vary greatly in their properties.} }

%0 = inproceedings %A = Bade, Korinna and Benz, Dominik %B = Proceedings of the 32nd Annual Conference of the German Classification Society - Advances in Data Analysis, Data Handling and Business Intelligence (GfKl 2008) %C = Berlin-Heidelberg %D = 2008 %I = Springer %T = Evaluation Strategies for Learning Algorithms of Hierarchical Structures %U = http://www.kde.cs.uni-kassel.de/pub/pdf/bade2008evaluation.pdf

The objective of our group was to exploit state-of-the-art Information Retrieval methods for finding associations and dependencies between tags, capturing and representing differences in tagging behavior and vocabulary of various folksonomies, with the overall aim to better understand the semantics of tags and the tagging process. Therefore we analyze the semantic content of tags in the Flickr and Delicious folksonomies. We find that: tag context similarity leads to meaningful results in Flickr, despite its narrow folksonomy character; the comparison of tags across Flickr and Delicious shows little semantic overlap, being tags in Flickr associated more to visual aspects rather than technological as it seems to be in Delicious; there are regions in the tag-tag space, provided with the cosine similarity metric, that are characterized by high density; the order of tags inside a post has a semantic relevance.

@inproceedings{benz2008analyzing, author = {Benz, Dominik and Grobelnik, Marko and Hotho, Andreas and Jäschke, Robert and Mladenic, Dunja and Servedio, Vito D. P. and Sizov, Sergej and Szomszor, Martin}, title = {Analyzing Tag Semantics Across Collaborative Tagging Systems}, editor = {Alani, Harith and Staab, Steffen and Stumme, Gerd}, booktitle = {Proceedings of the Dagstuhl Seminar on Social Web Communities}, year = {2008}, number = {08391}, url = {http://www.kde.cs.uni-kassel.de/pub/pdf/benz2008analyzing.pdf}, keywords = {2008, dagstuhl, iin2009, itegpub, myown, ol_web2.0, tag_semantics, tagorapub, widely_related}, abstract = {The objective of our group was to exploit state-of-the-art Information Retrieval methods for finding associations and dependencies between tags, capturing and representing differences in tagging behavior and vocabulary of various folksonomies, with the overall aim to better understand the semantics of tags and the tagging process. Therefore we analyze the semantic content of tags in the Flickr and Delicious folksonomies. We find that: tag context similarity leads to meaningful results in Flickr, despite its narrow folksonomy character; the comparison of tags across Flickr and Delicious shows little semantic overlap, being tags in Flickr associated more to visual aspects rather than technological as it seems to be in Delicious; there are regions in the tag-tag space, provided with the cosine similarity metric, that are characterized by high density; the order of tags inside a post has a semantic relevance.} }

%0 = inproceedings %A = Benz, Dominik and Grobelnik, Marko and Hotho, Andreas and Jäschke, Robert and Mladenic, Dunja and Servedio, Vito D. P. and Sizov, Sergej and Szomszor, Martin %B = Proceedings of the Dagstuhl Seminar on Social Web Communities %D = 2008 %T = Analyzing Tag Semantics Across Collaborative Tagging Systems %U = http://www.kde.cs.uni-kassel.de/pub/pdf/benz2008analyzing.pdf

Social bookmarking systems allow users to organise collections of resources on the Web in a collaborative fashion. The increasing popularity of these systems as well as first insights into their emergent semantics have made them relevant to disciplines like knowledge extraction and ontology learning. The problem of devising methods to measure the semantic relatedness between tags and characterizing it semantically is still largely open. Here we analyze three measures of tag relatedness: tag co-occurrence, cosine similarity of co-occurrence distributions, and FolkRank, an adaptation of the PageRank algorithm to folksonomies. Each measure is computed on tags from a large-scale dataset crawled from the social bookmarking system del.icio.us. To provide a semantic grounding of our findings, a connection to WordNet (a semantic lexicon for the English language) is established by mapping tags into synonym sets of WordNet, and applying there well-known metrics of semantic similarity. Our results clearly expose different characteristics of the selected measures of relatedness, making them applicable to different subtasks of knowledge extraction such as synonym detection or discovery of concept hierarchies.

@inproceedings{cattuto2008semantic, author = {Cattuto, Ciro and Benz, Dominik and Hotho, Andreas and Stumme, Gerd}, title = {Semantic Analysis of Tag Similarity Measures in Collaborative Tagging Systems}, booktitle = {Proceedings of the 3rd Workshop on Ontology Learning and Population (OLP3)}, address = {Patras, Greece}, year = {2008}, pages = {39--43}, note = {ISBN 978-960-89282-6-8}, url = {http://www.kde.cs.uni-kassel.de/pub/pdf/cattuto2008semantic.pdf}, isbn = {978-960-89282-6-8}, keywords = {2008, itegpub, methods_concepthierarchy, methods_concepts, myown, ol_web2.0, olp3, semantic_relatedness, tag_relatedness, tagorapub}, abstract = {Social bookmarking systems allow users to organise collections of resources on the Web in a collaborative fashion. The increasing popularity of these systems as well as first insights into their emergent semantics have made them relevant to disciplines like knowledge extraction and ontology learning. The problem of devising methods to measure the semantic relatedness between tags and characterizing it semantically is still largely open. Here we analyze three measures of tag relatedness: tag co-occurrence, cosine similarity of co-occurrence distributions, and FolkRank, an adaptation of the PageRank algorithm to folksonomies. Each measure is computed on tags from a large-scale dataset crawled from the social bookmarking system del.icio.us. To provide a semantic grounding of our findings, a connection to WordNet (a semantic lexicon for the English language) is established by mapping tags into synonym sets of WordNet, and applying there well-known metrics of semantic similarity. Our results clearly expose different characteristics of the selected measures of relatedness, making them applicable to different subtasks of knowledge extraction such as synonym detection or discovery of concept hierarchies.} }

%0 = inproceedings %A = Cattuto, Ciro and Benz, Dominik and Hotho, Andreas and Stumme, Gerd %B = Proceedings of the 3rd Workshop on Ontology Learning and Population (OLP3) %C = Patras, Greece %D = 2008 %T = Semantic Analysis of Tag Similarity Measures in Collaborative Tagging Systems %U = http://www.kde.cs.uni-kassel.de/pub/pdf/cattuto2008semantic.pdf

Collaborative tagging systems have nowadays become important data sources for populating semantic web applications. For taskslike synonym detection and discovery of concept hierarchies, many researchers introduced measures of tag similarity. Eventhough most of these measures appear very natural, their design often seems to be rather ad hoc, and the underlying assumptionson the notion of similarity are not made explicit. A more systematic characterization and validation of tag similarity interms of formal representations of knowledge is still lacking. Here we address this issue and analyze several measures oftag similarity: Each measure is computed on data from the social bookmarking system del.icio.us and a semantic grounding isprovided by mapping pairs of similar tags in the folksonomy to pairs of synsets in Wordnet, where we use validated measuresof semantic distance to characterize the semantic relation between the mapped tags. This exposes important features of theinvestigated similarity measures and indicates which ones are better suited in the context of a given semantic application.

@inproceedings{cattuto2008semantica, author = {Cattuto, Ciro and Benz, Dominik and Hotho, Andreas and Stumme, Gerd}, title = {Semantic Grounding of Tag Relatedness in Social Bookmarking Systems}, editor = {Sheth, Amit P. and Staab, Steffen and Dean, Mike and Paolucci, Massimo and Maynard, Diana and Finin, Timothy W. and Thirunarayan, Krishnaprasad}, booktitle = {The Semantic Web -- ISWC 2008, Proc.Intl. Semantic Web Conference 2008}, series = {LNAI}, publisher = {Springer}, address = {Heidelberg}, year = {2008}, volume = {5318}, pages = {615--631}, url = {http://www.kde.cs.uni-kassel.de/pub/pdf/cattuto2008semantica.pdf}, doi = {http://dx.doi.org/10.1007/978-3-540-88564-1_39}, keywords = {2008, grounding, iswc2008, itegpub, methods_concepthierarchy, methods_concepts, myown, ol_web2.0, relatedness, semantic, semantic_relatedness, similarity, sw, tag, tagging, tagorapub}, abstract = {Collaborative tagging systems have nowadays become important data sources for populating semantic web applications. For taskslike synonym detection and discovery of concept hierarchies, many researchers introduced measures of tag similarity. Eventhough most of these measures appear very natural, their design often seems to be rather ad hoc, and the underlying assumptionson the notion of similarity are not made explicit. A more systematic characterization and validation of tag similarity interms of formal representations of knowledge is still lacking. Here we address this issue and analyze several measures oftag similarity: Each measure is computed on data from the social bookmarking system del.icio.us and a semantic grounding isprovided by mapping pairs of similar tags in the folksonomy to pairs of synsets in Wordnet, where we use validated measuresof semantic distance to characterize the semantic relation between the mapped tags. This exposes important features of theinvestigated similarity measures and indicates which ones are better suited in the context of a given semantic application.} }

%0 = inproceedings %A = Cattuto, Ciro and Benz, Dominik and Hotho, Andreas and Stumme, Gerd %B = The Semantic Web -- ISWC 2008, Proc.Intl. Semantic Web Conference 2008 %C = Heidelberg %D = 2008 %I = Springer %T = Semantic Grounding of Tag Relatedness in Social Bookmarking Systems %U = http://www.kde.cs.uni-kassel.de/pub/pdf/cattuto2008semantica.pdf

The emergence of collaborative tagging systems with their underlying flat and uncontrolled resource organization paradigm has led to a large number of research activities focussing on a formal description and analysis of the resulting “folksonomies??. An interesting outcome is that the characteristic qualities of these systems seem to be inverse to more traditional knowledge structuring approaches like taxonomies or ontologies: The latter provide rich and precise semantics, but suffer - amongst others - from a knowledge acquisition bottleneck. An important step towards exploiting the possible synergies by bridging the gap between both paradigms is the automatic extraction of relations between tags in a folksonomy. This position paper presents preliminary results of ongoing work to induce hierarchical relationships among tags by analyzing the aggregated data of collaborative tagging systems as a basis for an ontology learning procedure.

@inproceedings{benz2007position, author = {Benz, Dominik and Hotho, Andreas}, title = {Position Paper: Ontology Learning from Folksonomies}, editor = {Hinneburg, Alexander}, booktitle = {Workshop Proceedings of Lernen - Wissensentdeckung - Adaptivität (LWA 2007)}, publisher = {Martin-Luther-Universität Halle-Wittenberg}, year = {2007}, pages = {109--112}, note = {http://lwa07.informatik.uni-halle.de/kdml07/kdml07.htm}, url = {http://www.kde.cs.uni-kassel.de/pub/pdf/benz2007position.pdf}, isbn = {978-3-86010-907-6}, keywords = {2007, diploma_thesis, ibm-kde-tagging, itegpub, methods_concepthierarchy, myown, ol_web2.0, ontology_learning, rel1, rel2, tagorapub}, abstract = {The emergence of collaborative tagging systems with their underlying flat and uncontrolled resource organization paradigm has led to a large number of research activities focussing on a formal description and analysis of the resulting “folksonomies??. An interesting outcome is that the characteristic qualities of these systems seem to be inverse to more traditional knowledge structuring approaches like taxonomies or ontologies: The latter provide rich and precise semantics, but suffer - amongst others - from a knowledge acquisition bottleneck. An important step towards exploiting the possible synergies by bridging the gap between both paradigms is the automatic extraction of relations between tags in a folksonomy. This position paper presents preliminary results of ongoing work to induce hierarchical relationships among tags by analyzing the aggregated data of collaborative tagging systems as a basis for an ontology learning procedure.} }

%0 = inproceedings %A = Benz, Dominik and Hotho, Andreas %B = Workshop Proceedings of Lernen - Wissensentdeckung - Adaptivität (LWA 2007) %D = 2007 %I = Martin-Luther-Universität Halle-Wittenberg %T = Position Paper: Ontology Learning from Folksonomies %U = http://www.kde.cs.uni-kassel.de/pub/pdf/benz2007position.pdf

Social bookmark tools are rapidly emerging on the Web. In suchsystems users are setting up lightweight conceptual structurescalled folksonomies. The reason for their immediate success is thefact that no specific skills are needed for participating. In thispaper we specify a formal model for folksonomies, briefly describeour own system BibSonomy, which allows for sharing both bookmarks andpublication references, and discuss first steps towards emergent semantics.

@inproceedings{hotho2006emergent, author = {Hotho, Andreas and Jäschke, Robert and Schmitz, Christoph and Stumme, Gerd}, title = {Emergent Semantics in BibSonomy}, editor = {Hochberger, Christian and Liskowsky, Rüdiger}, booktitle = {Informatik 2006 -- Informatik für Menschen. Band 2}, series = {Lecture Notes in Informatics}, publisher = {Gesellschaft für Informatik}, address = {Bonn}, year = {2006}, volume = {P-94}, note = {Proc. Workshop on Applications of Semantic Technologies, Informatik 2006}, url = {http://www.kde.cs.uni-kassel.de/stumme/papers/2006/hotho2006emergent.pdf}, keywords = {2006, UniK, bibsonomy, emergence, emergent, emergentsemantics_evidence, folksonomy, hotho, itegpub, jaeschke, l3s, myown, nepomuk, ol_web2.0, schmitz, semantics, stumme, tagorapub}, abstract = {Social bookmark tools are rapidly emerging on the Web. In suchsystems users are setting up lightweight conceptual structurescalled folksonomies. The reason for their immediate success is thefact that no specific skills are needed for participating. In thispaper we specify a formal model for folksonomies, briefly describeour own system BibSonomy, which allows for sharing both bookmarks andpublication references, and discuss first steps towards emergent semantics.} }

%0 = inproceedings %A = Hotho, Andreas and Jäschke, Robert and Schmitz, Christoph and Stumme, Gerd %B = Informatik 2006 -- Informatik für Menschen. Band 2 %C = Bonn %D = 2006 %I = Gesellschaft für Informatik %T = Emergent Semantics in BibSonomy %U = http://www.kde.cs.uni-kassel.de/stumme/papers/2006/hotho2006emergent.pdf