@inproceedings{bade2008evaluation,
  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.},
  address = {Berlin-Heidelberg},
  author = {Bade, Korinna and Benz, Dominik},
  booktitle = {Proceedings of the 32nd Annual Conference of the German Classification Society - Advances in Data Analysis, Data Handling and Business Intelligence (GfKl 2008)},
  file = {bade2008evaluation.pdf:bade2008evaluation.pdf:PDF},
  groups = {public},
  interhash = {8bb09e3197d01f7c23481c2cd68533af},
  intrahash = {ec033805bc90ab87c99860e29f0d00dd},
  note = {in press},
  publisher = {Springer},
  series = {Studies in Classification, Data Analysis, and Knowledge Organization},
  title = {Evaluation Strategies for Learning Algorithms of Hierarchical Structures},
  url = {http://www.kde.cs.uni-kassel.de/pub/pdf/bade2008evaluation.pdf},
  username = {dbenz},
  year = 2008
}

@inproceedings{benz2008analyzing,
  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.},
  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},
  booktitle = {Proceedings of the Dagstuhl Seminar on Social Web Communities},
  editor = {Alani, Harith and Staab, Steffen and Stumme, Gerd},
  file = {benz2008analyzing.pdf:benz2008analyzing.pdf:PDF},
  groups = {public},
  interhash = {d738d9d90c1c466ee0a73ac0cc3dc4c1},
  intrahash = {6918e578527dec96abb5718f105d9f78},
  issn = {1862-4405},
  number = 08391,
  title = {Analyzing Tag Semantics Across Collaborative Tagging Systems},
  url = {http://www.kde.cs.uni-kassel.de/pub/pdf/benz2008analyzing.pdf},
  username = {dbenz},
  year = 2008
}

@inproceedings{benz2009characterizing,
  address = {Bled, Slovenia},
  author = {Benz, Dominik and Krause, Beate and Kumar, G. Praveen and Hotho, Andreas and Stumme, Gerd},
  booktitle = {Proceedings of the 1st Workshop on Explorative Analytics of Information Networks (EIN2009)},
  file = {benz2009characterizing.pdf:benz2009characterizing.pdf:PDF},
  groups = {public},
  interhash = {de5e58b26200e44112d9791f39e7523d},
  intrahash = {b697a98a7340585594455ee2e81d238a},
  month = {September},
  title = {Characterizing Semantic Relatedness of Search Query Terms},
  url = {http://www.kde.cs.uni-kassel.de/pub/pdf/benz2009characterizing.pdf},
  username = {dbenz},
  year = 2009
}

@inproceedings{bullinaria2008semantic,
  author = {Bullinaria, J.A.},
  file = {bullinaria2008semantic.pdf:bullinaria2008semantic.pdf:PDF},
  groups = {public},
  interhash = {cdb7b1ff0e89f61f84e2c15a0e46c221},
  intrahash = {efae206c0f89363a3273a8d57c87eff5},
  journal = {ESSLLI Workshop on Distributional Lexical Semantics},
  timestamp = {2011-01-28 09:53:43},
  title = {Semantic Categorization Using Simple Word Co-occurrence statistics},
  username = {dbenz},
  year = 2008
}

@inproceedings{christiaens2006metadata,
  abstract = {In this paper we give a brief overview of different metadata mechanisms (like ontologies and folksonomies) and how they relate to each other. We identify major strengths and weaknesses of these mechanisms. We claim that these mechanisms can be classified from restricted (e.g., ontology) to free (e.g., free text tagging). In our view, these mechanisms should not be used in isolation, but rather as complementary solutions, in a continuous process wherein the strong points of one increase the semantic depth of the other. We give an overview of early active research already going on in this direction and propose that methodologies to support this process be developed. We demonstrate a possible approach, in which we mix tagging, taxonomy and ontology.},
  author = {Christiaens, Stijn},
  booktitle = {Lecture Notes in Computer Science: On the Move to Meaningful Internet Systems 2006: OTM 2006 Workshops},
  file = {christiaens2006metadata.pdf:christiaens2006metadata.pdf:PDF},
  groups = {public},
  interhash = {f733d993459329ed1ef9f26d303ba0d9},
  intrahash = {efc1396e845f3db1688dc8ef154d9520},
  lastdatemodified = {2007-01-04},
  lastname = {Christiaens},
  own = {notown},
  pdf = {christiaens06-metadata.pdf},
  publisher = {Springer},
  read = {notread},
  timestamp = {2007-09-11 13:31:23},
  title = {Metadata Mechanisms: From Ontology to Folksonomy ... and Back},
  url = {http://www.springerlink.com/content/m370107220473394},
  username = {dbenz},
  workshoppub = {1},
  year = 2006
}

@misc{asur2010predicting,
  abstract = {In recent years, social media has become ubiquitous and important for socialnetworking and content sharing. And yet, the content that is generated fromthese websites remains largely untapped. In this paper, we demonstrate howsocial media content can be used to predict real-world outcomes. In particular,we use the chatter from Twitter.com to forecast box-office revenues for movies.We show that a simple model built from the rate at which tweets are createdabout particular topics can outperform market-based predictors. We furtherdemonstrate how sentiments extracted from Twitter can be further utilized toimprove the forecasting power of social media.},
  author = {Asur, Sitaram and Huberman, Bernardo A.},
  file = {asur2010predicting.pdf:asur2010predicting.pdf:PDF},
  groups = {public},
  interhash = {538607d6d5da7946a0c5a2114a7c44f5},
  intrahash = {9c23c0465529a60d9540ee29e74856f1},
  note = {cite arxiv:1003.5699},
  timestamp = {2010-11-09 10:12:57},
  title = {Predicting the Future with Social Media},
  url = {http://arxiv.org/abs/1003.5699},
  username = {dbenz},
  year = 2010
}

@inproceedings{heymann2010tagging,
  abstract = {A fundamental premise of tagging systems is that regular users can organize large collections for browsing and other tasks using uncontrolled vocabularies. Until now, that premise has remained relatively unexamined. Using library data, we test the tagging approach to organizing a collection. We find that tagging systems have three major large scale organizational features: consistency, quality, and completeness. In addition to testing these features, we present results suggesting that users produce tags similar to the topics designed by experts, that paid tagging can effectively supplement tags in a tagging system, and that information integration may be possible across tagging systems.},
  author = {Heymann, Paul and Paepcke, Andreas and Garcia-Molina, Hector},
  booktitle = {WSDM},
  crossref = {conf/wsdm/2010},
  date = {2010-02-18},
  editor = {Davison, Brian D. and Suel, Torsten and Craswell, Nick and Liu, Bing},
  ee = {http://doi.acm.org/10.1145/1718487.1718495},
  file = {:heyman2010tagging.pdf:PDF},
  groups = {public},
  interhash = {d4f72ed57e6b99dbe32e18e218d81ef5},
  intrahash = {12579231cd5449f9a40cba9924975f09},
  isbn = {978-1-60558-889-6},
  pages = {51-60},
  publisher = {ACM},
  timestamp = {2010-04-08 07:27:02},
  title = {Tagging human knowledge.},
  url = {http://dblp.uni-trier.de/db/conf/wsdm/wsdm2010.html#HeymannPG10},
  username = {dbenz},
  year = 2010
}

@inproceedings{takano1998dynamic,
  abstract = {This paper describes a management tool to support revisiting WWW pages, which we call “WWW Dynamic Bookmark (WDB).�? WDB watches and archives a user’s navigation behavior, analyses the archive, and shows analyzed results as clues for revisiting URLs. We have integrated link analysis and user behavior analysis to evaluate WWW page importance. WDB presents a list of sites that a user has visited, in importance order, via a landmark list in each site, and showing relationships among sites. Experimental implementation shows that importance calculation and structure displays help users to pick up useful URLs.},
  address = {New York, NY, USA},
  author = {Takano, Hajime and Winograd, Terry},
  booktitle = {HYPERTEXT '98: Proceedings of the ninth ACM conference on Hypertext and hypermedia : links, objects, time and space---structure in hypermedia systems},
  file = {takano1998dynamic.pdf:takano1998dynamic.pdf:PDF},
  interhash = {ee36d0006bdf2300dff4b6b21aa2f231},
  intrahash = {f909bc2f43cc4a7e86d5b675944e24fd},
  lastdatemodified = {2005-08-06},
  lastname = {Takano},
  own = {own},
  pages = {297--298},
  pdf = {takano98.pdf},
  publisher = {ACM Press},
  read = {notread},
  title = {Dynamic bookmarks for the WWW},
  url = {doi.acm.org/10.1145/276627.276667},
  year = 1998
}

@book{salton1989automatic,
  address = {Boston, MA, USA},
  author = {Salton, Gerard},
  interhash = {2b94348e77f1c272f794db647d599892},
  intrahash = {b7c429f46ea331fc3a10d8124b1cb8ad},
  lastdatemodified = {2005-11-16},
  lastname = {Salton},
  own = {notown},
  publisher = {Addison-Wesley Longman Publishing Co., Inc.},
  read = {notread},
  title = {Automatic text processing: the transformation, analysis, and retrieval of information by computer},
  year = 1989
}

@misc{mukherjee2004semantic,
  abstract = {Bookmarks are shortcuts that enable quick access of the desired Web content. They have become a standard feature in any browser and recent studies have shown that they can be very useful for non-visual Web access as well. Current bookmarking techniques in assistive Web browsers are rigidly tied to the structure of Web pages. Consequently they are susceptible to even slight changes in the structure of Web pages. In this paper we propose semantic bookmarking for non-visual Web access. With the help of an ontology that represents concepts in a domain, content in Web pages can be semantically associated with bookmarks. As long as these associations can be identified, semantic bookmarks are resilient in the face of structural changes to the Web page. The use of ontologies allows semantic bookmarks to span multiple Web sites covered by a common domain. This contributes to the ease of information retrieval and bookmark maintenance. In this paper we describe highly automated techniques for creating and retrieving semantic bookmarks. These techniques have been incorporated into an assistive Web browser. Preliminary experimental evidence suggests the effectiveness of semantic bookmarks for non-visual Web access.},
  author = {Mukherjee, S. and Ramakrishnan, I. and Kifer, M.},
  file = {mukherjee2004semantic.pdf:mukherjee2004semantic.pdf:PDF},
  interhash = {d09f86aa5f36ea8eccd009c1be2bef2f},
  intrahash = {0b103520e1f873a93f66ef8115c1bffb},
  lastdatemodified = {2005-08-07},
  lastname = {Mukherjee},
  own = {own},
  pdf = {mukherjee04.pdf},
  read = {notread},
  title = {Semantic bookmarking for non-visual web access},
  url = {mukherjee04.ps},
  year = 2004
}

@misc{lacher2001facilitating,
  abstract = {In this paper, we give an overview of a system (CAIMAN) that can facilitate the exchange of relevant documents between geographically dispersed people in Communities of Interest. The nature of Communities of Interest prevents the creation and enforcement of a common organizational scheme for documents, to which all community members adhere. Each community member organizes her documents according to her own categorization scheme (ontology). CAIMAN exploits this personal ontology, which is essentially the perspective of a user on a domain, for information retrieval. Related documents are retrieved on a concept granularity level from a central community document repository. To find the related concepts in the queried ontology, CAIMAN performs an ontology mapping. The ontology mapping in CAIMAN is based on a novel approach, which considers the concepts in an ontology implicitly represented by the documents assigned to each concept. Using machine learning techniques for te...},
  author = {Lacher, M. and Groh, G.},
  file = {lacher2001facilitating.pdf:lacher2001facilitating.pdf:PDF},
  interhash = {c1b25b9d7f6cc1d0fd155811fe698839},
  intrahash = {141fb64c315972315dfab1d77451c117},
  lastdatemodified = {2005-08-07},
  lastname = {Lacher},
  own = {own},
  pdf = {lacher01.pdf},
  read = {notread},
  title = {Facilitating the exchange of explicit knowledge through ontology mappings},
  url = {lacher01.ps},
  year = 2001
}

@inproceedings{kaasten2001integrating,
  abstract = {Most Web browsers include Back, History and Bookmark facilities that simplify how people return to previously seen pages. While useful, these three facilities all operate on quite different underlying models, which undermines their usability. Our alternative revisitation system uses a single model of a recency-ordered history list to integrate Back, History and Bookmarks. Enhancements include: Back as a way to step through this list; implicit and explicit 'dog-ears' to mark pages on the list (replacing Bookmarks); searching/filtering the list through dynamic queries; and visual thumbnails to promote page recognition.},
  address = {New York, NY, USA},
  author = {Kaasten, Shaun and Greenberg, Saul},
  booktitle = {CHI '01: CHI '01 extended abstracts on Human factors in computing systems},
  interhash = {6f78c3258715fe82008530454538e7e6},
  intrahash = {f99a12a83dfaedd7c1f998a4c428e26d},
  lastdatemodified = {2005-08-06},
  lastname = {Kaasten},
  own = {own},
  pages = {379--380},
  pdf = {kaasten01.pdf},
  publisher = {ACM Press},
  read = {notread},
  title = {Integrating back, history and bookmarks in web browsers},
  url = {doi.acm.org/10.1145/634067.634291},
  year = 2001
}

@article{denoue2000annotation,
  abstract = {With bookmark programs, current Web browsers provide a limited support to personalize the Web. We present a new Web annotation tool which uses the Document Object Model Level 2 and Dynamic HTML to deliver a system where speed and privacy are important issues. We report on several experiments showing how annotations improve document access and retrieval by providing user-directed document summaries. Preliminary results also show that annotations can be used to produce user-directed document clustering and classification.},
  author = {Denoue, L. and Vignollet, L.},
  dateadded = {2005-08-06},
  file = {denoue2000annotation.pdf:denoue2000annotation.pdf:PDF},
  interhash = {25cb506db1cce228ab5e750e025f2f03},
  intrahash = {8fe69cc8f204724adb3eb5f990b79437},
  journal = {Proceedings of RIAO2000},
  lastdatemodified = {2005-08-06},
  lastname = {Denoue},
  month = {April},
  own = {own},
  pdf = {denoue00.pdf},
  read = {notread},
  title = {An annotation tool for Web browsers and its applications to information retrieval},
  url = {http://. univ-savoie.fr/labos/syscom/Laurent.Denoue/riao2000.doc},
  year = 2000
}

@article{cattuto2006semiotic,
  abstract = {Abstract A distributed classification paradigm known as collaborative tagging has been successfully deployed in large-scale web applications designed to manage and share diverse online resources. Users of these applications organize resources by associating with them freely chosen text labels, or tags. Here we regard tags as basic dynamical entities and study the semiotic dynamics underlying collaborative tagging. We collect data from a popular system and focus on tags associated with a given resource. We find that the frequencies of tags obey to a generalized Zipf�s law and show that a Yule�Simon process with memory can be used to explain the observed frequency distributions in terms of a simple model of user behavior},
  author = {Cattuto, Ciro},
  file = {cattuto2006semiotic.pdf:cattuto2006semiotic.pdf:PDF},
  groups = {public},
  interhash = {6651fe8b8916e8407f738325c092b860},
  intrahash = {df2a1161a75e3f328f82c204f942bb8a},
  journal = {The European Physical Journal C - Particles and Fields},
  journalpub = {1},
  lastdatemodified = {2006-09-25},
  lastname = {Cattuto},
  longnotes = {doi:10.1140/epjcd/s2006-03-004-4},
  month = {August},
  own = {notown},
  pages = {33--37},
  pdf = {cattuto2006-semiotic.pdf},
  read = {notread},
  timestamp = {2007-09-11 13:31:22},
  title = {Semiotic dynamics in online social communities},
  url = {http://www.springerlink.com/content/t964j63030507341},
  username = {dbenz},
  volume = 46,
  year = 2006
}

@article{cattuto2007semiotic,
  abstract = {Collaborative tagging has been quickly gaining ground because of its ability to recruit the activity of web users into effectively organizing and sharing vast amounts of information. Here we collect data from a popular system and investigate the statistical properties of tag co-occurrence. We introduce a stochastic model of user behavior embodying two main aspects of collaborative tagging: (i) a frequency-bias mechanism related to the idea that users are exposed to each other's tagging activity; (ii) a notion of memory - or aging of resources - in the form of a heavy-tailed access to the past state of the system. Remarkably, our simple modeling is able to account quantitatively for the observed experimental features, with a surprisingly high accuracy. This points in the direction of a universal behavior of users, who - despite the complexity of their own cognitive processes and the uncoordinated and selfish nature of their tagging activity - appear to follow simple activity patterns.},
  author = {Cattuto, Ciro and Loreto, Vittorio and Pietronero, Luciano},
  file = {cattuto2007semiotic.pdf:cattuto2007semiotic.pdf:PDF},
  groups = {public},
  interhash = {189402152f540f931a0eea5b8538411f},
  intrahash = {95a8e6a348e0acde9ce781004c45b94e},
  journal = {Proceedings of the National Academy of Sciences United States of America},
  journalpub = {1},
  lastdatemodified = {2007-05-14},
  lastname = {Cattuto},
  own = {notown},
  pages = 1461,
  pdf = {cattuto06-collaborative.pdf},
  read = {notread},
  timestamp = {2007-09-11 13:31:22},
  title = {Semiotic Dynamics and Collaborative Tagging},
  url = {http://www.citebase.org/abstract?id=oai:arXiv.org:cs/0605015},
  username = {dbenz},
  volume = 104,
  year = 2007
}

@inproceedings{haase2005usagedriven,
  abstract = {Large information repositories as digital libraries, online shops, etc. rely on a taxonomy of the objects under consideration to structure the vast contents and facilitate browsing and searching (e.g., ACM topic classification for computer science literature, Amazon product taxonomy, etc.). As in heterogeneous communities users typically will use different parts of such an ontology with varying intensity, customization and personalization of the ontologies is desirable. In this paper we adapt a collaborative filtering recommender system to assist users in the management and evolution of their personal ontology by providing detailed suggestions of ontology changes. Such a system has been implemented in the context of Bibster, a peer-to-peer based personal bibliography management tool. Finally, we report on an in-situ experiment with the Bibster community that shows the performance improvements over non-personalized recommendations.},
  address = {Las Vegas, Nevada USA},
  author = {Haase, Peter and Hotho, Andreas and Schmidt-Thieme, Lars and Sure, York},
  booktitle = {Proceedings of the 3rd International Conference on Universal A ccess in Human-Computer Interaction (UAHCI)},
  file = {haase2005usagedriven.pdf:haase2005usagedriven.pdf:PDF},
  groups = {public},
  interhash = {e5681c379cdfe126e44d034dac3fddad},
  intrahash = {05b1daa45aedb5c1f63683f961f17a9e},
  lastdatemodified = {2006-07-06},
  lastname = {Haase},
  month = {22-27 July},
  own = {notown},
  pdf = {Haase05.pdf},
  read = {notread},
  timestamp = {2007-05-25 16:05:53},
  title = {Usage-driven Evolution of Personal Ontologies},
  username = {dbenz},
  year = 2005
}

@inproceedings{jung2004collaborative,
  abstract = {This paper proposes the collaborative web browsing system sharing knowledge with other users. We have specifically focused on user interests extracted from bookmarks. A simple URL based-bookmark is provided with structural information by the conceptualization of the ontology. Furthermore, ontology learning based on a hierarchical clustering method can be applied to handle dynamic changes in bookmarks. As a result of our experiments, with respect to recall, about 53.1% of the total time was saved during collaborative browsing for seeking the equivalent set of information, as compared with single web browsing.},
  author = {Jung, Jason J. and Yu, Young-Hoon and Jo, GeunSik},
  booktitle = {International Conference on Computational Science},
  file = {jung2004collaborative.pdf:jung2004collaborative.pdf:PDF},
  groups = {public},
  interhash = {e1b4ebe8a3ae831c9372c7ca4f042256},
  intrahash = {ff3838b29b4d15654173fc724dcd383a},
  lastdatemodified = {2005-08-06},
  lastname = {Jung},
  note = {bibsource: DBLP, http://dblp.uni-trier.de},
  own = {own},
  pages = {513-520},
  pdf = {jung04.pdf},
  read = {read},
  timestamp = {2007-05-25 16:05:53},
  title = {Collaborative Web Browsing Based on Ontology Learning from Bookmarks.},
  url = {springerlink.metapress.com/openurl.asp?genre=article{\&}issn=0302-9743{\&}volume=3038{\&}spage=513},
  username = {dbenz},
  year = 2004
}

@misc{leydesdorff2010semantic,
  abstract = { Meaning can be generated when information is related at a systemic level.Such a system can be an observer, but also a discourse, for example,operationalized as a set of documents. The measurement of semantics assimilarity in patterns (correlations) and latent variables (factor analysis)has been enhanced by computer techniques and the use of statistics; forexample, in "Latent Semantic Analysis". This communication provides anintroduction, an example, pointers to relevant software, and summarizes thechoices that can be made by the analyst. Visualization ("semantic mapping") isthus made more accessible.},
  author = {Leydesdorff, Loet and Welbers, Kasper},
  file = {leydesdorff2010semantic.pdf:leydesdorff2010semantic.pdf:PDF},
  groups = {public},
  interhash = {ae82f513935ffa158395683303d52517},
  intrahash = {cee25eef0438b6ac2998041dbd93016e},
  note = {cite arxiv:1011.5209},
  timestamp = {2010-11-24 10:21:03},
  title = {The semantic mapping of words and co-words in contexts},
  url = {http://arxiv.org/abs/1011.5209},
  username = {dbenz},
  year = 2010
}

@inproceedings{marlow2006position,
  abstract = {In recent years, tagging systems have become increasingly popular. These systems enable users to add keywords (i.e., �tags�) to Internet resources (e.g., web pages, images, videos) without relying on a controlled vocabulary. Tagging systems have the potential to improve search, spam detection, reputation systems, and personal organization while introducing new modalities of social communication and opportunities for data mining. This potential is largely due to the social structure that underlies many of the current systems. Despite the rapid expansion of applications that support tagging of resources, tagging systems are still not well studied or understood. In this paper, we provide a short description of the academic related work to date. We offer a model of tagging systems, specifically in the context of web-based systems, to help us illustrate the possible benefits of these tools. Since many such systems already exist, we provide a taxonomy of tagging systems to help inform their analysis and design, and thus enable researchers to frame and compare evidence for the sustainability of such systems. We also provide a simple taxonomy of incentives and contribution models to inform potential evaluative frameworks. While this work does not present comprehensive empirical results, we present a preliminary study of the photosharing and tagging system Flickr to demonstrate our model and explore some of the issues in one sample system. This analysis helps us outline and motivate possible future directions of research in tagging systems.},
  address = {Edinburgh, Scotland},
  author = {Marlow, Cameron and Naaman, Mor and Boyd, Danah and Davis, Marc},
  booktitle = {Proceedings of the Collaborative Web Tagging Workshop at the WWW 2006},
  file = {marlow2006position.pdf:marlow2006position.pdf:PDF},
  groups = {public},
  interhash = {7446351e0d902ee4f36fb750f82c50a5},
  intrahash = {d9f433de0945351fa2157c1424d9fe67},
  lastdatemodified = {2006-07-17},
  lastname = {Marlow},
  month = May,
  own = {own},
  pdf = {marlow06-tagging.pdf},
  read = {readnext},
  timestamp = {2007-09-11 13:31:31},
  title = {{Position Paper, Tagging, Taxonomy, Flickr, Article, ToRead}},
  url = {http://.rawsugar.com/www2006/cfp.html},
  username = {dbenz},
  year = 2006
}

@inproceedings{ramage2009clustering,
  abstract = {Automatically clustering web pages into semantic groups promises improved search and browsing on the web. In this paper, we demonstrate how user-generated tags from largescale social bookmarking websites such as del.icio.us can be used as a complementary data source to page text and anchor text for improving automatic clustering of web pages. This paper explores the use of tags in 1) K-means clustering in an extended vector space model that includes tags as well as page text and 2) a novel generative clustering algorithm based on latent Dirichlet allocation that jointly models text and tags. We evaluate the models by comparing their output to an established web directory. We find that the naive inclusion of tagging data improves cluster quality versus page text alone, but a more principled inclusion can substantially improve the quality of all models with a statistically significant absolute F-score increase of 4%. The generative model outperforms K-means with another 8% F-score increase.},
  address = {New York, NY, USA},
  author = {Ramage, Daniel and Heymann, Paul and Manning, Christopher D. and Garcia-Molina, Hector},
  booktitle = {WSDM '09: Proceedings of the Second ACM International Conference on Web Search and Data Mining},
  doi = {http://doi.acm.org/10.1145/1498759.1498809},
  file = {ramage2009clustering.pdf:ramage2009clustering.pdf:PDF},
  groups = {public},
  interhash = {5595f06f88310ed67fd6fe23f813c69b},
  intrahash = {75c4bad29d7eb4b34f68da27f0353516},
  isbn = {978-1-60558-390-7},
  location = {Barcelona, Spain},
  pages = {54--63},
  publisher = {ACM},
  timestamp = {2009-04-24 10:19:45},
  title = {Clustering the tagged web},
  url = {http://portal.acm.org/citation.cfm?id=1498809},
  username = {dbenz},
  year = 2009
}