Andreas Heß // Veröffentlichungen // 2010 :: 2009 :: 2008 :: 2007 :: 2006 :: 2005 :: 2004 :: 2003 :: 2001 English version

Veröffentlichungen

Wenn Sie eine der folgenden Veröffentlichungen zitieren möchten, so können Sie diese BibTeX Bibliographiedatei verwenden.

Mit Ausnahme meiner Diplomarbeit (2001) und dem Artikel im Handbuch Internet-Suchmaschinen (2009) sind alle Veröffentlichungen auf English.

2010

PDFCONTENTUS - towards semantic multi-media libraries

Jan Nandzik, Andreas Heß, Jan Hannemann, Nicolas Flores-Herr and Klaus Bossert

World Library and Information Congress: 76th IFLA General Conference and Assembly, Göteborg, Schweden, August 10-15, 2010.

Eintrag im Katalog der Deutschen Nationalbibliothek

Full Paper:Lade 'CONTENTUS - towards semantic multi-media libraries' als PDF von der IFLA Website  Presentation slides: Download Presentation Slides for 'CONTENTUS - towards semantic multi-media libraries' as PDF

The ever-growing amount of content and knowledge published online makes it possible for libraries to complement their own data and to present their collections in novel ways. Conceptually related information can be semantically linked so that users may benefit from richer data collections and novel search possibilities that capitalize on the inherent relationships between media, local metadata and external information sources. This paper presents potential solutions for the fundamental challenges of integrating heterogeneous data sources and providing innovative semantic search approaches, as they are developed for libraries and multimedia archives within the CONTENTUS project.

2009

PDFStealing Anchors to Link the Wiki

Philipp Dopichaj, Andre Skusa and Andreas Heß

Advances in Focused Retrieval, 7th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2008, Dagstuhl Castle, Germany, December 15-18, 2008. Revised and Selected Papers

© Springer Verlag, Lecture Notes in Computer Science

Full Paper: Lade 'Stealing Anchors to Link the Wiki' als PDF

PDFDer Markt für Internet-Suchmaschinen

(The market for internet search engines, in German)   Auf Deutsch / In German

Christian Maaß, Andre Skusa, Andreas Heß und Gotthard Pietsch

In: Handbuch Internet-Suchmaschinen, Dirk Lewandowski (Hrsg.), AKA Verlag Heidelberg

Eintrag im Katalog der Deutschen Nationalbibliothek

Artikel: Lade 'Der Markt für Internet-Suchmaschinen' als PDF

2008

PDFMulti-Value Classification of Very Short Texts

Andreas Heß, Philipp Dopichaj and Christian Maaß

31st Annual German Conference on Artificial Intelligence (KI 2008), Kaiserslautern, Germany

Springer Best Paper Award

© Springer Verlag, Lecture Notes in Computer Science

Full Paper: Download 'Multi-Value Classification of Very Short Texts' as PDF

We introduce a new stacking-like approach for multi-value classification. We apply this classification scheme using Naive Bayes and Rocchio classifiers on the well-known Reuters dataset. We use part-of-speech tagging for stopword removal. We show that our setup performs just as well as other approaches that use the full article text even though we only classify headlines. Finally, we apply a Rocchio classifier on a dataset from a Web 2.0-site and show that it is suitable for semi-automated labelling (often called tagging) of short texts and is faster than other approaches.

PDFPlayful Validation of Automatically Extracted Data

Francis Dierick, Philipp Dopichaj, Uwe Fleischer, Andreas Heß, Andre Skusa and Christian Maaß

Workshop Nutzerinteraktion im Social Semantic Web bei der Tagung Mensch & Computer, Lübeck, Germany

Full Paper: Download 'Playful Validation of Automatically Extracted Data' as PDF

PDFFrom Web 2.0 to Semantic Web: A Semi-Automated Approach

Andreas Heß, Christian Maaß and Francis Dierick

ESWC 2008 Workshop on Collective Semantics: Collective Intelligence and the Semantic Web (CISWeb 2008), Tenerife, Spain

Full Paper: Download 'From Web 2.0 to Semantic Web: A Semi-Automated Approach' as PDF  Presentation slides: Download Presentation Slides for 'From Web 2.0 to Semantic Web: A Semi-Automated Approach' as PDF

Web 2.0 and Semantic Web are regarded as two complementary paradigms that will probably converge in the future. However, whereas the Semantic Web is an established field of research, there has been little analysis devoted to Web 2.0 applications. For this reason it remains unclear how the advantages of both paradigms could be merged. In this paper we make three contributions in this direction. First, we discuss why merging Web 2.0 and the Semantic Web is beneficial and propose five approaches. Second, we show that (semi-) automated tagging of content improves the quality of annotations. Third, we present an automatic approach for improving the tag quality by using duplicate detection techniques. We verify our approach on a large-scale data set from the social search service Lycos iQ (now COSMiQ).

2007

PDFAlternative Searching Services: Seven Theses on the Importance of Social Bookmarking

Gernot Gräfe, Christian Maaß and Andreas Heß

Software, Agents and Services for Business, Research and E-Sciences — Conference on Social Semantic Web
(SABRE/CSSW 2007), Leipzig, Germany

Full Paper: Download 'Alternative Searching Services:
  Seven Theses on the Importance of Social Bookmarking' as PDF

In recent years social bookmark systems like del.icio.us or Furl have become increasingly popular. These systems sometimes are regarded as alternatives to algorithmic search engines like Google. In this paper we develop seven theses on the potential of these systems in order to establish a conceptual basis for future research in this area. Thereby it becomes clear that social bookmarking systems complement rather than threaten algorithmic search engines.

2006

PDFAn Iterative Algorithm for Ontology Mapping Capable of Using Training Data

Andreas Heß

3rd European Semantic Web Conference (ESWC 2006), Budva, Montenegro

© Springer Verlag, Lecture Notes in Computer Science

Full Paper: Download 'An Iterative Algorithm for Ontology Mapping Capable of Using Training Data' as PDF

We present a new iterative algorithm for ontology mapping where we combine standard string distance metrics with a structural similarity measure that is based on a vector representation. After all pairwise similarities between concepts have been calculated we apply well-known graph algorithms to obtain an optimal matching. Our algorithm is also capable of using existing mappings to a third ontology as training data to improve accuracy. We compare the performance of our algorithm with the performance of other alignment algorithms and show that our algorithm can compete well against the current state-of-the-art.

PDFSupervised and Unsupervised Ensemble Learning for the Semantic Web

Andreas Heß

Doktorarbeit, University College Dublin, School of Computer Science and Informatics

Eintrag im Katalog der Deutschen Nationalbibliothek

Doktorarbeit: Download 'Supervised and
Unsupervised Ensemble Learning for the Semantic Web' as PDF

Doktorvater: Nicholas Kushmerick

The World Wide Web offers an ocean of information and services for everyone, and many of us rely on it on a daily basis. One of the reasons for the Web's success is that the threshold for creating content on the Web was always very low, because comfortable and easy to use HTML editors are available. However, today's Web is limited by the fact that it is machine-readable, but not machine-understandable. The Semantic Web promises a solution to this problem by adding explicit semantics with the goal of making the Web machine-understandable by using description logics and ontologies. However, the threshold for creating this extra markup is very high, and no comfortable tools exist at present. The goal of this work is to develop such tools. Machine Learning techniques have been used in literature for various classification tasks. The central claim of this thesis is that such algorithms, supervised and unsupervised, can be used for this purpose.

PDF The Dublin Algorithm for Ontology Alignment

Andreas Heß

Kapitel in Stuckenschmidt et al.: Alignment implementation and benchmarking results, Seiten 7—10, Knowledge Web Network of Excellence, Deliverable D2.2.4

Deliverable: Download Deliverable D2.2.4 as PDF

2005

PDFEnsembles of Biased Classifiers

Rinat Khoussainov, Andreas Heß and Nicholas Kushmerick

The 22nd International Conference on Machine Learning (ICML 2005), Bonn, Germany

Full paper: Download 'Ensembles of Biased Classifiers' as PDF

We propose a novel ensemble learning algorithm called Triskel, which has two interesting features. First, Triskel learns an ensemble of classifiers, each biased to have high precision on instances from a single class (as opposed to, for example, boosting, where the ensemble members are biased to maximise accuracy over a subset of instances from all classes). Second, the ensemble members' voting weights are assigned so that certain pairs of biased classifiers outweigh the rest of the ensemble, if their predictions agree. Our experiments demonstrate that Triskel often outperforms boosting, in terms of both accuracy and training time. We also present an ROC analysis, which shows that Triskel's iterative structure corresponds to a sequence of nested ROC spaces. The analysis predicts that Triskel works best when there are concavities in the ROC curves; this prediction agrees with our empirical results.

Please see also the Triskel project pages for a reference implementation and more empirical results.

PDFEnsemble Learning with Biased Classifiers: The Triskel Algorithm

Andreas Heß, Rinat Khoussainov and Nicholas Kushmerick

6th International Workshop on Multiple Classifier Systems (MCS 2005), Monterey Bay, California, USA

© Springer Verlag, Lecture Notes in Computer Science

Full paper: Download 'Ensemble Learning with Biased Classifiers: The Triskel Algorithm' as PDF

Please see also the Triskel project pages for a reference implementation and more empirical results.

PDFMachine Learning Techniques for Annotating Semantic Web Services

Andreas Heß, Eddie Johnston and Nicholas Kushmerick

Dagstuhl Seminar on Machine Learning for the Semantic Web

Full paper: Download 'ASSAM: A Tool for
Semi-Automatically Annotating Semantic Web Services' as PDF

2004

PDFASSAM: A Tool for Semi-Automatically Annotating Semantic Web Services

Andreas Heß, Eddie Johnston and Nicholas Kushmerick

3rd International Semantic Web Conference (ISWC2004), Hiroshima, Japan

© Springer Verlag, Lecture Notes in Computer Science

Full paper: Download 'ASSAM: A Tool for
Semi-Automatically Annotating Semantic Web Services' as PDF   Presentation slides: Download Presentation Slides for 'ASSAM: A Tool for
Semi-Automatically Annotating Semantic Web Services' as PDF   RDF annotated abstract: View RDF annotated abstract for 'ASSAM: A Tool for
Semi-Automatically Annotating Semantic Web Services'
Demonstration Paper: Download Demo Paper 'ASSAM: A Tool for
Semi-Automatically Annotating Semantic Web Services' as PDF   Poster: Download Poster 'ASSAM: A Tool for
Semi-Automatically Annotating Semantic Web Services' as PDF   Flyer: Download Flyer 'ASSAM: A Tool for
Semi-Automatically Annotating Semantic Web Services' as PDF

The Semantic Web Services vision requires that each service be annotated with semantic metadata. Manually creating such metadata is tedious and error-prone, and many software engineers, accustomed to tools that automatically generate WSDL, might not want to invest the additional effort. We therefore propose ASSAM, a tool that assists a user in creating semantic metadata for Web Services. ASSAM is intended for service consumers who want to integrate a number of services and therefore must annotate them according to some shared ontology. ASSAM is also relevant for service producers who have deployed a Web Service and want to make it compatible with an existing ontology. ASSAM's capabilities to automatically create semantic metadata are supported by two machine learning algorithms. First, we have developed an iterative relational classification algorithm for semantically classifying Web Services, their operations, and input and output messages. Second, to aggregate the data returned by multiple semantically related Web Services, we have developed a schema mapping algorithm that is based on an ensemble of string distance metrics.

PDFIterative Ensemble Classification for Relational Data: A Case Study of Semantic Web Services

Andreas Heß and Nicholas Kushmerick

15th European Conference on Machine Learning (ECML2004), Pisa, Italy

© Springer Verlag, Lecture Notes in Computer Science

Full paper: Download 'Iterative Ensemble
Classification for Relational Data' as PDF  Presentation slides: Download Presentation Slides for 'Iterative Ensemble Classification for Relational Data' as PDF

For the classification of relational data, iterative algorithms that feed back predicted labels of associated objects have been used. In this paper we show two extensions to existing approaches. First, we propose to use two separate classifiers for the intrinsic and the relational (extrinsic) attributes and vote their predictions. Second, we introduce a new way of exploiting the relational structure. When the extrinsic attributes alone are not sufficient to make a prediction, we train specialised classifiers on the intrinsic features and use the extrinsic features as a selector. We apply these techniques to the task of semi-automated Web Service annotation, a task with a rich relational structure.

PDFSemi-Automatically Annotating Semantic Web Services

Andreas Heß, Eddie Johnston and Nicholas Kushmerick

Workshop on Information Integration on the Web (IIWeb2004),
held in conjunction with the 30th International Conference on Very Large Data Bases (VLDB2004), Toronto, Canada

Full paper: Download 'Semi-Automatically
Annotating Semantic Web Services' as PDF

PDFMachine Learning for Annotating Semantic Web Services

Andreas Heß and Nicholas Kushmerick

AAAI Spring Symposium Semantic Web Services 2004, Stanford, California, USA

Full paper: Download 'Machine Learning for
Annotating Semantic Web Services' as PDF Download 'Machine Learning for Annotating Semantic Web Services' as gzipped Postscript  Presentation slides: Download Presentation Slides 'Machine Learning for
Annotating Semantic Web Services' as PDF

2003

PDFLearning to Attach Semantic Metadata to Web Services

Andreas Heß and Nicholas Kushmerick

2nd International Semantic Web Conference (ISWC2003), Sanibel Island, Florida, USA

© Springer Verlag, Lecture Notes in Computer Science

Full paper: Download 'Learning to Attach Semantic Metadata to
Web Services' as PDF Download 'Learning to Attach Semantic Metadata to Web Services' as gzipped Postscript   Presentation slides: Download Presentation Slides for 'Learning to Attach Semantic Metadata to Web Services' as PDF   RDF annotated abstract: View RDF annotated abstract for 'Learning to Attach Semantic Metadata to Web Services'

Emerging Web standards promise a network of heterogeneous yet interoperable Web Services. Web Services would greatly simplify the development of many kinds of data integration and knowledge management applications. Unfortunately, this vision requires that services describe themselves with large amounts of semantic metadata "glue". We explore a variety of machine learning techniques to semi-automatically create such metadata.

We make three contributions. First, we describe a Bayesian learning and inference algorithm for classifying HTML forms into semantic categories, as well as assigning semantic labels to the form's fields. These techniques are important as legacy HTML interfaces are migrated to Web Services. Second, we describe the application of the Naive Bayes and SVM algorithms to the task of Web Service classification. We show that an ensemble approach that treats Web Services as structured objects is more accurate than an unstructured approach. Finally, we describe a clustering algorithm that automatically discovers the semantic categories of Web Services. All of our algorithms are evaluated using large collections of real HTML forms and Web Services.

PDFAutomatically Attaching Semantic Metadata to Web Services

Andreas Heß and Nicholas Kushmerick

Workshop on Information Integration on the Web (IIWeb2003)
at the 18th International Joint Conference on Artificial Intelligence (IJCAI2003), Acapulco, Mexico

Full paper: Download 'Automatically Attaching Semantic Metadata to Web Services' as PDF Download 'Automatically Attaching Semantic Metadata to Web
Services' as gzipped Postscript

2001

gzipped PostscriptThemenextraktion aus Semantischen Netzen

(Topic extraction from semantic networks, in German)   Auf Deutsch / In German

Andreas Heß

Diplomarbeit, Fachhochschule Darmstadt, Fachbereich Informatik 

Eintrag im Katalog der Deutschen Nationalbibliothek

Diplomarbeit: Lade 'Themenextraktion aus Semantischen Netzen' als gzip-komprimiertes Postscript Lade 'Themenextraktion aus Semantischen Netzen' als PDF


Andreas Hess, mail at andreas-hess dot info