Home

Special Issue on Induction on the Semantic Web

Submission deadline: October 4, 2010


Overview

Increasingly, real-world data is published in the Semantic Web languages. The vast availability of these data has uncovered one of the main current limitations of deductive reasoning (generally adopted in the Semantic Web context), i.e., its severe limitation when scaling to large amounts of data. Alternative approaches such as data mining and machine learning methods could effectively cope with the web's scale and can also be used to capture new knowledge emerging from the data that is not logically derivable.

However, exploiting this global resource of data requires new perspective in perfoming data mining and machine learning that need to be able to deal with the heterogeneity and complexity of Semantic Web data. Depending on the data sources under consideration and the point of view of the individual researcher, the idiosyncrasies of the Semantic web -- e.g., the expressivity of the employed language, the richness of the ontologies novel assumptions (e.g., "open world") -- might play a major role in the analysis..

The primary goal of the special issue is to showcase cutting edge research on the intersection of the Semantic Web with Knowledge Discovery and Machine Learning, e.g.:

  • How can machine learning techniques, such as statistical learning methods and inductive forms of reasoning, work directly on the richly structured Semantic Web data and exploit the Semantic Web technologies?
  • How could machine learning techniques contribute to the full realization of the Semantic Web view?
  • What are the challenges for developers of machine learning techniques for the Semantic Web data?

Topics

The topics of interest of the special issue include, but are not limited to:

  • Knowledge Discovery and Ontologies:
    • data mining techniques using ontologies,
    • ontology mining and knowledge discovery from ontological knowledge bases,
    • ontology-based interpretation and validation of discovered knowledge,
    • whole knowledge discovery process guided by ontologies
  • Knowledge Discovery and Linked Data:
    • learning ontologies from Linked Data,
    • discovering hidden knowledge from Linked Data,
    • learning semantic relationship from Linked Data
  • Inductive Reasoning with Concept Languages:
    • inductive aggregation,
    • concept retrieval and query answering,
    • approximate classification,
    • inductive methods and fuzzy reasoning for ontology mapping,
    • construction, refinement and evolution of ontologies
    • concept change and novelty detection for ontology evolution
  • Statistical learning for the Semantic Web:
    • refinement operators for concept and rule languages,
    • concept and rules learning,
    • kernels and instance-based learning for structured representations,
    • semantic (dis-)similarity measures and conceptual clustering,
    • probabilistic methods for concept and rule languages
  • Other topics:
    • Open World Assumption (OWA) vs. Closed World Assumption (CWA) in learning,
    • applicability of relational learning in the Semantic Web context,
    • integration of induction and deduction,
    • evaluation methodologies and metrics for machine learning methods applied to ontologies
  • Applications:
    • challenges in practical applications of Machine Learning/Data Mining on the Semantic Web
    • life sciences,
    • cultural heritage,
    • semantic multimedia,
    • geo-informatics,
    • bio-informatics,
    • Semantic Web Services,
    • and others

    Submission Process

    Submissions to this special issue should follow the journal's guidelines for submission, and be made via the IJSWIS Submission System. After submitting a paper, please also inform the guest editors by email. Papers must be of high quality and should clearly state the technical issue(s) being addressed as related to Induction on the Semantic Web. Wherever possible, submissions should demonstrate the contribution of the research by reporting on a systematic evaluation of the work. If a submission is based on a prior publication in a workshop or conference, the journal submission must involve substantial advance (a minimum of 30%) in conceptual terms as well as in exposition (e.g., more comprehensive testing/evaluation/validation or additional applications/usage). If this applies to your submission, please explicitly reveal the relevant previous publications and describe enhancements to the previous version as an appendix so the reviewers have easy access to the details.

    The recommended length of submitted papers is between 5,500 to 8,000 words. All papers are subject to peer review performed by at least three established researchers drawn from a panel of experts selected for this special issue. Accepted papers will undergo for a second cycle of revision and reviewer feedback. Please submit manuscripts as a PDF file using the online submission system.

    The International Journal on Semantic Web and Information Systems (IJSWIS) is the first Semantic Web journal to be included in the Thomson ISI citation index. More information on the journal can be found at http://www.ijswis.org.

    Important Dates

    • Submission Deadline: October 4, 2010
    • Notifications: January 1, 2011
    • Revised Papers (after Revisions): February 15, 2011
    • Final Versions: March 15, 2011
    • Publication in a 2011 issue: 2011

    Organizing Committee

    Special Issue Guest Editors: