COALA - Correlation-Aware Active Learning of Link Specifications

Created by W.Langdon from gp-bibliography.bib Revision:1.4549

  author =       "Axel-Cyrille {Ngonga Ngomo} and Klaus Lyko and 
                 Victor Christen",
  title =        "{COALA} - Correlation-Aware Active Learning of Link
  booktitle =    "Proceedings of the 10th European Semantic Web
                 Conference: Semantics and Big Data, ESWC 2013",
  year =         "2013",
  volume =       "7882",
  series =       "LNCS",
  pages =        "442--456",
  address =      "Montpellier, France",
  month =        may # " 26-30",
  keywords =     "genetic algorithms, genetic programming, Active
                 Learning, Link Discovery",
  URL =          "",
  DOI =          "doi:10.1007/978-3-642-38288-8_30",
  timestamp =    "Tue, 26 Jun 2018 14:12:58 +0200",
  biburl =       "",
  bibsource =    "dblp computer science bibliography,",
  abstract =     "Link Discovery plays a central role in the creation of
                 knowledge bases that abide by the five Linked Data
                 principles. Over the last years, several active
                 learning approaches have been developed and used to
                 facilitate the supervised learning of link
                 specifications. Yet so far, these approaches have not
                 taken the correlation between unlabelled examples into
                 account when requiring labels from their user. In this
                 paper, we address exactly this drawback by presenting
                 the concept of the correlation-aware active learning of
                 link specifications. We then present two generic
                 approaches that implement this concept. The first
                 approach is based on graph clustering and can make use
                 of intra-class correlation. The second relies on the
                 activation-spreading paradigm and can make use of both
                 intra- and inter-class correlations. We evaluate the
                 accuracy of these approaches and compare them against a
                 state-of-the-art link specification learning approach
                 in ten different settings. Our results show that our
                 approaches outperform the state of the art by leading
                 to specifications with higher F-scores.",

Genetic Programming entries for Axel-Cyrille Ngonga Ngomo Klaus Lyko Victor Christen