• Publications
  • Influence
Boosting for Text Classification with Semantic Features
TLDR
This paper proposes an enhancement of the classical document representation through concepts extracted from background knowledge through Boosting, a successful machine learning technique used for classification. Expand
Kernel Methods for Mining Instance Data in Ontologies
TLDR
This work investigates how machine learning algorithms can be made amenable for directly taking advantage of the rich knowledge expressed in ontologies and associated instance data through decomposing the kernel computation into specialized kernels for selected characteristics of an ontology which can be flexibly assembled and tuned. Expand
The SWRC Ontology - Semantic Web for Research Communities
TLDR
This paper describes the publicly available ‘Semantic Web for Research Communities’ (SWRC) ontology, in which research communities and relevant related concepts are modelled, and describes the design decisions that underlie the ontology. Expand
Semantic Kernels for Text Classification Based on Topological Measures of Feature Similarity
TLDR
A new approach to the design of semantic smoothing kernels for text classification that implicitly encode a superconcept expansion in a semantic network using well-known measures of term similarity. Expand
Graph Kernels for RDF Data
TLDR
This paper introduces two versatile families of graph kernels specifically suited for RDF, based on intersection graphs and intersection trees, and shows that the novel RDF graph kernels used with (SVMs) achieve competitive predictive performance when compared to specialized techniques for both tasks. Expand
TagFS - Tag Semantics for Hierarchical File Systems
TLDR
This paper analyzes the different semantics between strictly hierarchical and tagging-based organisation and maps non-hierarchical tagging and query semantics to the commonly used hierarchical file system semantics, thus combining the benefits of both worlds. Expand
Semantic Annotation of Images and Videos for Multimedia Analysis
TLDR
This paper uses M-OntoMat-Annotizer in order to construct ontologies that include prototypical instances of high-level domain concepts together with a formal specification of corresponding visual descriptors, allowing for new kinds of multimedia content analysis and reasoning. Expand
Combined Syntactic and Semantic Kernels for Text Classification
TLDR
A new type of kernel is proposed, the Semantic Syntactic Tree Kernel (SSTK), which incorporates linguistic structures, e.g. syntactic dependencies and semantic background knowledge, to automatically learn question categories in QA. Expand
An Approach to Formal and Semantic Representation of Logistics Services
Modern logistics systems are characterized by an increasing structural complexity and dynamicity, which arises from trends in the global economy, decomposition of supply chains and individualExpand
Structure and semantics for expressive text kernels
TLDR
This work proposes a generalized framework consisting of a family of kernels that jointly incorporate syntactic and semantic similarity and demonstrates the power of this approach in a series of experiments. Expand
...
1
2
3
4
5
...