Zero-shot learning aims to classify visual objects without any training data via knowledge transfer between seen and unseen classes. This is typically achieved by exploring a semantic embedding space where the seen and unseen classes can be related. Previous works differ in what embedding space is used and how different classes and a test image can be related. In this paper, we utilize the anno...

Spoken dialogue systems (SDS) typically require a predefined semantic ontology to train a spoken language understanding (SLU) module. In addition to the annotation cost, a key challenge for designing such an ontology is to define a coherent slot set while considering their complex relations. This paper introduces a novel matrix factorization (MF) approach to learn latent feature vectors for utt...

BioUML is integrated extensible workbench for systems biology (http://www.biouml.org). Its core is BioUML meta model that provides a formalism for comprehensive description, visualization and simulation of wide range of biological and other complex systems. The suggested model is problem domain neutral and splits the system description into three interconnected levels: 1) graph structure the sy...

Information visualisation benefits from the Semantic Web: multimodal mobile interfaces to the Semantic Web offer access to complex knowledge and information structures. Natural language dialogue systems are ideal interfaces to personal digital assistants (PDAs) or other handheld clients. We explore more fine-grained co-ordination of multimodal presentations as answers to natural language questi...

In this paper, we present a novel lexical knowledge graph called SRDF and describe an extraction system that automatically generates a SRDF graph from the Korean natural language sentence. In the semantic web, knowledge is expressed in the RDF triple form but natural language sentences consist of multiple relationships between the predicates and arguments. For this reason, we design a SRDF grap...

Bootstrapping has a tendency, called semantic drift, to select instances unrelated to the seed instances as the iteration proceeds. We demonstrate the semantic drift of bootstrapping has the same root as the topic drift of Kleinberg’s HITS, using a simplified graphbased reformulation of bootstrapping. We confirm that two graph-based algorithms, the von Neumann kernels and the regularized Laplac...

Several works have been dedicated to the authoring and presentation of interactive multimedia applications to be distributed over the web, but few of them address semantic verification issues of these documents. This paper presents a formal approach for the design and semantic verification of SMIL documents which relies on the classical reachability analysis performed on RT-LOTOS specifications...

This paper introduces Explicit Semantic Ranking (ESR), a new ranking technique that leverages knowledge graph embedding. Analysis of the query log from our academic search engine, SemanticScholar.org, reveals that a major error source is its inability to understand the meaning of research concepts in queries. To addresses this challenge, ESR represents queries and documents in the entity space ...

We propose a novel semantic parsing framework for question answering using a knowledge base. We define a query graph that resembles subgraphs of the knowledge base and can be directly mapped to a logical form. Semantic parsing is reduced to query graph generation, formulated as a staged search problem. Unlike traditional approaches, our method leverages the knowledge base in an early stage to p...

In this paper we discuss a kernel approach to learning in semantic graphs. To scale up the performance to large data sets, we employ the Nyström approximation. We derive a kernel derived from semantic relations in a local neighborhood of a node. One can apply our approach to problems in multi-relational domains with several thousand graph nodes and more than a million potential links. We apply ...