The idea of the Semantic Web provides new opportunities for
semantically-enabled user interfaces. The explicit representation of the
meaning of data allows us to (1) transcend traditional keyboard and
mouse interaction metaphors, and (2) provide representation structures
for more complex interaction scenarios that aim to address semantic
In this talk I will give an overview of DFKI's Semantic Dialogue Shell
developed in the context of the THESEUS project. THESEUS is the German
flagship project on the Internet of Services, where the user can
delegate complex tasks to dynamically composed semantic web services by
utilising multimodal interaction.
Over the last years, we have adhered strictly to the developed rule "No
presentation without representation." The idea is to implement a
generic, and semantic, dialogue shell that can be configured for and
applied to domain-specific dialogue applications. The dialogue system
also acts as the middleware between the clients and the backend
services. The spoken dialogue input is used to generate SPARQL queries
to address information sources such as DBpedia, Yago, or YouTube. The
backend system also integrates multiple Linked Data sources.
We will describe our semantic search architecture in two specific
use-case scenarios: In the first scenario, a user is interested in music
and is able to retrieve musician's information using natural speech. In
the second use case scenario, we examine first ideas on the
applicability of Linked Data, in particular a subset of the Linked Open
Drug Data (LODD), to connect radiology, human anatomy, and drug
information for improved medical image annotation and subsequent search.
We will explain this with an example in the THESEUS use case Medico.
Whereas the dialogue-based access to keyword-based search engines has
only moderate success, semantic (ontology-based) interpretations of
dialogue utterances may become the key advancement in semantic search,
thereby mediating and addressing dynamic semantic search engines which
are already freely available.