Semantic Web Seminars at LiU - Spring 2017
Tuesday, January 24, 3.15 pm, 2017.Recent Developments in Bridging the Semantic Gap Problem
Marjan Alirezaie (Örebro University)
Abstract: In this talk, I will present a summary of my PhD thesis which is about bridging the semantic gap between sensor data and ontological knowledge. The focus will be on the recent developments in heterogeneous knowledge integration as a solution for the semantic gap issue. I will also introduce the two research projects: Semantic Robot and E-care@home, with the goal of addressing the knowledge integration problem. In Semantic Robot, we are aiming to provide a semantic layer to a 3D topographic map and making the objects in the 3D map reasoning ready for different purposes such as querying and navigation. Likewise, in Ecare@home which is a Swedish interdisciplinary distributed research environment, our focus is on the development of methods that provide interpretation of the heterogeneous data coming from different types of sensors in conjunction with both medical and environmental knowledge in order to provide e-services for the elderly residing in their homes.
Location: Room "Alan Turing"
Organizer: Eva Blomqvist
Tuesday, February 7, 3.15 pm, 2017.Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes
Kim Ahlstrøm Meyn Mathiassen (Aalborg University)
Abstract: The SPARQL 1.1 standard has made it possible to formulate analytical queries in SPARQL. While some approaches have become available for processing analytical queries on RDF data cubes, little attention has been paid to answering provenance-enabled queries over such data. Yet, considering provenance is a prerequisite to being able to validate if a query result is trustworthy. The main challenge for existing triple stores is the way provenance can be encoded in standard triple stores based on context values (named graphs). In this talk, I will present shortcomings in existing techniques, and we propose an index to handle the high number of context values that provenance encoding typically entails. Our experimental results using the Star Schema Benchmark show the feasibility and scalability of our index and query evaluation strategies.
Location: Room "Charles Babbage"
Organizer: Olaf Hartig
Page responsible: Olaf Hartig
Last updated: 2017-01-24