European Commission logo
español español
CORDIS - Resultados de investigaciones de la UE
CORDIS

Semantic EnrichmEnt of trajectory Knowledge discovery

Final Report Summary - SEEK (Semantic EnrichmEnt of trajectory Knowledge discovery)

The main objective of the SEEK project is to envisage a new semantic enriched knowledge discovery process. In contrast to the classical knowledge discovery process for relational data, SEEK tries to formulate a new vision where the semantic aspect (in the sense of the meaning of the movement) is embedded in each step. The challenge is to propose a new process that, taking advantage from the semantic information during the discovery process, may produce exploitable results in several application domains.
This can be done at several levels. The basic step in the semantic enriched knowledge discovery process is the Semantic Trajectory representation, storage and OLAP analysis - objective of Working Package 2. This task is about the investigation of semantic representations of trajectories and techniques for cleaning, transforming, enriching, storing and analysing movement data by using a specialised Data Warehouse. Our goal has been to tailor methods for the newly defined semantic trajectory warehouse, often done exploiting ontologies.
Another step in the discovery process is the “Semantic Knowledge Discovery” - Working Package 3: the objective here has been to investigate knowledge discovery methods that take into account the semantics at several levels: (1) studying methods that use the contextual information to mine different properties of a trajectory (2) proposing post-processing and visualization methods to give a context-dependent meaning to the extracted patterns.
A fourth Working Package is called “New challenges: social aspects of the movement”. Here we traces some ideas for future directions that can benefit from the semantically enriched trajectories, like the use of Linked Open Data for the enrichment, the use of complex networks to represent semantically enriched movement data or the analysis of social networks to predict traffic related problems.
The project produced a total of 40 published or accepted papers, while several other works are ongoing or waiting notification. These papers are published at international high level peer review journals and conferences. Among them, we count three “best papers” in international conferences.
The list of publications is available and updated at the project web site: http://www.seek-project.eu.
Among many other activities, the project also organised seven workshops and created the opportunity for two PhD students co-tutela between project partners