Project Description

SEEK - Semantic Enrichment of trajectory Knowledge discovery is a Marie Curie Project funded by EU in the PEOPLE program as an IRSES 2011 scheme. The project activity starts on March, 1 2012 and lasts 42 months. This project funds the exchange of research personnel between participant organizations from EU to international partners and vice-versa.

Abstract

A flood of data pertinent to moving objects is available today, and will be more in the near future, particularly due to the automated collection of data from personal devices such as mobile phones and other location-aware devices. Such wealth of data, referenced both in space and time, may enable novel classes of applications of high societal and economic impact, provided that the discovery of consumable and concise knowledge out of these raw data is made possible.

The fundamental hypothesis is that it is possible, in principle, to aid citizens in their mobile activities by analysing the traces of their past activities by means of data mining techniques. For instance, behavioural patterns derived from mobile trajectories may allow inducing traffic flow information, capable to help people travelling efficiently, to help public administrations in traffic-related decision making for sustainable mobility and security management.
Behavioral patterns can be extracted through a knowledge discovery process where positioning data collected from mobile devices are first transformed in semantically enriched trajectory data stored in a database. Then, these data are loaded in a data warehouse and analysed with OLAP operations that allow summarization of the trajectories features. Mobility patterns, the most common movements emerging from data, are computed with suitable spatio-temporal data mining algorithms. A further semantic enrichment step is needed to give context-dependent meaning to the discovered patterns.

The scientific activity is organized into 4 working packages:

  1. Application Scenarios: Semantics is a key aspect of this project, and semantics is mainly given by the application. Therefore, we believe that the development and the experimentation of the new techniques must be carried out and verified by specific case studies representing different application domains. The choice of three application scenarios offers the possibility to develop techniques that are valid in very different contexts. The objective of each task is to provide a showcase providing the list of the application requirements, contextual information such as maps, domain knowledge and a dataset.
  2. Semantic Trajectory representation, storage and OLAP analysis: This work package has two main goals: investigation of semantic representations of trajectories and techniques for cleaning, transforming, enriching, storing and analysing movement data by using a specialised Data Warehouse. The traditional representation of a trajectory only takes into account the geometric aspect of the movement, disregarding the representation of the meaning of the movement. In oder to discover the important places visited and for how long, as well as the behavior represented by such movement we have to investigate an adequate data model for semantic trajectory database and novel operations. A step forward is the definition of a Data Warehouse for semantic trajectories. DWs offer a powerful technological support to visual analysis of movement data by efficiently aggregating the data in various ways and at different spatial and temporal scales. Our goal is to tailor existing methods for the newly defined semantic trajectory warehouse possibly exploiting ontologies.
  3. Semantic Knowledge Discovery: The objective of this work package is to investigate knowledge discovery methods that take into account the semantic aspect of the movement data not only the geometric aspect of a raw trajectory. This can be done at different levels: (1) studying methods that use the contextual information to mine different properties of a trajectory taking into account these semantic aspects (2) proposing post-processing and visualisation methods to give a context-dependent meaning to the extracted patterns. These approaches can be integrated in a new notion of semantic-enhanced knowledge discovery process that defines a new notion of progressive and interactive semantic-enrichment process. We also propose to have a task studying the specific problem of privacy in the knowledge discovery process, disclosing possible attacks and proposing methods to cope with them.
  4. New challenges: social aspects of the movement. The semantic - enriched knowledge discovery process outlined in the previous work packages enables a number of innovative challenging research issues. Here, we intend to build a roadmap on some "on the frontier" research issues that may arise from such movement knowledge. This will be essential in establishing long term collaborations among the members of the project on the new proposed issues. We identified two main topics: social networks and trajectory interactions. Here, two partners, CNR and UNB may bring to the consortium the necessary knowledge about these two aspects to outline a roadmap of research in these two main directions.