Effective organisational knowledge management is one of the main challenges that organisations face today namely, the management of information that appears in both structured and unstructured heterogeneous, autonomous, data sources. The challenge is exacerbated where organisations use multiple knowledge sources such as news feeds, scientific, technical and management reports and where these sources are spread over different internal and external sites. The ability to access, manage and analyse such disparate data sources, especially unstructured textual data, is paramount in order to gain competitive advantage for every organisation.
The overall objective of PARMENIDES is to develop an ontology driven systematic approach for integrating the entire process of information gathering, processing and analysis.
Detailed objectives of this output oriented project are:
-Automatic detection of entities and events from unstructured textual data, including identification of temporal attributes and temporal relations between them.
-Extending existing ontologies by automatic discovery of terms and terminological relations, as the underlying document collections change.
-Integrating factual and derived information at the database level.
-Developing mechanisms to permanently monitor pattern changes.
-Determination of temporal validity of information extracted from document collections.
-Prediction of events of interest from extracted temporal information.
-Automatic categorisation, contextualisation and summarisation of documents aiming at leveraging passive information collections into knowledge assets.
DESCRIPTION OF WORK
-WP1 Project management. Co-ordinate and manage the project and provide communication media between partners and the commission.
Also consolidate project planning, control, progress and milestone reports, cost statements and budgetary overviews and motivate and monitor exploitation of results.
-WP2 Domain and system requirements specification and design. Characterise case study domains and identification of user requirements on these case studies. Definition of standards to be used for knowledge structures and exchange formats and overall architecture supporting the case studies.
-WP3 Semi-automatic ontology construction. Development of domain ontologies and terminology supporting the case studies.
-WP4 Semantic-based annotation. Development of NKRL-like tools to enable semi-automatic conceptual annotation of documents, especially focussing on handling temporal phenomena.
-WP5 Integration in document warehouse. Implementation of data model and integration of required tools into a document warehouse supporting temporal issues.
-WP6 Temporal knowledge discovery. Development of components for temporal knowledge discovery including association rules, clustering, and time series. Also aiming at components for pattern and ontology evolution discovery.
-WP7 Domain case studies and evaluation. Validation of methods and tools developed in the project in three domains: foods, biotechnology and defence. Evaluation from a quality in use viewpoint using the EAGLES/ISLE framework.
-WP8 Dissemination and Exploitation. Consolidate results of the project through dissemination and exploitation.
Funding SchemeCSC - Cost-sharing contracts
16777 Ellinikon, Athens
15500 Holargos, Athens
14231 N. Ionia, Athens
3013 AL Rotterdam
1211 Geneve 4
BS8 4AN Bristol