The world of procurements and eProcurement generates daily large amounts of data, that represent knowledge of great economical value both for individual companies and for public organisations wishing to achieve a better understanding of a given market. However, such data remains dificult to explore and analyze as it is being kept isolated from other sources of knowledge, in dedicated systems. In this paper, we present an ongoing work on extracting and linking data from the European 'Tenders Electronic Daily' system, which publishes approximately 1,500 tenders five times a week. We specifically show how such information is dynamically extracted and linked to external datasets, and how the created links enrich the original data, introducing new perspectives to its analysis. We show tools we developed to support such 'linked data-based' analysis of data, and report on the lessons learnt from our experience in building a linked data application with potential for real-life use in knowledge extraction.
LOTED: Exploiting Linked Data in Analyzing European Procurement Notices / Valle, F; D'Aquin, M; DI NOIA, Tommaso; Motta, E.. - 631:(2010), pp. 52-63. (Intervento presentato al convegno 1st Workshop on Knowledge Injection into and Extraction from Linked Data, KIELD 2010 tenutosi a Lisbon- Portugal nel October 15, 2010).
LOTED: Exploiting Linked Data in Analyzing European Procurement Notices
DI NOIA, Tommaso;
2010-01-01
Abstract
The world of procurements and eProcurement generates daily large amounts of data, that represent knowledge of great economical value both for individual companies and for public organisations wishing to achieve a better understanding of a given market. However, such data remains dificult to explore and analyze as it is being kept isolated from other sources of knowledge, in dedicated systems. In this paper, we present an ongoing work on extracting and linking data from the European 'Tenders Electronic Daily' system, which publishes approximately 1,500 tenders five times a week. We specifically show how such information is dynamically extracted and linked to external datasets, and how the created links enrich the original data, introducing new perspectives to its analysis. We show tools we developed to support such 'linked data-based' analysis of data, and report on the lessons learnt from our experience in building a linked data application with potential for real-life use in knowledge extraction.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.