A5023940852- Advanced Database Systems and Queries
- Web Data Mining and Analysis
- Data Management and Algorithms
- Social Sciences and Governance
- COVID-19 and healthcare impacts
- Per- and polyfluoroalkyl substances research
- Semantic Web and Ontologies
- Smoking Behavior and Cessation
- Air Quality and Health Impacts
- Multiculturalism, Politics, Migration, Gender
- Complex Network Analysis Techniques
- Weed Control and Herbicide Applications
- Toxic Organic Pollutants Impact
- Scientific Computing and Data Management
- COVID-19 and Mental Health
Le Petit Monde
2021
World Federation of Science Journalists
2017
While the extent of environmental contamination by per- and polyfluoroalkyl substances (PFAS) has mobilized considerable efforts around globe in recent years, publicly available data on PFAS Europe were very limited. In an unprecedented experiment "expert-reviewed journalism" involving 29 journalists seven scientific advisers, a cross-border collaborative project, "Forever Pollution Project" (FPP), drew both methods investigative journalism techniques such as open-source intelligence (OSINT)...
Scientific papers suggesting that smokers are less likely to fall ill with covid-19 being discredited as links the tobacco industry revealed, report <b>Stéphane Horel</b> and <b>Ties Keyzer</b>
Investigative Journalism (IJ, in short) requires combining highly heterogeneous digital datasets coming from a wide variety of sources. We have developed ConnectionLens, system that integrates such sources into single graph and enables users to query the using keywords. The first iteration [7] followed mediator architecture which severely constrained its scalability. Thus, we fully re-engineered system, moving it warehouse architecture, replacing core components (information extraction, data...
Investigative Journalism (IJ, in short) is staple of modern, democratic societies. IJ often necessitates working with large, dynamic sets heterogeneous, schema-less data sources, which can be structured, semi-structured, or textual, limiting the applicability classical integration approaches. In prior work, we have developed ConnectionLens, a system capable integrating such sources into single heterogeneous graph, leveraging Information Extraction (IE) techniques; users then query graph by...