DEMON PRIN22

About the DEMON Project

To study information manipulation we plan to adapt and extend techniques from NLP (to automatically process text) and data/network science (to study information diffusion in online social networks), such as to eventually detect and evaluate the extent to which specific stories are being exploited or weaponized by (malicious) actors in the online information ecosystem.

Data Science

Data collection, clustering of stories and categorization of sub-narratives.

Natural Language Processing

Content-based analysis of manipulation of sub-narratives.

Social Network Analysis

Network-based analysis of manipulation of sub-narratives.

Deliverables and Papers

In the following the list of Deliverables and Papers produced by the project

Deliverable on WP1

WP1: Data collection, clustering of stories and categorization of sub-narratives

Deliverable on WP2

WP2: Content-based analysis of stories and sub-narratives

Deliverable on WP3

WP3: Network-based analysis of manipulation of sub-narratives

Deliverable on WP4

WP4: Integration of content- and network-based approaches to measure manipulation of online information

Matteo Fenu

University of Cagliari

The DEMON project puts together two partners with strong, complementary research teams: UNICA has a high international reputation in the Semantic Web and the Data Mining communities and has already successfully applied NLP methods for information extraction from unlabeled text such as those available in social media; POLIMI has worked on the spread of online mis/disinformation using network science and machine learning techniques, showing how social media interactions allow to identify and characterize mis/disinformation in a variety of contexts.