Nowadays, publishing and retrieving information through web services (e.g., social networks, news aggregators) has become extremely easy. However, the lack of quality controls has made possible for fake news – inaccurate or sometimes forged information – to be spread.

The blockchain is seen as a promising technology to address this issue. In particular, the blockchain allows to immutably and persistently store information on the provenance of each piece of news. However, the blockchain alone cannot certify the authenticity of the information. Instead, it must be combined with a so-called “oracle”. An oracle is a service made of machines or humans that validate the information before storing it in a blockchain.

In our paper, we present DeepThought, a blockchain oracle that combines voting and reputation schemes in order to certify with high confidence the authenticity of the information. Existing web services can exploit DeepThought to validate information or to retrieve already certified ones. Compared to existing techniques, DeepThought makes it extremely difficult for malicious users to collude and corrupt the service.

This work is partially funded by the Otto Moensted Foundation:


Marco Di Gennaro is a Computer Science and Engineering MSc Student at Politecnico di Milano. His research interests include Cybersecurity, Graph Representation Learning, and Blockchain.

Lorenzo Italiano is a PhD student in Information Technology at Politecnico di Milano. His interests are related to cybersecurity and machine learning topics mostly applied in the smart mobility field.

Giovanni Meroni is an Assistant Professor at Technical University of Denmark. His research interests include Information Systems, Business Process Management, Blockchain, and the Internet of Things.

Giovanni Quattrocchi is an adjunct professor and post-doc at Politecnico di Milano, Italy. His research interests include self-adaptive systems, Edge- and Blockchain-based systems, and software engineering for AI.

Categories: General