NTT Group paper selected for IEEE TIFS, a competitive international journal on information security

The following NTT researcher's paper has been accepted by the IEEE Transactions on Information Forensics and Security (TIFS) (Impact Factor: 7.231), known as the top international journal in the security field.

Social Informatics Laboratories of Service Innovation Laboratory Group

Satoshi Furutani

Social Informatics Laboratories of Service Innovation Laboratory Group

Toshiki Shibahara

Social Informatics Laboratories of Service Innovation Laboratory Group

Mitsuaki Akiyama

The accepted paper, "Interpreting Graph-based Sybil Detection Methods as Low-Pass Filtering", addresses the problem of fake accounts (Sybil) exploited for spreading spam and fake news in social media. The paper establishes a methodology to theoretically compare and analyze various existing detection methods by interpreting Sybil detection problem as signal recovering problem in graph signal processing. Based on this methodology, the paper identifies the requirements for high performance of the Sybil detection method and proposes a new detection method that satisfies these requirements. Numerical experiments show that it performs more consistently and better than existing methods on graphs with various structural properties. This technology provides the basis for early detection and countermeasures against fake accounts systematically operated by attackers, leading to a safe and healthy online space.

NTT will continue to work on fundamental studies aimed at creating innovative technologies, and contribute to a safe and secure society through cyber security research and development.

[Reference]

S. Furutani, T. Shibahara, M. Akiyama and M. Aida, "Interpreting Graph-Based Sybil Detection Methods as Low-Pass Filtering," in IEEE Transactions on Information Forensics and Security, vol. 18, pp. 1225-1236, 2023, doi: 10.1109/TIFS.2023.3237364.
https://ieeexplore.ieee.org/document/10018255