Improving spam filtering in enterprise email systems with blockchain-based token incentive mechanism
Abstract
Spam has caused serious problems for email systems. To address this issue, numerous spam filter algorithms have been developed, all of which require extensive training on labeled spam datasets to obtain the desired filter performance. However, users' privacy concerns and apathy make it difficult to acquire personalized spam data in real-world applications. When it comes to enterprise email systems, the problem worsens because enterprises are extremely sensitive to the possible disclosure of confidential information during the reporting of spam to the cloud. Targeting these obstacles, this study proposes a blockchain-based token incentive mechanism, with the aim of encouraging users to report spam while protecting business secrets and ensuring the transparency of reward rules. The proposed mechanism also enables a decentralized ecosystem for token circulation, fully utilizing the advantages of blockchain technologies. We developed a prototype of the proposed system, on which we conducted a user experiment to verify our design. Results indicate that the proposed incentive mechanism is effective and can raise the probability of spam reporting by more than 1.4 times.
Recommended Citation
Xu, Xiaoying; Tian, Minghao; and Li, Zhihong, "Improving spam filtering in enterprise email systems with blockchain-based token incentive mechanism" (2022). ICEB 2022 Proceedings (Bangkok, Thailand). 23.
https://aisel.aisnet.org/iceb2022/23