The 6th Frontier Forum for Digital Technology and Finance

Abstract:
Virtual and reality are constantly interacting, and the content of cyberspace is accumulating at an unimaginable speed. In order to implement the "Regulations on the Ecological Governance of Network Information Content", further protect the national network security and user information security, and maintain a good network ecology, the traditional method of relying on content reviewers cannot adapt to the current network information content governance needs. Based on artificial intelligence, deep learning, and user interaction models, it automatically detects images, texts, videos and audio uploaded by users, and identifies pornographic, advertising, and political-related issues, explosives, sensitive people and other content, empowering intelligent content auditing and network ecological governance. The research team proposes text review for metaphors, novel review content combined with user reading behavior, and product review combined with user search behavior. By extracting commonalities in different scenarios, identifying the characteristics of different scenarios, solving the problem of data sparseness in specific scenarios, and providing solutions for the governance of network information content ecology.

Speaker Introduction:
Wei Lu, Dean of the School of Information Management of Wuhan University, Distinguished Professor of Changjiang Scholar, Distinguished Professor of Luojia, Doctoral Supervisor, Director of the Institute of Information Retrieval and Knowledge Mining. He went to the United Kingdom, Denmark, the Netherlands and the United States for visits or postdoctoral research. He has published more than 100 papers, including more than 50 SSCI/SCI/EI papers, and presided over and participated in the preparation of many books.