In his doctoral dissertation, M.Sc. (Tech.) Zhen Yang studied deep learning-based methods for multi-label text classification, a task in which a single text may belong to several categories at the same time. His work introduces several novel deep learning approaches to this problem, with a particular focus on methods based on convolutional neural networks, and shows that these methods achieve strong and competitive classification performance on benchmark datasets. The dissertation also demonstrates that combining deep learning with binary relevance can improve the accuracy and scalability of multi-label text classification. Overall, the work contributes to research on how deep learning can support more accurate and reliable classification of complex text data when documents need to be assigned multiple labels simultaneously.
The doctoral dissertation of M.Sc. (Tech.) Zhen Yang in the field of Computing Sciences titled Multi-label Text Classification with Deep Learning Models will be publicly examined at the Faculty of Information Technology and Communication Sciences at Tampere University at 12:00 on Monday, 30 March 2026 via Zoom.
The Opponent will be Associate Professor WANG Zhaoxia from Singapore Management University, Singapore. The Custos will be Professor Frank Emmert-Streib from the Faculty of Information Technology and Communication Sciences, Tampere University, Finland.