Thien Huu Nguyen

We develop novel multilingual datasets and transfer learning methods for information extraction in natural language processing.

We develop novel representation learning methods for authorship attribution, privacy preservation, and model explainability.

We develop novel methods to induce text structures to improve the performance, robustness, and flexibility of event extraction systems.

Our research introduces effective methods, systems, and datasets for the business of Adobe.

Our CBL consortium focuses on the development of large-scale deep learning algorithms (DL) and DL-enabled big data applications in a broad spectrum of disciplines (i.e., healthcare, science, security, and education).

We develop state-of-the-art multilingual methods for information extraction that can perform event extraction for multiple languages with increasing fine-grained event types and argument roles.