In this talk, the speaker will share his experience of using Snorkel, an open-source library for natural language data processing. Snorkel is designed for processing textual data, which can perform weakly supervised data labeling and data augmentation. The content will be interspersed with data processing concepts, key introductions of the thesis, and Snorkel implementation cases. I hope you can understand the advantages and limitations of data augmentation after listening, and know how to use Snorkel to create a data set in your own NLP project.
Eason is a senior student at National Taiwan Normal University (NTNU) and an R&D Engineer at Cohart. He is a passionate researcher and loves to use programming skills and creativity to solve problems. For more about Eason please visit: https://eason.best