Efficient Labelling of Large Datasets for NLP Tasks

Explore efficient techniques for labeling large datasets in Natural Language Processing tasks in this 29-minute conference talk by Nichola Roberts at the Data Science Festival. Learn how to grow a comprehensive labeled dataset from a small initial set of examples, automating the process as much as possible. Discover various methods, including traditional approaches like Support Vector Machines and cutting-edge techniques such as few-shot and zero-shot learning. Gain insights into overcoming the challenge of creating high-quality labeled datasets for NLP projects, essential for most data scientists in the field.