Skip to collection list
Skip to video grid
USA: Day 2 - Hands-On Labs
Using Deep Learning for Entity Detection and Intent Extraction in Natural Language
Peter Izsak, Deep Learning Data Scientist, NLU Israel, AI Products Group, Intel
In this session, we’ll get a glimpse into how algorithms can extract meaning out of our language. Entity detection and Intent extraction algorithms are key components in natural language understanding applications, such as, virtual assistants, chat-bots and information extraction systems. Recent Recurrent Neural Networks utilizing LSTMs and sequence tagging techniques helped us to improve the accuracy of the models compared to prior approaches. We will briefly present the tasks of named entity recognition and intent extraction, show best practices for developing deep learning based models for sequential tagging, and provide a hands-on demonstration of how to prepare, train and deploy such a model.