Abstract:
Topic models are statistical models that discover thematic "topics" that pervade a large unstructured collection of documents. Topic Modeling algorithms can alternatively be interpreted as dimensionality reduction algorithms that provide sparse outputs. These models are among the most commonly seen Bayesian Machine Learning models in industry and are used heavily in different contexts from document tagging, topical search indexing to data visualization, image tagging etc.
This talk aims to demystify these models, show how to productionize such things, and also give some examples of how they are used in industry.
Speaker Bio:
Arnab Bhadury is a Machine Learning engineer and built the topic extraction engine in production at Flipboard. His interests lie predominantly in Bayesian Machine Learning, topic modeling and recommender systems.
Agenda:
17:45 - Doors Open & Mingle
18:15 - Doors Closed & Introductions
18:30 - Talk
19:25 - Questions & Discussion
19:45 - Close