Topic : NATURAL LANGUAGE PROCESSING IN PRACTICE – DO’S AND DONT’S
Abstract: While frequently sophisticated NLP methods may show promising results, these methods may not scale out of the box, require expensive infrastructure that your company may not want to afford and can be a maintenance nightmare. In this use-case driven talk, I will discuss how various text analytics problems can be solved using time and resource sensitive directions, how you can later add complexity involved models as well as discuss pitfalls when you immune.
Bio: Kavita Ganesan is a Senior Machine Learning Data Scientist at GitHub. She’s launch the first Machine Learning and Natural Language Processing pipeline at GitHub with the release of GitHub Topics. She was previously an NLP specialist at 3M Health Information Systems where she developed several supervised models to assist with medical billing code predictions. Kavita’s current interest spans graphical methods for large scale text mining and from meaning from source code. Kavita received her Ph.D in Text Analytics and Search from the University of Illinois at Urbana Champaign. Some of her Ph.D level work in sentiment analysis have been adopted by industry practitioners.