Machine Learning 301: Learning from Text
Enjoy watching the new video on InterSystems Developers YouTube Channel:
⏯ Machine Learning 301: Learning from Text
In this hour, we will build off of the algorithms and material that we learned in our prior Machine Learning Lunch and Learns.
🗣 @Don Woodlock, Vice President, HealthShare Platforms
🗣 Aliaa Atwi, Software Developer, HealthShare Platforms
So far, we have covered Machine Learning examples and datasets that all are numeric. But much of the data in the world, and especially in healthcare, is textual data – clinical notes, coded descriptions and classification of information, etc. In this session, we will dive into the world of text. We will learn how to deal with category data through working with the Titanic dataset. We will use this information to predict who will survive the Titanic disaster based on each passenger’s attributes – numbers and text like age, fare, class of travel, gender, city they embarked from, etc. Secondly we will dive into free text and see how to learn from documents. In this case study, we will explore the IMDB Movie Review database and learn how we can train a model to read the textual movie review and predict whether the author liked or disliked the movie automatically. This is the field of Sentiment Analysis and is used by many organizations to review their public perception based on news and social media content.
2 good datasets, 2 good speakers, 1 good hour. Please welcome! 👏🏼
Don't forget to watch two previous webinars:
➡️ Machine Learning 101: How does it actually work?
➡️ Machine Learning 201: The Mighty Neural Network
Enjoy and stay tuned!