Sentiment Analysis: Past, Present and Future
Sentiment analysis is both linguistically interesting and crucial to business intelligence. In this talk, I will first describe overly simple methods for sentiment analysis that have been used in the past. Next, I will describe current methods of sentiment analysis and demo an easy to use tool for text and sentiment classification: etcml.com. In the third part, I will introduce more sophisticated models based on recursive deep learning which I believe will eventually supersede currently used algorithms thanks to their improved performance.
Richard Socher is the co-founder and CTO of MetaMind, a young machine learning company focused on pushing the state of the art in AI and making it accessible to many people. More concretely, MetaMind uses deep learning and machine learning techniques to solve many different natural language processing and computer vision problems. In 2014, Richard received his PhD from Stanford University working with Chris Manning and Andrew Ng. He was awarded the 2011 Yahoo! Key Scientific Challenges Award, the Distinguished Application Paper Award at the International Conference on Machine Learning 2011, a Microsoft Research PhD Fellowship in 2012 and a 2013 "Magic Grant" from the Brown Institute for Media Innovation.