University of Cambridge > > Machine Learning @ CUED > Higher Order Learning for Classification in Emergency Situations

Higher Order Learning for Classification in Emergency Situations

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Zoubin Ghahramani.

Traditional machine learning methods (like Naïve Bayes and Latent Dirichlet Allocation) only consider relationships between feature values within individual data instances while disregarding the dependencies that link features across instances. My group at DIMA Cs has developed Higher Order Learning—a general approach to supervised learning that leverages higher-order dependencies between features across instances. Higher Order Learning has been shown to outperform traditional machine learning techniques in multiple accounts. This talk will focus on two Higher Order Learning methods: Higher Order Naïve Bayes (HONB) and Higher Order Latent Dirichlet Allocation (HOLDA). More specifically, my group has been interested in applications of HONB and HOLDA to situations concerning national security and emergency response. Some examples of this include nuclear detection and classification of needs during an emergency using crowdsourced data from social media and text messages.

This talk is part of the Machine Learning @ CUED series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2024, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity