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

  Browse A-Z

Log in

University Account External (via Google)

🔐 Log In

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

ℹ️ Information

Higher Order Learning for Classification in Emergency Situations

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.

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

MRC LMB - Sanger Seminar room Economic Epidemiology Seminar Series (supported by CReMic) Seminar room 3.37, Wellcome Trust/MRC Building, Cambridge Biomedical Campus

Other talks

Bigger Picture Talk: Developing future marine engines running on ammonia and other green fuels Dinner Title and abstract tba Content, Caching and Kubernetes: Performance and User Experience in the Browser title abstract tbc Argentina 2005
 

© 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity