Inferring latent activities is a challenging task that has received little research attention. Most of the existing approaches have used parametric methods from machine learning and data mining. Standard supervised learning and unsupervised learning techniques are the most popular ones used. These include Support Vector Machines (SVMs), Naive Bayes, and hidden Markov models (HMM), to name a few. Typical unsupervised learning methods include Gaussian mixture models (GMM), K-means, and latent Dirichlet allocation (LDA) (see Section 6.2). These methods are parametric models in the sense that, once models are learned from a particular datase...
- 6: Learning Latent Activities from Social Signals with Hierarchical Dirichlet Processes
- from Plan, Activity, and Intent Recognition
- Publisher: Morgan Kaufmann
- Released: March 2014
Test note - look at this!
Share this highlighthttp://www.safaribooksonline.com/a/plan-activity-and/33899/