The feature interaction problem is prominent in telephone service development. Through a number of case studies, we have discovered that no single semantic framework is suitable f...
J. Paul Gibson, Geoff Hamilton, Dominique Mé...
This paper proposes an approach to classification of adjacent segments of a time series as being either of classes. We use a hierarchical model that consists of a feature extract...
We present Promodes, an algorithm for unsupervised word decomposition, which is based on a probabilistic generative model. The model considers segment boundaries as hidden variable...
The problem of joint modeling the text and image components of multimedia documents is studied. The text component is represented as a sample from a hidden topic model, learned wi...
Nikhil Rasiwasia, Jose Costa Pereira, Emanuele Cov...
We develop a new component analysis framework, the Noisy-Or Component Analyzer (NOCA), that targets high-dimensional binary data. NOCA is a probabilistic latent variable model tha...