Luke Carrivick

Learn More
We apply a new Bayesian data analysis technique (latent process decomposition) to four recent microarray datasets for breast cancer. Compared to hierarchical cluster analysis, for example, this technique has advantages such as objective assessment of the optimal number of sample or gene clusters in the data, penalization of overcomplex models fitting to(More)
In this paper we compare a variety of unsupervised probabilistic models used to represent a data set consisting of textual and image information. We show that those based on latent Dirichlet allocation (LDA) out perform traditional mixture models in likelihood comparison. The data set is taken from radiology; a combination of medical images and consultants(More)
Background: We present a variational Bayesian approach to inference in a probabilistic model for microarray gene expression data. The algorithmic approach efficiently maximises the probability of the model given the data and provides an unbiased indication of the most probable number of processes or soft clusters in the data. Compared to hierarchical(More)
  • 1