Cluster analysis of heterogeneous rank data

Cluster analysis of ranking data, which occurs in consumer questionnaires, voting forms or other inquiries of preferences, attempts to identify typical groups of rank choices. Empirically measured rankings are often incomplete, i.e. different numbers of filled rank positions cause heterogeneity in the data. We propose a mixture approach for clustering of heterogeneous rank data. Rankings of different lengths can be described and compared by means of a single probabilistic model. A maximum entropy approach avoids hidden assumptions about missing rank positions. Parameter estimators and an efficient EM algorithm for unsupervised inference are derived for the ranking mixture model. Experiments on both synthetic data and real-world data demonstrate significantly improved parameter estimates on heterogeneous data when the incomplete rankings are included in the inference process.

Fracture and energy partitioning in uncooked and cooked noodles

Humans perform fascinating science experiments at home on a daily basis when they undertake the modification of natural and naturally-derived materials by a cooking process prior to consumption. The material properties of such foods are of interest to food scientists (texture is often fundamental to food acceptability), oral biologists (foods modulate feeding behavior), anthropologists (cooking is probably as old as the genus Homo and distinguishes us from all other creatures) and dentists (foods interact with tooth and tooth replacement materials). Materials scientists may be interested in the drastic changes in food properties observed over relatively short cooking times. In the current study, the mechanical properties of one of the most common (and oldest at 4,000+ years) foods on earth, the noodle, were examined as a function of cooking time. Two types of noodles were studied, each made from natural materials (wheat flour, salt, alkali and water) by kneading dough and passing them through a pasta-making machine. These were boiled for between 2-14 min and tested at regular intervals from raw to an overcooked state. Cyclic tensile tests at small strain levels were used to examine energy dissipation characteristics. Energy dissipation was >50% per cycle in uncooked noodles, but decreased by an order of magnitude with cooking. Fractional dissipation values remained approximately constant at cooking times greater than 7 min. Overall, a greater effect of cooking was on viscoplastic dissipation characteristics rather than on fracture resistance. The results of the current study plot the evolution of a viscoplastic mixture into an essentially elastic material in the space of 7 minutes and have broad implications for understanding what cooking does to food materials. In particular, they suggest that textural assessment by consumers of the optimally cooked state of food has a definite physical definition. © 2007 Materials Research Society.

Gaussian mixture modeling with Gaussian process latent variable models

Density modeling is notoriously difficult for high dimensional data. One approach to the problem is to search for a lower dimensional manifold which captures the main characteristics of the data. Recently, the Gaussian Process Latent Variable Model (GPLVM) has successfully been used to find low dimensional manifolds in a variety of complex data. The GPLVM consists of a set of points in a low dimensional latent space, and a stochastic map to the observed space. We show how it can be interpreted as a density model in the observed space. However, the GPLVM is not trained as a density model and therefore yields bad density estimates. We propose a new training strategy and obtain improved generalisation performance and better density estimates in comparative evaluations on several benchmark data sets. © 2010 Springer-Verlag.

Structured precision modelling with Cholesky basis superposition for speech recognition

Structured precision modelling is an important approach to improve the intra-frame correlation modelling of the standard HMM, where Gaussian mixture model with diagonal covariance are used. Previous work has all been focused on direct structured representation of the precision matrices. In this paper, a new framework is proposed, where the structure of the Cholesky square root of the precision matrix is investigated, referred to as Cholesky Basis Superposition (CBS). Each Cholesky matrix associated with a particular Gaussian distribution is represented as a linear combination of a set of Gaussian independent basis upper-triangular matrices. Efficient optimization methods are derived for both combination weights and basis matrices. Experiments on a Chinese dictation task showed that the proposed approach can significantly outperformed the direct structured precision modelling with similar number of parameters as well as full covariance modelling. © 2011 IEEE.