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.

Biomimetic mineral-protein composites formed by an automated alternate soaking process

Biomineralized composite materials found in nature have a compromise of good mechanical properties and relatively small embodied energies in the process of their formation. The Alternate Soaking Process (ASP) is a laboratory technique that has only recently been applied to replicating composite biomineralization. The nexus of the ASP - heterogeneous nucleation - makes it ideal for replicating biominerals where the mineral is templated onto an organic substrate, such as occurs in avian eggshell. Here we demonstrate the deposition of a calcium carbonate gelatin composite on either glass cover slips or demineralized eggshell membranes using an automated ASP. SEM images and FTIR spectra of the resulting mineral show that by altering the amount of gelatin in the growth solutions the final organic component can be controlled accurately in the range of 1-10%, similar to that of natural eggshell. This study shows for the first time the co-precipitation of a CaCO3 - gelatin composite by an ASP and that the organic fraction of this mineral can be tuned to mimic that of natural biomineralized composites. © 2012 Materials Research Society.