With the proliferation of an electronic commerce and knowledge economy environment, both organizations and individuals generate and consume a large amount of online information, typically available as textual documents. To manage the ever-increasing volume of documents, organizations and individuals typically organize their documents into categories to facilitate document management and subsequent information access and browsing. However, document grouping behaviors are intentional acts, reflecting individuals¡¦ (or organizations¡¦) preferential perspective on semantic coherency or relevant groupings between subjects. Thus, an effective document clustering needs to address the described preferential perspective on document grouping and support personalized document clustering. In this thesis, we designed and implemented a personalized document clustering approach by incorporating individual¡¦s partial clustering into the document clustering process. Combining two document representation methods (i.e., feature refinement and feature weighting) with two clustering processes (i.e., pre-cluster-based and atomic-based), four personalized document clustering techniques are proposed. Using the clustering effectiveness achieved by a traditional content-based document clustering technique as performance benchmarks, our evaluation results suggest that use of partial clusters would improve the document clustering effectiveness. Moreover, the pre-cluster-based technique outperforms the atomic-based one, and the feature weighting method for document representation achieves a higher clustering effectiveness than the feature refinement method does.

Considerable evidence exists that the majority of low tension water flow through highly structured clayey soil occurs in a small fraction of total pore space and that the flow paths converge as depth increases. In structured clayey soils, water tends to flow in locations where macroporosity is high and roots tend to enjoy this condition as well. Water reduces the strength and mechanical impedance of the soil. Mechanical impedance of clayey soils tends to be extremely high when the soils are dry so one might expect that there would be a positive spatial correlation between the location of roots and the location of water flow paths in highly structured clayey soils. Understanding the relationship between the location of roots in soil relative to the location of water flow paths is important in understanding how plants obtain nutrients and water for growth, and it would also be of considerable importance in phytoremediation research and research into the prevention of groundwater contamination. This experiment was designed to map the locations of flow paths and roots and then measure the spatial association of the two. A pasture on Ship’s clay along the Brazos River was chosen as the research site. Three plots were irrigated with an Erioglaucine dye solution used to stain flow paths. After irrigation the soil was excavated to a depth of 25 cm. On the resulting horizontal plane the dye stain pattern was mapped using photography. The locations of roots were mapped on clear plastic sheets. During mapping the roots were categorized by size. The mapping procedure was repeated at depth of 45 cm and 75 cm for all plots. The root maps were overlaid on the photographic images and analyzed for a spatial association. There was no evidence the smallest (> 1 mm diameter) roots were not randomly distributed. The results did show that the larger roots were not randomly distributed, and evidence pointed to a clustering of roots in and around the dye stained flow paths. However, the data fell short of establishing a spatial association. The lack of more conclusive data was likely the result of inaccuracies in the mapping.

The Objective of this thesis is to talk about the usage of Fuzzy Logic in pattern recognition. There are different fuzzy approaches to recognize the pattern and the structure in data. The fuzzy approach that we choose to process the data is completely depends on the type of data. Pattern reorganization as we know involves various mathematical transforms so as to render the pattern or structure with the desired properties such as the identification of a probabilistic model which provides the explaination of the process generating the data clarity seen and so on and so forth. With this basic school of thought we plunge into the world of Fuzzy Logic for the process of pattern recognition. Fuzzy Logic like any other mathematical field has its own set of principles, types, representations, usage so on and so forth. Hence our job primarily would focus to venture the ways in which Fuzzy Logic is applied to pattern recognition and knowledge of the results. That is what will be said in topics to follow. Pattern recognition is the collection of all approaches that understand, represent and process the data as segments and features by using fuzzy sets. The representation and processing depend on the selected fuzzy technique and on the problem to be solved. In the broadest sense, pattern recognition is any form of information processing for which both the input and output are different kind of data, medical records, aerial photos, market trends, library catalogs, galactic positions, fingerprints, psychological profiles, cash flows, chemical constituents, demographic features, stock options, military decisions.. Most pattern recognition techniques involve treating the data as a variable and applying standard processing techniques to it.

I consider statistical inference for clustering, that is the arrangement of experimental units in homogeneous groups. In particular, I discuss clustering for multivariate binary outcomes. Binary data is not very informative, making it less meaningful to proceed with traditional (deterministic) clustering methods. Meaningful inference needs to account for and report the considerable uncertainty related with any reported cluster arrangement. I review and implement an approach that was proposed in the recent literature. / text

This thesis covers the analysis of sparse and irregular fuel consumption data of long distance haulage articulate trucks. It is shown that this kind of data is hard to analyse with multivariate as well as with functional methods. To be able to analyse the data, Principal Components Analysis through Conditional Expectation (PACE) is used, which enables the use of observations from many trucks to compensate for the sparsity of observations in order to get continuous results. The principal component scores generated by PACE, can then be used to get rough estimates of the trajectories for single trucks as well as to detect outliers. The data centric approach of PACE is very useful to enable functional analysis of sparse and irregular data. Functional analysis is desirable for this data to sidestep feature extraction and enabling a more natural view on the data.

Clustering, which is partitioning data into groups of similar objects, has a wide range of applications. In many cases unstructured data makes up a significant part of the input. Attempting to cluster such part of the data, which can be referred to as noise, can disturb the clustering on the remaining domain points. Despite the practical need for a framework of clustering that allows a portion of the data to remain unclustered, little research has been done so far in that direction. In this thesis, we take a step towards addressing the issue of clustering in the presence of noise in two parts. First, we develop a platform for clustering that has a cluster devoted to the "noise" points. Second, we examine the problem of "robustness" of clustering algorithms to the addition of noise. In the first part, we develop a formal framework for clustering that has a designated noise cluster. We formalize intuitively desirable input-output properties of clustering algorithms that have a noise cluster. We review some previously known algorithms, introduce new algorithms for this setting, and examine them with respect to the introduced properties. In the second part, we address the problem of robustness of clustering algorithms to the addition of unstructured data. We propose a simple and efficient method to turn any centroid-based clustering algorithm into a noise robust one that has a noise cluster. We discuss several rigorous measures of robustness and prove performance guarantees for our method with respect to these measures under the assumption that the noise-free data satisfies some niceness properties and the noise satisfies some mildness properties. We also prove that more straightforward ways of adding robustness to clustering algorithms fail to achieve the above mentioned guarantees.

Clustering microarray data is a helpful way of identifying genes which are biologically related. Unfortunately, when attempting to cluster microarray data, certain issues must be considered including: the uncertainty in the number of true clusters; the expression of a given gene is often a ected by the expression of other genes; and microarray data is usually high dimensional. This thesis outlines a Bayesian in nite Gaussian mixture model which addresses the issues outlined above by: not requiring the researcher to specify the number of clusters expected, applying a non-diagonal covariance structure, and using mixtures of factor analyzers and extensions thereof to structure the covariance matrix such that it is based on a few latent variables. This approach will be illustrated on real and simulated data.

No description available.

In this Thesis we investigate the evolutionary histories of faint field galaxies using extremely deep optical and near-infrared photometry. Our work is centred on a 50 arcmin(^2) region at high galactic latitude which we call "The William Herschel Deep Field" (WHDF). In this work we describe three new near-infrared surveys of this field. In considering both this infrared data and the existing optical data, our broad aims are to increase our understanding of both the growth of galaxy clustering in the Universe and also to determine the star-formation histories of the field galaxy population. We consider our observations primarily in the context of luminosity evolution models in low density universes, but alternative scenarios are considered. Near-infrared galaxy counts derived from our catalogues are consistent with the predictions of our models, without the need for a steep faint-end slope for the galaxy luminosity function. We find that optical-infrared colour distributions of infrared-selected galaxies in the WHDF are deficient in red, early-type galaxies. This is consistent with the predictions of evolutionary models in which these systems have a small amount of on-going star-formation. We measure the amplitude of galaxy clustering in the WHDF for galaxies selected in optical and near-infrared bandpasses using the projected two-point correlation function. By comparing our measured clustering amplitudes with the predictions of our models we find that in all bandpasses the growth of galaxy clustering is approximately fixed in proper co-ordinates, again assuming a low-density Universe. Finally, an analysis of errors on the correlation function measurements suggest that discrepancies between our work and those of other authors may be explained by an underestimation of statistical errors.

This thesis describes a study of the clustering properties of galaxies and clusters of galaxies as measured by the power spectrum (P(k)) and the counts in cells statistic. The samples used are the optical Stromlo-APM galaxy survey, the APM cluster survey and the IRAS 1.2Jy, QDOT and PSCz surveys. Throughout, N-body simulations, for a variety of cosmological models, are used to test methods and to supplement analytic error estimates. For the Stromlo-APM sample the amplitude of the power spectrum is dependent on galaxy morphology. Early-type galaxies show a higher clustering amplitude than late-type galaxies by a factor of ~ 1.8. There is also tentative evidence for some dependence of the clustering amplitude on galaxy luminosity. The parameter Ω0.6/b is estimated via a comparison with the real-space power spectrum of the two-dimensional APM galaxy survey. For APM clusters the power spectrum is measured to very small wave numbers, with a possible detection of the expected turn-over. The results are inconsistent with the standard cold dark matter model. The shape of P( k) for clusters is approximately the same as that for Stromlo-APM galaxies but amplified by a factor of ~ 3.5. The power spectrum of the QDOT sample depends sensitively on the galaxy weighting scheme, probably due the manner in which the region of the Hercules supercluster is sampled. A best estimate of the power spectrum of IRAS galaxies is computed by combining the IRAS l.2Jy and QDOT samples. The PSCz galaxy power spectrum is also computed. The PSCz galaxies have a clustering amplitude twice that of optical galaxies. A similar result is found from a joint counts in cells analysis. Redshift-space distortions in the PSCz sample are analysed using a spherical harmonic decomposition of the density field. The value of Ω0.6/b = 1 is ruled out by this analysis at the 2σ significance level.