Murray Browne

Algorithms and applications for approximate nonnegative matrix factorization

The development and use of low-rank approximate nonnegative matrix factorization (NMF) algorithms for feature extraction and identification in the fields of text mining and spectral data analysis are presented. The evolution and convergence properties of hybrid methods based on both sparsity and smoothness constraints for the resulting nonnegative matrix factors are discussed. The interpretability of NMF outputs in specific contexts are provided along with opportunities for future work in the modification of NMF algorithms for large-scale and time-varying data sets.

Understanding Search Engines: Mathematical Modeling and Text Retrieval, Second Edition

A discussion of many of the key design issues for building search engines. It emphasizes the important roles that applied mathematics can play in improving information retrieval. The authors discuss not only important data structures, algorithms and software, but also user-centred issues such as interfaces, manual indexing, and document preparation. The authors bridge the gap between applied mathematics and information retrieval. They discuss some of the current problems in information retrieval that may not be familiar to applied mathematicians and computer scientists and present some of the driving computational methods (SVD, SDD) for automated conceptual indexing. This book introduces topics in a non-technical way and provides insights into common problems found in information retrieval. The more mathematical details are provided in sidebars or offset from the regular text.

Modeling the Effects of Land Use on the Quality of Water, Air, Noise, and Habitat for a Five-County Region in Georgia

A computer simulation model, the Regional Simulator (RSim), was constructed to project how landuse changes affect the quality of water, air, noise, and habitat of species of special concern. RSim was designed to simulate these environmental impacts for five counties in Georgia that surround and include Fort Benning. The model combines existing data and modeling approaches to simulate the effects of land-cover changes on: nutrient export by hydrological unit; peak 8-h average ozone concentrations; noise caused by small arms and blasts; and habitat changes for the rare Red-cockaded Woodpecker (Picoides borealis) and gopher tortoise (Gopherus polyphemus). The model also includes submodules for urban growth, new urbanization influenced by existing roads, nonurban land cover transitions, and a new military training area under development at Fort Benning. The model was run under scenarios of business as usual (BAU) and greatly increased urban growth for the region. The projections show that the effects of high urban growth will likely differ from those of BAU for noise and nitrogen and phosphorus loadings to surface water, but not for peak airborne ozone concentrations, at least in the absence of associated increases in industry and transportation use or technology changes. In both scenarios, no effects of urban growth are anticipated for existing populations of the federally endangered Red-cockaded Woodpecker. In contrast, habitat for gopher tortoise in the five-county region is projected to decline by 5 and 40% in the BAU and high urban growth scenarios, respectively. RSim is designed to assess the relative environmental impacts of planned activities both inside and outside military installations and to address concerns related to encroachment and transboundary influences.