The application of adaptive partitioned random search in feature selection problem

Feature selection is one of important and frequently used techniques in data preprocessing. It can improve the efficiency and the effectiveness of data mining by reducing the dimensions of feature space and removing the irrelevant and redundant information. Feature selection can be viewed as a global optimization problem of finding a minimum set of M relevant features that describes the dataset as well as the original N attributes. In this paper, we apply the adaptive partitioned random search strategy into our feature selection algorithm. Under this search strategy, the partition structure and evaluation function is proposed for feature selection problem. This algorithm ensures the global optimal solution in theory and avoids complete randomness in search direction. The good property of our algorithm is shown through the theoretical analysis.

Lo-Fi matchmaking: A study of social pairing for backpackers

It is technically feasible for mobile social software such as pairing or ‘matchmaking’ systems to introduce people to others and assist information exchange. However, little is known about the social structure of many mobile communities or why they would want such pairing systems. While engaged in other work determining requirements for a mobile travel assistant we saw a potentially useful application for a pairing system to facilitate the exchange of travel information between backpackers. To explore this area, we designed two studies involving usage of a low-fidelity role prototype of a social pairing system for backpackers. Backpackers rated the utility of different pairing types, and provided feedback on the social implications of being paired based on travel histories. Practical usage of the social network pairing activity and the implications of broader societal usage are discussed.

Long-term survivor mixture model with random effects: application to a multi-centre clinical trial of carcinoma

A mixture model incorporating long-term survivors has been adopted in the field of biostatistics where some individuals may never experience the failure event under study. The surviving fractions may be considered as cured. In most applications, the survival times are assumed to be independent. However, when the survival data are obtained from a multi-centre clinical trial, it is conceived that the environ mental conditions and facilities shared within clinic affects the proportion cured as well as the failure risk for the uncured individuals. It necessitates a long-term survivor mixture model with random effects. In this paper, the long-term survivor mixture model is extended for the analysis of multivariate failure time data using the generalized linear mixed model (GLMM) approach. The proposed model is applied to analyse a numerical data set from a multi-centre clinical trial of carcinoma as an illustration. Some simulation experiments are performed to assess the applicability of the model based on the average biases of the estimates formed. Copyright (C) 2001 John Wiley & Sons, Ltd.

A sensitive vorticity gauge using rotated porphyroblasts, and its application to rocks adjacent to the Alpine Fault, New Zealand

Variable aspect ratio porphyroblasts deformed in non-coaxial flow. and internally containing rotated relicts of an external foliation, can be used to characterise plane strain flow regimes. The distribution obtained by plotting the orientation of the long axis of such grains, classified by aspect ratio, against the orientation of the internal foliation is potentially a sensitive gauge of both the bulk shear strain (as previously suggested) and kinematic vorticity number. We illustrate the method using rotated biotite porphyroblasts in the Alpine Schist: a sequence of mid-crustal rocks that have been ramped to the surface along the Alpine Fault. a major transpressional plate boundary. Results indicate that, at distances greater than or equal to similar to1 km from the fault, the rocks have undergone a combination of irrotational fattening and dextral-oblique, normal-sense shear, with a bulk shear strain of similar to0.6 and kinematic vorticity number of similar to0.2. The vorticity analysis is compatible with estimates of strongly oblate bulk strain of similar to 75% maximum shortening. Dextral-reverse transpressional flow characterises higher strain S-tectonite mylonite within similar to1 km of the Alpine Fault. These relationships provide insight into the kinematics of flow and distribution of strain in the hangingwall of the Alpine Fault and place constraints on numerical mechanical models for the exhumation of these mid-crustal rocks. (C) 2001 Elsevier Science Ltd. All rights reserved.

The application of electron microscopy and microanalysis in conjunction with organic petrology to further the understanding of organic-mineral association: examples from Mount Isa and McArthur Basins, Australia

Organic petrology supported by electron microscopical and micro-analytical techniques was applied to organic matter in Proterozoic sediments to better understand hydrothermal processes responsible for ore-grade mineralisation. It was shown that organic maturation was not only closely linked to the geological history of the sediments, but also highlighted heat transfer by convection as differentiated from conduction solely through sediment burial and step-wise subsidence. Water-rock ratios effect organic maturation in hydrothermal systems, and erratic reflectance profiles are indicators of convective heat transfer. Identification and characterisation of organic materials in terms of source rock and migrated hydrocarbons was shown to be a powerful tool in reconstructing the thermal history of sediments, identifying hydrothermal episodes, fluid pathways and heat source in the northern Australian Proterozoic basins. Higher reflectance of organic matter towards the central parts of the Mount Isa Basin and some of the most northerly parts point to proximity to higher heat flow at times, in contrast to relatively low temperatures prevailing in the western parts of the basin, next to the Murphy Inlier. A close correlation shown between peak organic reflectance values and super-sequence boundaries farther highlighted the valuable information to be gained from organic petrology, by allowing the separation of processes responsible for metal dissolution and transportation from those inducing precipitation. (C) 2001 Elsevier Science B.V All rights reserved.

The structure of middle management remuneration packages: An application to Australian mine managers

This paper investigates the composition of remuneration packages for middle managers and relates the structure of remuneration contracts to firm-specific attributes. A statutorily defined position in a single industry is studied as an example of middle management. This allows us to control for differences in task complexity across managers and industry-induced factors that could determine differences in remuneration contracts. Higher-risk firms are expected to pay their mine managers a greater proportion of variable salaries and market and/or accounting-based compensation than low-risk firms. Results indicate that high-risk firms pay a higher proportion of variable salaries and more compensation based on market and/or accounting performance.