HColonies: a new hybrid metaheuristic for medical data classification

Medical data feature a number of characteristics that make their classification a complex task. Yet, the societal significance of the subject and the computational challenge it presents has caused the classification of medical datasets to be a popular research area. A new hybrid metaheuristic is presented for the classification task of medical datasets. The hybrid ant–bee colonies (HColonies) consists of two phases: an ant colony optimization (ACO) phase and an artificial bee colony (ABC) phase. The food sources of ABC are initialized into decision lists, constructed during the ACO phase using different subsets of the training data. The task of the ABC is to optimize the obtained decision lists. New variants of the ABC operators are proposed to suit the classification task. Results on a number of benchmark, real-world medical datasets show the usefulness of the proposed approach. Classification models obtained feature good predictive accuracy and relatively small model size.     

Ignition and combustion of pyrotechnic compositions based on micro- and nanoparticles of aluminum diboride in air flow in a two-zone combustion chamber

This paper presents the experimental study of ignition and combustion of micro- and nanoparticles of aluminum diboride as part of pyrotechnic energy-saturated compositions in a gas generator with an air afterburner chamber and the thermodynamic calculations of combustion of pyrotechnic compositions based on aluminum diboride in air. The discharge of the combustion products from the afterburner chamber is recorded on video. It is shown that replacing micron aluminum diboride by powdered diboride with a mass-average diameter of particles equal to ≈270 nm in the pyrotechnic composition and increasing the pressure in the afterburner chamber cause the combustion efficiency in air to increase by 5–20%.     

Effects of gamma radiation on the progeny of irradiated Ephestia cautella (Walker) (Lepidoptera: Pyralidae) males

One-day-old P₁ males of Ephestia cautella, Baghdad strain, were treated with 0.15, 0.20 and 0.25 kGy of γ radiation and mated in groups (15 pairs) with untreated virgin females. Radiation effects were assayed by determining percentage of egg hatch, mating ability and F₁ sex ratio. Our results showed the greater the initial dose, the greater the reduction in the percent egg hatch, and the greater the male-to-female ratio. Radiation effects on F₁ males were greater than the effect on the P₁ males. Furthermore, the results showed that the effects on F₁ males were greater than the effects on F₁ females. These results were confirmed cytologically by examination of the developing meiotic germ cells of F₁ males which showed that the meiotic nuclei carried multiple translocations. Such chromosomal aberrations explain the sterility observed in the F₁ male generation.     

Evolutionary computation approaches to the Curriculum Sequencing problem

Within the field of e-Learning, a learning path represents a match between a learner profile and his preferences from one side, and the learning content presentation and the pedagogical requirements from the other side. The Curriculum Sequencing problem (CS) concerns the dynamic generation of a personal optimal learning path for a learner. This problem has gained an increased research interest in the last decade, as it is not possible to have a single learning path that suits every learner in the widely heterogeneous e-Learning environment. Since this problem is NP-hard, heuristics and meta-heuristics are usually used to approximate its solutions, in particular Evolutionary Computation approaches (EC). In this paper, a review of recent developments in the application of EC approaches to the CS problem is presented. A classification of these approaches is provided with emphasis on the tools necessary for facilitating learning content reusability and automated sequencing.     

Effect of titanium dioxide nanoparticles on polydimethylsiloxane/polyethersulfone composite membranes for gas separation

Introducing inorganic nanoparticles into the structure of polymeric membranes is an interesting approach for the enhancement of physical, chemical, and separation properties of the membranes. In this article, the performance of a two‐layer nanocomposite membrane for gas separation was studied. Three different methods for embedding titanium dioxide (TiO₂) nanoparticle were employed for the membrane preparation. The techniques include blending TiO₂ in the polydimethylsiloxane (PDMS) coating layer, blending TiO₂ in the polyethersulfone (PES) support and dip coating of PES support with TiO₂ accompanied by PDMS coating. The aim of the current research was finding the optimum technique for introducing TiO₂ into the membrane to obtain superior performance for gas separation. The results indicated that PES support containing TiO₂ nanoparticles possessed favorable effect on gas separation capability. The optimum performance was obtained by PDMS‐coated membranes prepared with 7 wt% TiO₂‐embedded PES support. Carbon dioxide (CO₂) permeance, CO₂/nitrogen, and CO₂/methane selectivity were obtained as 188.7 GPU, 8.6, and 3.4, respectively. POLYM. ENG. SCI., 2012. © 2012 Society of Plastics Engineers     

IWD-Miner: A Novel Metaheuristic Algorithm for Medical Data Classification

Medical data classification (MDC) refers to the application of classification methods on medical datasets. This work focuses on applying a classification task to medical datasets related to specific diseases in order to predict the associated diagnosis or prognosis. To gain experts’ trust, the prediction and the reasoning behind it are equally important. Accordingly, we confine our research to learn rule-based models because they are transparent and comprehensible. One approach to MDC involves the use of metaheuristic (MH) algorithms. Here we report on the development and testing of a novel MH algorithm: IWD-Miner. This algorithm can be viewed as a fusion of Intelligent Water Drops (IWDs) and AntMiner+. It was subjected to a four-stage sensitivity analysis to optimize its performance. For this purpose, 21 publicly available medical datasets were used from the Machine Learning Repository at the University of California Irvine. Interestingly, there were only limited differences in performance between IWD-Miner variants which is suggestive of its robustness. Finally, using the same 21 datasets, we compared the performance of the optimized IWD-Miner against two extant algorithms, AntMiner+ and J48. The experiments showed that both rival algorithms are considered comparable in the effectiveness to IWD-Miner, as confirmed by the Wilcoxon nonparametric statistical test. Results suggest that IWD-Miner is more efficient than AntMiner+ as measured by the average number of fitness evaluations to a solution (1,386,621.30 vs. 2,827,283.88 fitness evaluations, respectively). J48 exhibited higher accuracy on average than IWD-Miner (79.58 vs. 73.65, respectively) but produced larger models (32.82 leaves vs. 8.38 terms, respectively).     

Optical and structural features of silicon-rich hydrogenated amorphous silicon nitride thin films

A systematic study of fabrication and effect of post-deposition processing on the optical and structural features of silicon-rich hydrogenated amorphous silicon nitride thin films deposited by Hg-sensitized Photo-Chemical Vapour Deposition technique is presented. Both deposition parameters and post-deposition thermal treatment resulted into substantial change in the refractive index associated with the densification of the film. Our studies reveal that the presence of hydrogen and its out-diffusion upon thermal treatment play a crucial role in the overall structural evolution, specially the stabilization of individual phases such as Si and Si3N4. We further report the room-temperature photoluminescence from as-deposited films, which is due to formation of silicon nanostructures in crystalline and amorphous forms. These studies are of great interest from the prospective of commercially viable Si-based technology.     

Growth of long aligned carbon nanotubes on amorphous hydrogenated silicon nitride by thermal chemical vapor deposition

Vertically aligned long carbon nanotubes in the range of 80-100 µm have been synthesized on amorphous hydrogenated silicon nitride (a-SiN_x:H) coated silicon substrate by thermal chemical vapor deposition of ferrocene and xylene. It is observed that high temperature annealing in oxygen ambient results in formation of crystalline silicon dioxide in the matrix of amorphous silicon nitride due to out diffusion of hydrogen. It is suggested that active sites created on silicon dioxide and a-SiN_x:H clusters provide mechanical support for the alignment of long carbon nanotubes. It is proposed that a thin layer of a-SiN_x:H prevents silicide formation between the catalyst (Fe) and silicon thus lengthening the catalyst life.