Extracting rules for classification problems: AIS based approach

Although Artificial Neural Network (ANN) usually reaches high classification accuracy, the obtained results in most cases may be incomprehensible. This fact is causing a serious problem in data mining applications. The rules that are derived from ANN are needed to be formed to solve this problem and various methods have been improved to extract these rules. In our previous work, a hybrid neural network was presented for classification (Kahramanli & Allahverdi, 2008). In this study a method that uses Artificial Immune Systems (AIS) algorithm has been presented to extract rules from trained hybrid neural network. The data were obtained from the University of California at Irvine (UCI) machine learning repository. The datasets are Cleveland heart disease and Hepatitis data. The proposed method achieved accuracy values 96.4% and 96.8% for Cleveland heart disease dataset and Hepatitis dataset respectively. It is been observed that these results are one of the best results comparing with results obtained from related previous studies and reported in UCI web sites.     

Parallel inference in dynamic decision support systems of a semiotic type

The organization of parallel inference in dynamic decision support systems (DDSS) of a semiotic type, oriented towards a solving of ill-formed problems in dynamic applied domains, is considered. As a knowledge representation model, there are used production rules reflecting expert knowledge about a problem domain, an environment and decision making processes. The main concepts and assertions defining possibility and impossibility of parallel executing the production rules are given. Several types of parallelism in an inference process are introduced. The corresponding algorithm of parallel inference is described. Thus, the purpose of this paper is to develop and to research parallel inference methods and procedures that provide efficient processing a large amount of production rules for DDSS of a semiotic type.     

Landscaper: A Modeling System for 3D Printing Scale Models of Landscapes

Landscape models of geospatial regions provide an intuitive mechanism for exploring complex geospatial information. However, the methods currently used to create these scale models require a large amount of resources, which restricts the availability of these models to a limited number of popular public places, such as museums and airports. In this paper, we have proposed a system for creating these physical models using an affordable 3D printer in order to make the creation of these models more widely accessible. Our system retrieves GIS relevant to creating a physical model of a geospatial region and then addresses the two major limitations of affordable 3D printers, namely the limited number of materials and available printing volume. This is accomplished by separating features into distinct extruded layers and splitting large models into smaller pieces, allowing us to employ different methods for the visualization of different geospatial features, like vegetation and residential areas, in a 3D printing context. We confirm the functionality of our system by printing two large physical models of relatively complex landscape regions.     

The two-stage assembly flowshop scheduling problem with bicriteria of makespan and mean completion time

In this paper, we address the two-stage assembly flowshop scheduling problem with a weighted sum of makespan and mean completion time criteria, known as bicriteria. Since the problem is NP-hard, we propose heuristics to solve the problem. Specifically, we propose three heuristics; simulated annealing (SA), ant colony optimization (ACO), and self-adaptive differential evolution (SDE). We have conducted computational experiments to compare the performance of the proposed heuristics. It is statistically shown that both SA and SDE perform better than ACO. Moreover, the experiments reveal that SA, in general, performs better than SDE, while SA consumes less CPU time than both SDE and ACO. Therefore, SA is shown to be the best heuristic for the problem.     

Investigating photocatalytic property of TiO2 nanorods fabricated hydrothermally at equal molar ratio of KOH and NaOH

TiO₂ photocatalysts were prepared by hydrothermal method at temperatures of 110, 155, and 200°C for 24 hours, using KOH (5 mol/L):NaOH (5 mol/L) solution and anatase TiO₂ powder with an average particle size of 0.13 μm as precursor and by calcining at two different temperatures of 450 and 700°C. An overall photocatalytic removal efficiency of 93% was achieved for Rhodamine B dye under UV light irradiation by the photocatalyst obtained at synthesis and calcination temperatures of 155 and 700°C, respectively, which indicated a nanorod morphology with a mean diameter of about 34 nm.     

Zinc incorporation into CdTe quantum dots in glass

We report zinc incorporation into CdTe nanoparticles grown by two-step solid phase precipitation in commercial borosilicate glass quenched from the melt, based on a co-evaluation of the results of resonant Raman and optical absorption measurements. Resonant Raman spectra display a two-peak structure at wave-number positions corresponding to ternary compound Zn_xCd_1?xTe. We attribute the higher intensity peak between 190 and 195 cm^?1 to the first harmonic of the zone-center longitudinal optical mode (LO₁) and, the lower intensity peak between 157 and 160 cm^?1 to the second harmonic (LO₂) of Zn_xCd_1?xTe crystal. The wave-number of vibrational Raman modes indicates that zinc content varies between 39 and 50% during the growth of quantum dots. The asymptotic absorption edge against heat-treatment time plot extrapolates to a bulk band gap of 1.714 eV which sets a lower limit of 31% for zinc incorporated into quantum dots which is consistent with the results of resonant Raman measurements. The energetic position of asymptotic absorption edge of 1.592 eV and an additional unresolved weak structure in Raman spectrum between 166 and 182 cm^?1 observed for as-received glass might serve as a evidence for the occurrence of a different nanocrystalline phase with 13% zinc content.     

CdS<Subscript><Emphasis Type="Italic">x</Emphasis></Subscript>Se<Subscript>1−<Emphasis Type="Italic">x</Emphasis></Subscript> quantum dots studied through optical absorption,steady-state photoluminescence,and resonant Raman spectroscopies

CdS _x Se_1−x nanoparticles or quantum dots (QDs) were grown in borosilicate glass by a two-step heat-treatment process from a melt-quenched color filter glass. We incorporate the results of optical absorption, steady-state photoluminescence (PL), and resonant Raman spectroscopies in the study of growth kinetics of CdS _x Se_1−x QDs. A modeling of PL spectra employing two Gaussian emission bands and a quantized-state effective mass model in the strong confinement regime reveals that (i) the average particle size ranges from 1.7 to 8.5 nm, (ii) the size dispersion narrows down to 0.22 nm for a single sample, and (iii) QDs form by diffusion-limited growth. We presume that size-dependent higher energy PL band close to the asymptotic absorption edge is due to surface-assisted electron-hole recombination since the difference between optical absorption and PL bands decreases from 239 to 122 meV with increasing average radius.     

Development and Electromechanical Properties of Multimaterial Piezoelectric and Electrostrictive PMN-PT Monomorph Actuators

Monolithic multimaterial monomorphs, comprised of varying ratios of piezoelectric 0.65Pb(Mg_1/3 Nb_2/3)O₃-0.35PbTiO₃ to electrostrictive 0.90Pb(Mg_1/3Nb_2/3)O₃-0.10PbTiO₃, have been co-fired at 1150^∘C. The relative permittivity, displacement, and polarization hysteresis were investigated for varying ratios of piezoelectric to electrostrictive material. The permittivity of the 1:1 multimaterial monomorphs followed the dielectric mixing laws, showing a dielectric constant of 5,500 at room temperature. The P-E hysteresis loop of the 1:1 sample exhibited a maximum and remnant polarization slightly less than the piezoelectric PMN-PT 65/35, but higher than the electrostrictive PMN-PT 90/10. Displacement was found to be higher for the 3:1 monolithic monomorph actuators, reaching 76 μ m at 6 kV/cm. The results indicate that by minimizing the electrostrictive layer thickness the tip displacement can be substantially increased while maintaining a lower hysteresis than that of the purely piezoelectric counterpart.