Eigenclassifiers for combining correlated classifiers

In practice, classifiers in an ensemble are not independent. This paper is the continuation of our previous work on ensemble subset selection [A. Ula?, M. Semerci, O.T. Y?ld?z, E. Alpayd?n, Incremental construction of classifier and discriminant ensembles, Information Sciences, 179 (9) (2009) 1298–1318] and has two parts: first, we investigate the effect of four factors on correlation: (i) algorithms used for training, (ii) hyperparameters of the algorithms, (iii) resampled training sets, (iv) input feature subsets. Simulations using 14 classifiers on 38 data sets indicate that hyperparameters and overlapping training sets have higher effect on positive correlation than features and algorithms. Second, we propose postprocessing before fusing using principal component analysis (PCA) to form uncorrelated eigenclassifiers from a set of correlated experts. Combining the information from all classifiers may be better than subset selection where some base classifiers are pruned before combination, because using all allows redundancy.     

Cost-conscious comparison of supervised learning algorithms over multiple data sets

In the literature, there exist statistical tests to compare supervised learning algorithms on multiple data sets in terms of accuracy but they do not always generate an ordering. We propose Multi²Test, a generalization of our previous work, for ordering multiple learning algorithms on multiple data sets from “best” to “worst” where our goodness measure is composed of a prior cost term additional to generalization error. Our simulations show that Multi²Test generates orderings using pairwise tests on error and different types of cost using time and space complexity of the learning algorithms.     

Design of a framework for modeling, integration and simulation of physiological models

Multiscale modeling and integration of physiological models carry challenges due to the complex nature of physiological processes. High coupling within and among scales present a significant challenge in constructing and integrating multiscale physiological models. In order to deal with such challenges in a systematic way, there is a significant need for an information technology framework together with related analytical and computational tools that will facilitate integration of models and simulations of complex biological systems. Physiological Model Simulation, Integration and Modeling Framework (Phy-SIM) is an information technology framework providing the tools to facilitate development, integration and simulation of integrated models of human physiology. Phy-SIM brings software level solutions to the challenges raised by the complex nature of physiological systems. The aim of Phy-SIM, and this paper is to lay some foundation with the new approaches such as information flow and modular representation of the physiological models. The ultimate goal is to enhance the development of both the models and the integration approaches of multiscale physiological processes and thus this paper focuses on the design approaches that would achieve such a goal.     

Iterative performance improvement of fuzzy control systems for three tank systems

This paper suggests the performance improvement of fuzzy control systems (FCSs) for three tank systems using iterative feedback tuning (IFT). The stable design of Takagi–Sugeno–Kang fuzzy controllers is guaranteed by means of a stability theorem based on LaSalle’s global invariant set theorem formulated for a class of multi input-multi output (MIMO) nonlinear processes. An IFT algorithm characterized by setting the step size to guarantee the FCS stability is proposed. The theoretical approaches are applied in a case study that deals with the IFT-based stable design of fuzzy controllers dedicated to the level control of a cylindrical three tank system as a representative MIMO system. A set of experimental results for a laboratory setup illustrates the performance improvement.     

The ratio of crystallinity and thermodynamical interactions of polycaprolactone with some aliphatic esters and aromatic solvents by inverse gas chromatography

Summary The retention diagrams of benzene, toluene, ethyl benzene, chloro benzene, n-propyl benzene, isopropyl benzene, methyl acetate, ethyl acetate, n-propyl acetate, n-butyl acetate and isoamyl acetate on the polycaprolactone were plotted at temperatures between 70 and 140^oC by inverse gas chromatography technique. Percent crystallinity of polycaprolactone were obtained at temperatures below melting point from the retention diagrams of benzene, toluene and ethyl benzene. It was concluded that the data obtained by inverse gas chromatography were comparable those of obtained by differential scanning calorimetry. Specific retention volume, V_g^o, Flory-Huggins polymer-solvent interaction parameters, ₁₂, the weight fraction activity coefficients of the solvents at infinite dilution, ₁, effective exchange energy parameters, X_eff were determined. Later, the partial molar heat of sorption, H_1,sorp and the partial molar heat of mixing, H₁ were obtained from the slope of the logarithm of specific retention volume, Ln V_g^o versus 1/T plot and from the slope of the logarithm of the weight fraction activity coefficients, ₁versus 1/T plot, respectively.     

Robust estimation and hypothesis testing under short-tailedness and inliers

Estimation and hypothesis testing based on normal samples censored in the middle are developed and shown to be remarkably efficient and robust to symmetric shorttailed distributions and to inliers in a sample. This negates the perception that sample mean and variance are the best robust estimators in such situations (Tiku, 1980; Dunnett, 1982). Professor Emeritus, Department of Mathematics and Statistics, McMaster University, Professor Emeritus, Department of Mathematics and Statistics, McMaster University,     

Comparison of OR and AI methods in discrete manufacturing using fuzzy logic

Since 1950s the techniques of Operations Research (OR) and Optimization have been utilized to increase the efficiency of the production systems. With the widespread use of computers, it has even become easier to deal with industrial problems. However the complexity of the problems still reveals the difficulty in providing solutions. The use of artificial intelligence (AI) seems to attract the attention of the researcher to overcome to the difficulties. This has already been realized with several successful applications. In this study, the use of AI and OR techniques is compared using fuzzy logic. The progress of manufacturing systems, characteristics of production processes, system managements and system behavior are taken into account. The study is focussed on only discrete manufacturing.     

Experimental and theoretical study for the determination of thermal conductivity of porous building material made with pumice and tragacanth

In this study, porous solid materials were produced with mixture combinations of materials such as tragacanth, cement, and pumice aggregate. Thermal conductivities of the produced materials were determined using hot wire methodology. The thermal conductivities of the produced samples ranged from 0.433 to 0.177 W/mK, depending on pumice diameter, pumice, tragacanth, and cement rates. A new model was developed to determine the effective thermal conductivities of the produced samples. The thermal conductivities obtained by the measurements were compared to the ones predicted by the model. The predicted values deviated as 1–31% from the measured values. It was finally shown that the experimental results were in good agreement to the predicted results.     

Biosorption of Acid Red P-2BX by lichens as low-cost biosorbents

The aim of this study is to examine the dye biosorption properties of lichen species called Cladonia convoluta and Evernia prunastri. Since lichens are extensively found in the environment, their suitability as a cheap adsorbent has been investigated in this study. The optimal parameters for textile dye biosorption were also determined. The dried lichen biomass showed better dye biosorption capacity than ash lichen biomass. C. convoluta had better dye biosorption capacity than E. prunastri. Dye biosorption rate was found as 71.41% at optimal conditions. This study concluded that C. convoluta was a successful and cheap biosorbent for treatment of water contaminated by Acid Red P-2BX dye.