A Human–Computer Cooperation Fuzzy c‐Means Clustering with Interval‐Valued Weights

In this paper, a fuzzy c‐means clustering algorithm based on interval‐valued weights is proposed for improving clustering performance. In the proposed algorithm, the interval‐valued weights are first constructed by synergy of the ReliefF algorithm and the analytic hierarchy process (AHP) method, and then they are transformed into a constraint condition associating with each weight variable in the weighted clustering objective function. In the sequence, the weighted clustering objective function is solved by combining the Lagrange multiplier method with the gradient‐based iteration computation. In the whole process of algorithm iteration, a compulsion strategy with human–computer cooperation is adopted to ensure each weight variable satisfies interval constraint itself. Three well‐known data set are used to perform profound experiments. Experimental results clearly show that the proposed algorithm has better clustering performance than other the weighted fuzzy c‐means clustering algorithm.     

A fuzzy ensemble of parallel polynomial neural networks with information granules formed by fuzzy clustering

In this paper, we introduce a new category of fuzzy models called a fuzzy ensemble of parallel polynomial neural network (FEP²N²), which consist of a series of polynomial neural networks weighted by activation levels of information granules formed with the use of fuzzy clustering. The two underlying design mechanisms of the proposed networks rely on information granules resulting from the use of fuzzy C-means clustering (FCM) and take advantage of polynomial neural networks (PNNs).The resulting model comes in the form of parallel polynomial neural networks. In the design procedure, in order to estimate the optimal values of the coefficients of polynomial neural networks we use a weighted least square estimation algorithm. We incorporate various types of structures as the consequent part of the fuzzy model when using the learning algorithm. Among the diverse structures being available, we consider polynomial neural networks, which exhibit highly nonlinear characteristics when being viewed as local learning models.We use FCM to form information granules and to overcome the high dimensionality problem. We adopt PNNs to find the optimal local models, which can describe the relationship between the input variables and output variable within some local region of the input space.We show that the generalization capabilities as well as the approximation abilities of the proposed model are improved as a result of using polynomial neural networks. The performance of the network is quantified through experimentation in which we use a number of benchmarks already exploited within the realm of fuzzy or neurofuzzy modeling.     

Oligomer and Polymer Formation in Hexamethylcyclotrisiloxane (D<Subscript>3</Subscript>) – Hydrosilane Systems Under Catalysis by tris(pentafluorophenyl)borane

Summary Reactions of hexamethylcyclotrisiloxane, D₃, with 1,1,3,3-tetramethyldisiloxane, ^HMM^H, 1,1,1,3,3-pentamethyldisiloxane, ^HMM, phenyldimethylsilane and phenylmethylsilane catalyzed by tris(pentafluorophenyl)borane were studied. These reactions lead to ring opening of D₃ by the SiH reactant producing open chain oligomers with hydrosilane functionality at one or both chain ends. The reactivity of the hydrosilanes toward D₃ decreases in the series: PhMeSiH₂ > ^HMM^H > PhMe₂SiH > ^HMM. Competitive self-oligomerization of ^HMM^H and ^HMM also occurs. Primary products of these processes are able to enter into reactions with the SiH and D₃ reactants; some also undergo cyclization. Thus, consecutive and competitive processes lead to a series of various oligohomologues. Gas chromatography in conjunction with chemical ionization mass spectroscopy permitted identification of structure and determination of the basic directions of these oligomerization processes. Polysiloxanes of higher molecular weight may be also formed in some of these systems. The reactions, which occur in the systems studied, are rationalized on the basis of the mechanism involving the hydride transfer from silicon to trivalent boron. This includes the transient formation of tertiary trisilyloxonium borate which decomposes by the hydride transfer to one of the silicon atoms of the trisilyloxonium center. Footnote: This paper is dedicated to Professor Ian Manners in recognition of his significant contributions to the field of organometallic polymers.     

In vitro availability of minerals from oat products

The content of macroelements P, Mg and Ca and microelements Mn, Fe, Zn and Cu was determined in 10 commercially sold oat products made by different technological processes (dehulling, instantinising, extrusion, flaking). Phosphorus was the most prevalent of the macroelements (from 240.8 ± 2.2 to 845.5 ± 8.1 mg per 100 g), followed by magnesium (from 73.2 ± 0.7 to 271.9 ± 2.7 mg per 100 g) and calcium (from 30.69 ± 0.01 to 112.7 ± 0.3 mg per 100 g). The Ca/P ratio ranged from 1:5.3 in crushed oat to 1:8.2 in oat flakes. Regarding the microelements, manganese was present at the highest concentrations (from 2.62 ± 0.02 to 8.69 ± 0.01 mg per 100 g). The content of iron was similar and that of zinc not much lower, whereas the amount of copper was considerably smaller (from 0.23 ± 0.002 to 0.59 ± 0.002 mg per 100 g). The highest concentrations of mineral elements were found in instant oat bran flakes and the lowest in extruded oat and corn crisps containing 50% corn grouts. Samples of the products analysed were subjected to in vitro enzymatic digestion, simulating the digestive process occurring in the human alimentary tract. The supernatants thus obtained were analysed for their content of the previously determined mineral components; the percentage of minerals released from the products was calculated. The following sequence of mineral components released was observed: Cu (57.2–95.6%) > P (39.7–60.9%) > Ca (18.2–39.5%) > Mg (16.4–39.8%) > Mn (6.4–24.7%) > Fe (6.5–29%) > Zn (11–17.2%). The Ca/P ratio in the supernatant worsened from 1:8.2 in crushed oat to 1:23.3 in extruded oat and corn crisps. Crushed oat released the highest amounts of mineral elements during enzymatic hydrolysis, with oat grouts coming second. As regards the other products, it is difficult to establish their relative sequence in the release of minerals. © 2002 Society of Chemical Industry     

Synthesis and characterization of novel alkyd-silicone hyperbranched nanoresins with high solid contents

Hyperbranched alkyd-silicone nanoresins (ASiHBRs) with high solid content were synthesized by etherification reaction between a hyperbranched alkyd resin (HABR) and Z-6018 silicone. ASiHBRs were characterized by nuclear magnetic resonance (NMR), gel permeation chromatography (GPC) and several other techniques. NMR spectra show the presence of aromatic rings, -Si-O and -C-O; grafting was successful. The molar masses of ASiHBRs determined by GPC are higher than that of HABR. The hydroxyl values decrease with increasing silicone content. ASiHBRs have low viscosities what allows easier film formation. Hydrodynamic dimensions, refractive indices, glass transition temperatures, gloss and hardness of ASiHBRs increase with increasing silicone contents. Low contents of volatile organic compounds provide more environmentally friendly coatings.     

Bond strength of polymer lightweight aggregate concrete

Bond strength, physical, and mechanical properties of lightweight PC were investigated with inclusion of pumice lightweight aggregate in maximum size of 12 mm. As binder material, epoxy resin‐based polymer was used with its hardener. The binder to aggregate ratio was 30% by weight. In addition, steel fibers were added to lightweight PC mixtures in ratio of 0, 0.5, and 1%. After lightweight PC mixture was prepared, it was poured in the molds with different type of steel‐bars in size of 100 × 100 × 100 mm³. The steel‐bars centered in the cubic molds, and they were in size of Ø12, Ø14, and Ø16. The specimens were cured at 60°C for 2 h. On the hardened polymer lightweight concrete (PLC), pull‐out test for bond strength and compressive strength tests were performed. Moreover, ultrasonic pulse velocity, water absorption by weight, specific porosity, and density experiments were carried out. The relation between physical and mechanical properties showed that PLCs become more durable when using ratio of steel fibers. POLYM. COMPOS., 34:2125–2132, 2013. © 2013 Society of Plastics Engineers     

Parameterized complexity of basic decision problems for tree automata

There are many decision problems in automata theory (including membership, emptiness, inclusion and universality problems) that are NP-hard for some classes of tree automata (TA). The study of their parameterized complexity allows us to find new bounds of their nonpolynomial time algorithmic behaviours. We present results of such a study for classical TA, rigid tree automata, TA with global equality and disequality and t-DAG automata. As parameters we consider the number of states, the cardinality of the signature, the size of the term or the t-dag and the size of the automaton.     

SDF classifier revisited

This paper is addressing problems related to the construction of classifiers based on the Similarity Discriminant Function (SDF), in which the traditional vector representation of a pattern is replaced with matrix data. We introduce potential modifications of the matrix data structure and propose new variants of the SDF. The algorithms that we present were tested on images of handwritten digits and on photographs of human faces, taken from the ORL and CMU‐PIE databases. The results of experiments show that our modifications significantly improved the performance of the original SDF classifier.