On A Maximum Likelihood Method for Clustering

Behaviormetrika - In this paper, we propose a new clustering method based on the concept of maximum likelihood (ML) estimation. In general, the problem of local minima arises when we try to use the...     

Relationship Between Local Atomic Arrangements and Properties of High-T c Superconductive Copper Oxides

Here various approaches to understanding the complex relationship among the local atomic arrangements, the distribution of charge/holes and the superconductivity characteristics are discussed for a variety of multi-layered copper-oxide superconductors. Regarding the structural parameters, i.e. bond lengths and angles, no single parameter is revealed to control the superconductivity characteristics alone. As a matter of fact, superconductivity properties are found to be correlated with the whole delicately balanced fine-structure that is described by a set of multiple parameters: it is shown that T _c and H_irr are enhanced in somewhat opposite ways. Expressing the delicately balanced fine-structure through the charge/hole distribution, it is concluded that the more homogeneously distributed the charge carriers are the more enhanced is the H_irr characteristics, while the opposite is likely to apply to the T _c value.     

Gas permeation property of polyaniline films

We examined the influence of polyaniline (PAn)'s unit sequence and doping with low molecular weight dopants or polymer dopants on permeation property. It was found that CO₂ permeability was increased by the formation of a quinonediimine unit in PAn with the oxidation. CO₂ sorption amount of PAn was decreased by oxidation. The increase of CO₂ permeability with oxidation, therefore, resulted from the increase of diffusivity, which was attributable to morphological variation by the increase of a quinonediimine unit. The permselectivity of PAn films was found to be remarkably improved by doping. In particular, the selectivity value of the PAn film doped with polyvinyl sulphonic acid as a polymer dopant went up to over 2,000. This remarkable increase of selectivity was found to result in the increase of selectivity, depending on diffusivity. It was also found that the permselectivity of the PAn film doped by polymer dopants was surpassed, as compared with that doped by low molecular dopants. © 1995 John Wiley & Sons, Inc.     

Effects of support and additive on oxidation state and activity of Pt catalyst in propane combustion

The effects of support and additive on the oxidation state and catalytic activity of Pt catalyst in the low temperature propane combustion were systematically investigated on Pt/MgO, Pt/Al₂O₃ and Pt/SiO₂–Al₂O₃. The catalytic activity varied much with both support materials and additives. The catalyst on the more acidic support showed higher activity, and the catalytic activity on every support materials increased as the electronegativity of additives increased, while some additives decreased the activity. The oxidation state of platinum, estimated by white line intensity of Pt L_III-edge XANES spectrum, also varied with the support and additives, and additives with higher electronegativity greatly prevented the platinum from its oxidation under oxidising atmosphere. Among almost all the catalysts with various supports and various additives, a clear relationship was observed between the oxidation state of platinum and the catalytic activity; the more metallic platinum showed higher activity. Thus, it was concluded that the total electrophilic/electrophobic property derived from those of the support and additive controls the oxidation state of platinum, which intensively affects the catalytic activity; i.e. higher electrophilic property provides less oxidised platinum, resulting in high catalytic activity. The mechanism of this effect was also discussed on the basis of thermochemical data, and it was proposed that the electrophobic materials promote the noble metal oxidation since the noble metal oxo-anion such as PtO^δ− is more stabilised with electrophobic cation.     

Structural significance of the alkyl substituent at the C-4 of (+)-trans-verbenyl acetate in sex pheromonal activity of the American cockroach

The significance of the alkyl group at the C-4 of (+)-trans-verbenyl acetate, which is the sex pheromone mimic of the American cockroach, was investigated. Seven alcohols possessing an ethyl, propyl, or dimethyl group at this position of the 6,6-dimethylbicyclo[3.1.1]heptane skeleton were synthesized and evaluated by behavioral assay. All of the alcohols were inactive, while three of four acetates of the 2-alcohols induced sexual behavior in male cockroaches at the 0.02 or 0.5 mg dosage level, either of which is many orders of magnitude higher than the threshold level of the natural sex pheromones (10^–8 mg). Among the acetates, the compounds with a methyl group or an -oriented ethyl group at C-4 showed the highest activity. The results are discussed in terms of spatial requirements of the molecules for interactions with the receptor.Studies on the sex pheromone mimic of the American cockroach, (+)-trans-verbenyl acetate. Part VIII. For Part VII, seeComp. Biochem. Physiol.,70A: 229–234 (1981).     

Fibrous support for immobilization of enzymes

A poly(vinyl alcohol) (PVA) fiber support incorporating various aminoacetal functional groups has been developed for immobilizing enzymes. The aminated PVA fiber seems to adsorb enzymes with electrostatic force of attraction; thus the immobilization procedure is simple. By the use of this fiber having immobilized enzymes, the reaction between enzymes and substrates is nearly independent of the size of subtstrates. This newly developed type of fiber, which is formed by a mass superfine fibers (SFF), each measuring 1 μm or less in diameter, permits much more increased surface area than the conventional enzyme immobilization supports. Our studies of the properties of the fiber for immobilization of enzymes show the following results: (1) SFF has a greater ability for the immobilization of invertase than ordinary fibers; (2) dimethyl-aminated SFF has the best performance for the immobilization of invertase. From these results, it is concluded that the dimethyl-aminated SFF is an excellent support for the immobilization of invertase.     

Damage to seagrass and seaweed beds in Matsushima Bay,Japan, caused by the huge tsunami of the Great East Japan Earthquake on 11 March 2011

Damage to seagrass and seaweed beds caused by the huge tsunami of the Great East Japan Earthquake on 11 March 2011 was investigated in Matsushima Bay, Miyagi Prefecture, Japan, by comparing satellite images taken before the tsunami in November 2009 and after the tsunami in May 2011. The results showed that the tsunami destroyed 220 ha of seagrass and seaweed beds, from 320 ha in 2009 to 100 ha just after the tsunami. Zostera marina beds on the sandy and muddy bottom were rather severely damaged compared with brown seaweed beds of Sargassum horneri on the rocky substrate. Substrates of the beds and directions of the tsunami seem to be responsible for the magnitude of the damage to the beds. Maps of the spatial distributions of the beds before and after the tsunami can provide information for the restoration of seagrass and seaweed beds as natural infrastructures of coastal fisheries resources.     

The analysis of an adaptive workload balancing strategy in computing system resources management

An adaptive workload balancing scheme is analyzed. This scheme is intended to automatically dissolve the imbalance in resource utilization of computing systems and to bring about the proper balance between throughput and turnaround time. This is done by means of the invisible hand due to the bidding of priorities of resource use, with no contradiction between the flexible bid for each resource and the stable price for the entire service. The result of this analysis shows that this scheme would regulate those imbalances, and that it includes the load balancing policy of IBM OS/VS2-2 as a special case. Some considerations and methods on an actual implementation are also presented.This paper is a revised version of material which appeared as Research Report RC5072 of IBM Thomas J. Watson Research Center, Yorktown Heights, New York.On leave from the University of Electro-Communications, Chofu, Tokyo, Japan.     

Performance evaluation of fuzzy rule-based classification systems obtained by multi-objective genetic algorithms

In this paper, we examine the classification performance of fuzzy if-then rules selected by a GA-based multi-objective rule selection method. This rule selection method can be applied to high-dimensional pattern classification problems with many continuous attributes by restricting the number of antecedent conditions of each candidate fuzzy if-then rule. As candidate rules, we only use fuzzy if-then rules with a small number of antecedent conditions. Thus it is easy for human users to understand each rule selected by our method. Our rule selection method has two objectives: to minimize the number of selected fuzzy if-then rules and to maximize the number of correctly classified patterns. In our multi-objective fuzzy rule selection problem, there exist several solutions (i.e., several rule sets) called “non-dominated solutions” because two conflicting objectives are considered. In this paper, we examine the performance of our GA-based rule selection method by computer simulations on a real-world pattern classification problem with many continuous attributes. First we examine the classification performance of our method for training patterns by computer simulations. Next we examine the generalization ability for test patterns. We show that a fuzzy rule-based classification system with an appropriate number of rules has high generalization ability.     

Dm-KDE: dynamical kernel density estimation by sequences of KDE estimators with fixed number of components over data streams

In many data stream mining applications, traditional density estimation methods such as kernel density estimation, reduced set density estimation can not be applied to the density estimation of data streams because of their high computational burden, processing time and intensive memory allocation requirement. In order to reduce the time and space complexity, a novel density estimation method Dm-KDE over data streams based on the proposed algorithm m-KDE which can be used to design a KDE estimator with the fixed number of kernel components for a dataset is proposed. In this method, Dm-KDE sequence entries are created by algorithm m-KDE instead of all kernels obtained from other density estimation methods. In order to further reduce the storage space, Dm-KDE sequence entries can be merged by calculating their KL divergences. Finally, the probability density functions over arbitrary time or entire time can be estimated through the obtained estimation model. In contrast to the state-of-the-art algorithm SOMKE, the distinctive advantage of the proposed algorithm Dm-KDE exists in that it can achieve the same accuracy with much less fixed number of kernel components such that it is suitable for the scenarios where higher on-line computation about the kernel density estimation over data streams is required.We compare Dm-KDE with SOMKE and M-kernel in terms of density estimation accuracy and running time for various stationary datasets. We also apply Dm-KDE to evolving data streams. Experimental results illustrate the effectiveness of the proposed method.