Constrained cascade generalization of decision trees

While decision tree techniques have been widely used in classification applications, a shortcoming of many decision tree inducers is that they do not learn intermediate concepts, i.e., at each node, only one of the original features is involved in the branching decision. Combining other classification methods, which learn intermediate concepts, with decision tree inducers can produce more flexible decision boundaries that separate different classes, potentially improving classification accuracy. We propose a generic algorithm for cascade generalization of decision tree inducers with the maximum cascading depth as a parameter to constrain the degree of cascading. Cascading methods proposed in the past, i.e., loose coupling and tight coupling, are strictly special cases of this new algorithm. We have empirically evaluated the proposed algorithm using logistic regression and C4.5 as base inducers on 32 UCI data sets and found that neither loose coupling nor tight coupling is always the best cascading strategy and that the maximum cascading depth in the proposed algorithm can be tuned for better classification accuracy. We have also empirically compared the proposed algorithm and ensemble methods such as bagging and boosting and found that the proposed algorithm performs marginally better than bagging and boosting on the average.     

Thermal expansion studies on As-Sb-Se glasses

The results of measurement of thermal expansion coefficient from 50°C to the softening temperature on eight glass compositions of the As-Sb-Se system are presented. Measurements have been made on (As, Sb)₄₀Se₆₀ and As _x Sb₁₅Se_85−x glasses at heating rates of 1°C/min and 5°C/min. The composition and temperature dependence of the thermal expansion coefficient is discussed in the light of results of other chalcogenide glasses.     

Equivalence of classicality and separability based on P phase–space representation of symmetric multiqubit states

Classical and quantum world views differ in peculiar ways. Understanding decisive quantum features—for which no classical explanation exist—and their interrelations is of foundational interest. Moreover, recognizing non-classical features carries practical significance in information processing tasks as it offers insights as to why quantum protocols work better than their classical counterparts. We focus here on two celebrated notions of non-classicality viz., negativity of P phase–space representation and entanglement in symmetric multiqubit systems. We prove that they imply each other.     

PHYSIOLOGICAL AND MORPHOLOGICAL CHANGES IN IMMOBILIZED L. PLANTARUM NCIM 2084 CELLS DURING REPEATED BATCH FERMENTATION FOR PRODUCTION OF LACTIC ACID

L. Plantarum NCIM 2084 cells are homofermentative, essentially producing lactic acid as the main end product of glucose fermentation. Prolonged use through repeated batch fermentation with the cells of L. plantarum immobilized on chitosan treated polypropylene matrix showed a shift in the metabolic pathway from the homofermentative to heterofermentative, accompanied by morphological changes in immobilized cells from the normal rod shape to the coccoid shape. These changes appear to be related to a shift in the bacterial cell metabolism, resulting in a decrease in lactic acid yields.     

Removal of Pb(II) ions from aqueous solution by adsorption using bael leaves (Aegle marmelos)

Biosorption of Pb(II) on bael leaves (Aegle marmelos) was investigated for the removal of Pb(II) from aqueous solution using different doses of adsorbent, initial pH, and contact time. The maximum Pb loading capacity of the bael leaves was 104 mg g^?1 at 50 mg L^?1 initial Pb(II) concentration at pH 5.1. SEM and FT-IR studies indicated that the adsorption of Pb(II) occurs inside the wall of the hollow tubes present in the bael leaves and carboxylic acid, thioester and sulphonamide groups are involved in the process. The sorption process was best described by pseudo second order kinetics. Among Freundlich and Langmuir isotherms, the latter had a better fit with the experimental data. The activation energy E_a confirmed that the nature of adsorption was physisorption. Bael leaves can selectively remove Pb(II) in the presence of other metal ions. This was demonstrated by removing Pb from the effluent of exhausted batteries.     

Spectral distribution of photocurrent in poly(6-tert-butyl-3-methyl-3,4-dihydro-2H-1,3-benzoxazine)

The spectral distribution of photoconductivity in poly(6-tert-butyl-3-methyl-3,4-dihydro-2H-1,3-benzoxazine) was studied in the sandwich configuration with indium tin oxide and silver electrodes. The photocurrent was found to originate mainly from the bulk dissociation of excitons with lesser contribution from electrode-mediated exciton dissociation. The molecule was fluorescent, but in the presence of the electron acceptor molecule 2,4,6-trinitrophenol (TNP), the intensity was reduced. The reduction in fluorescence intensity corresponded to an increase in photocurrent. The molecule showed photoconductivity in the visible without any sensitizers, which makes it suitable for stable photorefractive composites.     

Chemical bath deposition of different phases of copper selenide thin films by controlling bath parameters

In the preparation of copper selenide thin films using chemical bath deposition (CBD) technique, it is observed that the pH of the final reacting mixture is the major factor controlling the composition of the film. Thin films of cubic Cu_2−xSe and tetragonal Cu₃Se₂, of band gaps 2.20 and 2.83 eV, respectively, have been prepared using the CBD technique by adjusting the bath parameters like pH, temperature and the ratio between copper and selenium atoms in the reaction bath. X-ray diffraction analysis is used as the major tool for identification of these phases. The results have been confirmed using XPS, ICP and absorption studies.     

Learning decision rules for pattern classification under a familyof probability measures

In this paper, uniformly consistent estimation (learnability) of decision rules for pattern classification under a family of probability measures is investigated. In particular, it is shown that uniform boundedness of the metric entropy of the class of decision rules is both necessary and sufficient for learnability under each of two conditions: (i) the family of probability measures is totally bounded, with respect to the total variation metric, and (ii) the family of probability measures contains an interior point, when equipped with the same metric. In particular, this shows that insofar as uniform consistency is concerned, when the family of distributions contains a total variation neighborhood, nothing is gained by this knowledge about the distribution. Then two sufficient conditions for learnability are presented. Specifically, it is shown that learnability with respect to each of a finite collection of families of probability measures implies learnability with respect to their union; also, learnability with respect to each of a finite number of measures implies learnability with respect to the convex hull of the corresponding families of uniformly absolutely continuous probability measures     

A controllability criterion for linear systems with unbounded operators

A method is presented for 1) determining whether a given linear system with an unbounded system operator is controllable, and 2) determining the set of reachable states in systems which are not controllable.