Supervised clustering of label ranking data using label preference information

This paper studies supervised clustering in the context of label ranking data. The goal is to partition the feature space into K clusters, such that they are compact in both the feature and label ranking space. This type of clustering has many potential applications. For example, in target marketing we might want to come up with K different offers or marketing strategies for our target audience. Thus, we aim at clustering the customers’ feature space into K clusters by leveraging the revealed or stated, potentially incomplete customer preferences over products, such that the preferences of customers within one cluster are more similar to each other than to those of customers in other clusters. We establish several baseline algorithms and propose two principled algorithms for supervised clustering. In the first baseline, the clusters are created in an unsupervised manner, followed by assigning a representative label ranking to each cluster. In the second baseline, the label ranking space is clustered first, followed by partitioning the feature space based on the central rankings. In the third baseline, clustering is applied on a new feature space consisting of both features and label rankings, followed by mapping back to the original feature and ranking space. The RankTree principled approach is based on a Ranking Tree algorithm previously proposed for label ranking prediction. Our modification starts with K random label rankings and iteratively splits the feature space to minimize the ranking loss, followed by re-calculation of the K rankings based on cluster assignments. The MM-PL approach is a multi-prototype supervised clustering algorithm based on the Plackett-Luce (PL) probabilistic ranking model. It represents each cluster with a union of Voronoi cells that are defined by a set of prototypes, and assign each cluster with a set of PL label scores that determine the cluster central ranking. Cluster membership and ranking prediction for a new instance are determined by cluster membership of its nearest prototype. The unknown cluster PL parameters and prototype positions are learned by minimizing the ranking loss, based on two variants of the expectation-maximization algorithm. Evaluation of the proposed algorithms was conducted on synthetic and real-life label ranking data by considering several measures of cluster goodness: (1) cluster compactness in feature space, (2) cluster compactness in label ranking space and (3) label ranking prediction loss. Experimental results demonstrate that the proposed MM-PL and RankTree models are superior to the baseline models. Further, MM-PL is has shown to be much better than other algorithms at handling situations with significant fraction of missing label preferences.     

Surface modification of macroporous poly(glycidyl methacrylate-co-ethylene glycol dimethacrylate) resins for improved Candida antarctica lipase B immobilization

Crosslinked macroporous hydrophilic poly(glycidyl methacrylate-co-ethylene glycol dimethacrylate) [abbreviated poly(GMA-co-EGDMA)] with identical chemical structure (60% of glycidyl methacrylate) but with varied average pore sizes (from 30 to 560 nm), specific surface areas (from 13.2 to 106.0 m²/g), specific volumes (from 0.755 to 1.191 cm³/g) and particle sizes (<100 μm–630 μm) were synthesized via suspension polymerization. Modifications of poly(GMA-co-EGDMA) with various diamines (1,2-diaminoethane, 1,4-diaminobutane, 1,6-diaminohexane and 1,8-diaminooctane), 2-fluoroethylamine, glutaraldehyde and cyanuric chloride were carried out. The influence of the interaction between Candida antarctica lipase B (Cal-B) and various carriers during immobilization on the loading and hydrolytic activity (hydrolysis of para-nitrophenyl acetate) of the immobilized Cal-B were studied. Immobilization of Cal-B was performed at different temperatures and pH values. Cal-B immobilized at 30 °C and pH 6.8 was leading to increased activities. Purely physical adsorption between enzyme and copolymer was observed on carriers in which amine or fluorine groups were introduced into the carrier structure by modification with various diamines or 2-fluoroethylamine. As a consequence enzyme loading and activity decreases. In contrary, modification of the poly(GMA-co-EGDMA) with glutaraldehyde and cyanuric chloride results in a covalent connection between enzyme and carrier. The obtained results show a significant increase in Cal-B activity. The influence of the amount of glutaraldehyde and cyanuric chloride used for modification was screened. Increasing the amount of glutaraldehyde or cyanuric chloride used for modification resulted in an increase of the enzyme loading. Consequently, higher amount of glutaraldehyde used led to a higher fraction of the enzyme molecules that are covalently connected on to the carrier. As the amount of glutaraldehyde or cyanuric chloride used for modifications increases, activity of immobilized C. antarctica lipase B primarily increases, showing the highest value for 0.66% and 0.050% w/w, respectively, and subsequently decreases. We could show that Cal-B immobilized on epoxy-containing copolymer modified with glutaraldehyde and cyanuric chloride performs higher activity than free enzyme powder.     

Enhancement of electrocatalytic properties of carbonized polyaniline nanoparticles upon a hydrothermal treatment in alkaline medium

The electrocatalytic activity of carbonized polyaniline nanostructures (Carb-nanoPANI) towards oxygen reduction reaction (ORR), estimated in 0.1 mol dm⁻³ KOH solution, was significantly improved upon a hydrothermal treatment in 1 mol dm⁻³ KOH solution. Namely, the onset of ORR was shifted by ∼70 mV to more positive potentials, and the number of electrons consumed per O₂ molecule was enhanced in comparison to the original material. The number of electrons involved in ORR depended on loading, and with a loading of 0.5 mg cm⁻², for the potentials lower than −0.5 V vs SCE, the number of electrons approached 4. For this material, high stability of electrochemical behavior and resistance to the poisoning by ethanol was evidenced by potentiodynamic cycling.     

Enrichment of CLA isomers by selective esterification with l-menthol using Candida rugosa lipase

Commercially available preparations of CLA are composed of almost equal amounts of 9-cis,11-trans (9c,11t)-CLA and 10-trans,12-cis (10t,12c)-CLA. Each isomer was fractionated and enriched, for availability as a food supplement, by a process comprising selective esterification with l-menthol by Candida rugosa lipase, distillation, and n-hexane extraction. The first selective esterification of CLA isomers was conducted with an equimolar amount of l-menthol of 30°C. The oil phase of the reaction mixture was fractionated into an l-menthyl ester fraction (9c,11t-CLA rich) and an FFA fraction (10t,12c-CLA rich) by distillation. The FFA fraction was esterified again with an equimolar amount of l-menthol to enrich 10t,12c-CLA. The 10t,12c-CLA preparation was obtained as the resulting FFA fraction by distillation. 10t,12c-CLA was enriched to 91% with 40% recovery. To enrich 9c,11t-CLA, the l-menthyl ester fraction in the first esterification was chemically hydrolyzed, and the resulting FFA were esterified again with an equimolar amount of l-menthol. The 9c, 11t-CLA preparation was obtained by chemical hydrolysis of the resulting l-methyl ester fraction, followed by n-hexane extraction. 9c,11t-CLA was enriched to 94% with 42% recovery. This effective process for purification of CLA isomers using l-methol is applicable to the production of food supplements.     

Automatic domain modeling for human–robot interaction

Intelligent Service Robotics - This paper introduces an approach to automatic domain modeling for human–robot interaction. The proposed approach is symbolic and intended for semantically...     

Fewer Defects in the Surface Slows the Hydrolysis Rate,Decreases the ROS Generation Potential,and Improves the Non‐ROS Antimicrobial Activity of MgO

Magnesium oxide (MgO) is recognised as exhibiting a contact‐based antibacterial activity. However, a comprehensive study of the impact of atomic‐scale surface features on MgO's antibacterial activity is lacking. In this study, the nature and abundance of the native surface defects on different MgO powders are thoroughly investigated. Their impacts on the hydrolysis kinetics, antibacterial activity against Escherichia coli (ATCC 47076), Staphylococcus epidermidis and Pseudomonas aeruginosa and the reactive oxygen species (ROS) generation potential are determined and explained. It is shown that a reduction in the abundance of low‐coordinated oxygen atoms on the surface of the MgO improves its resistance to both hydrolysis and antibacterial activity. The ROS generation potential, determined in‐situ using a fluorescence microplate assay and electron paramagnetic resonance spectroscopy, is not an inherent property of the studied MgO, rather it is a side product of hydrolysis (only for the most highly defected MgO particles) and/or a consequence of the MgO/bacteria interaction. The evaluation of the mutual correlations of the hydrolysis, the antibacterial activity and the ROS generation, with their origin in the surface defects' peculiarities, led to the conclusion that the acid/base reaction between the MgO surface and the bacterial wall contributes considerably to the MgO's antibacterial activity.     

The synthesis of single phase WC nanoparticles/C composite by solid state reaction involving nitrogen-rich carbonized polyaniline

Single phase tungsten carbide nanoparticles (WC-NPs), (mean particle diameter 5.4 nm), distributed over carbonized polyaniline (C-PANI) nanotubes/nanosheets were synthesized by a solid state reaction between WO₃ and nitrogen-rich carbonized polyaniline at 1000 °C in a reducing atmosphere. The resulting composite was characterized by X-ray diffractometry, electron microscopy, thermogravimetry in oxidizing and reduction atmospheres and elemental analysis. We suggested that the synthesis of WC as a single phase was facilitated by reactive C atoms with dangling bonds, formed upon nitrogen removal.