Boosting k-nearest neighbor classifier by means of input space projection

The k-nearest neighbors classifier is one of the most widely used methods of classification due to several interesting features, such as good generalization and easy implementation. Although simple, it is usually able to match, and even beat, more sophisticated and complex methods. However, no successful method has been reported so far to apply boosting to k-NN. As boosting methods have proved very effective in improving the generalization capabilities of many classification algorithms, proposing an appropriate application of boosting to k-nearest neighbors is of great interest.Ensemble methods rely on the instability of the classifiers to improve their performance, as k-NN is fairly stable with respect to resampling, these methods fail in their attempt to improve the performance of k-NN classifier. On the other hand, k-NN is very sensitive to input selection. In this way, ensembles based on subspace methods are able to improve the performance of single k-NN classifiers. In this paper we make use of the sensitivity of k-NN to input space for developing two methods for boosting k-NN. The two approaches modify the view of the data that each classifier receives so that the accurate classification of difficult instances is favored.The two approaches are compared with the classifier alone and bagging and random subspace methods with a marked and significant improvement of the generalization error. The comparison is performed using a large test set of 45 problems from the UCI Machine Learning Repository. A further study on noise tolerance shows that the proposed methods are less affected by class label noise than the standard methods.     

Divided Attention in Younger and Older Adults: Effects of Strategy and Relatedness on Memory Performance and Secondary Task Costs.

Divided attention at encoding leads to a significant decline in memory performance, whereas divided attention during retrieval has relatively little effect; nevertheless, retrieval carries significant secondary task costs, especially for older adults. The authors further investigated the effects of divided attention in younger and older adults by using a cued-recall task and by measuring retrieval accuracy, retrieval latency, and the temporal distribution of attentional costs at encoding and retrieval. An age-related memory deficit was reduced by pair relatedness, whereas strategy instructions benefited both age groups equally. Attentional costs were greater for retrieval than for encoding, especially for older adults. These findings are interpreted in light of notions of an age-related associative deficit (M. Naveh-Benjamin, 2000) and age-related differences in the use of self-initiated activities and environmental support (F. I. M. Craik, 1983, 1986). (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Electrically insulating polymeric nanocomposites with enhanced thermal conductivity by visible‐light curing of epoxy–boron nitride nanotube formulations

Epoxy–boron nitride nanotube (BNNT) composites were prepared using visible light through a radical‐induced cationic polymerization method activated by camphorquinone. The fully cured films showed an enhancement of glass transition temperature in the presence of the filler. Electrical characterization showed a slight dielectric constant decrease with BNNT content. Finally, thermal conductivity measured using nano‐flash analysis showed a linear increase in the thermal conductivity of the materials with increasing BNNT content in the photocurable formulations. © 2017 Society of Chemical Industry     

Mechanical properties of freely suspended semiconducting graphene-like layers based on MoS2

We fabricate freely suspended nanosheets of molybdenum disulphide (MoS₂) which are characterized by quantitative optical microscopy and high-resolution friction force microscopy. We study the elastic deformation of freely suspended nanosheets of MoS₂ using an atomic force microscope. The Young''s modulus and the initial pre-tension of the nanosheets are determined by performing a nanoscopic version of a bending test experiment. MoS₂ sheets show high elasticity and an extremely high Young''s modulus (0.30 TPa, 50% larger than steel). These results make them a potential alternative to graphene in applications requiring flexible semiconductor materials.PACS, 73.61.Le, other inorganic semiconductors, 68.65.Ac, multilayers, 62.20.de, elastic moduli, 81.40.Jj, elasticity and anelasticity, stress-strain relations.     

Lung Cancer Assistant: a hybrid clinical decision support application for lung cancer care

Multidisciplinary team (MDT) meetings are becoming the model of care for cancer patients worldwide. While MDTs have improved the quality of cancer care, the meetings impose substantial time pressure on the members, who generally attend several such MDTs. We describe Lung Cancer Assistant (LCA), a clinical decision support (CDS) prototype designed to assist the experts in the treatment selection decisions in the lung cancer MDTs. A novel feature of LCA is its ability to provide rule-based and probabilistic decision support within a single platform. The guideline-based CDS is based on clinical guideline rules, while the probabilistic CDS is based on a Bayesian network trained on the English Lung Cancer Audit Database (LUCADA). We assess rule-based and probabilistic recommendations based on their concordances with the treatments recorded in LUCADA. Our results reveal that the guideline rule-based recommendations perform well in simulating the recorded treatments with exact and partial concordance rates of 0.57 and 0.79, respectively. On the other hand, the exact and partial concordance rates achieved with probabilistic results are relatively poorer with 0.27 and 0.76. However, probabilistic decision support fulfils a complementary role in providing accurate survival estimations. Compared to recorded treatments, both CDS approaches promote higher resection rates and multimodality treatments.     

Increasing espresso coffee brew antioxidant capacity using phenolic extract recovered from hazelnut skin waste

The simultaneous consumption of different classes of phytochemical antioxidants in the diet can result in more beneficial effects than when consumed alone. In the present study, the in vitro and in vivo antioxidative effects of espresso coffee brew (EC) (rich in chlorogenic acids) with added crude hazelnut skin phenolic extract (HSPE) from hazelnut skin waste (rich in flavonoids) were studied. Both post-brewing and pre-brewing phenolic-enriched espresso coffees (PE-ECs) were analysed for total phenols and screened for their in vitro antiradical ability. Moreover, the in vivo biological effect on the antioxidant potential of plasma in rats was evaluated. The PE-ECs showed increased both in vitro and in vivo antiradical activity proportional to the added HSPE. The in vivo experiments suggested that HSPE was much more antioxidant active than the phenolic fraction naturally contained in EC. Moreover, evidence of possible synergic effects of EC and HSPE phenolics was observed in vivo.     

Effect of different treatments on the microstructure and functional and pasting properties of pigeon pea (Cajanus cajan L.), dolichos bean (Dolichos lablab L.) and jack bean (Canavalia ensiformis) flours from the north‐east Argentina

The effect of germination (G; 5 days), soaking‐cooking (SC; 6 h–20 min, 6 h–40 min, 6 h–60 min) and microwave (M; 50%, 70%, 100%) treatments on pigeon pea (PP), dolichos bean (DB) and jack bean (JB) seeds was studied. Microstructure of seeds and functional (protein solubility, water‐holding capacity, oil‐holding capacity, emulsion stability) and pasting properties of flours were determined. Germination and microwave treatments modified the protein matrix of cotyledon cells preserving the shape of the starch granule, whereas the SC treatment (6 h–60 min) affected both. The soaking‐cooking is the most influential treatment on the functional properties of PP, DB and JB flours, as increased water absorption capacity (73–96%), decreased protein solubility (>80%) and the tendency to retrogradation of amylose (69–85%) also improved emulsion stability.     

Identification of unbalance forces by metaheuristic search algorithms

This paper discusses the identification of parameters in rotary systems, namely, the unbalance magnitude, phase and position in the rotor system. These parameters can be identified using the measured orbits in the hydrodynamic bearings. The oil film forces are evaluated in the different positions of the orbit of the journal and are applied to the model of the shaft. The model, integrated in time domain, allows with an assumed unbalance, to simulate the orbits. The objective function is basically the difference between measured and simulated orbits, and its minimum corresponds to the identified unbalance amount, phase and position along the shaft. With respect to traditional model based identification procedures, this approach using oil film forces instead of oil film linearized stiffness and damping coefficients, and unfiltered orbits instead of 1X vibration components is suitable to deal with non-linear behaviour of the system.     

Study of shrinkage porosity in spheroidal graphite cast iron

This study aims at identifying the relationship between the shrinkage cavities and the solidification structure in spheroidal graphite cast iron. Cast samples specially designed to contain shrinkage cavities were used. The solidification macrostructure was revealed using the Direct Austempering After Solidification method, while the solidification microstructure was revealed by using colour etching. At the midsection of the pieces, the shrinkage cavities and the solidification structure were observed jointly. The study showed that the classification of shrinkage porosity found in literature does not correspond to the ductile iron solidification model recognized by most of the scientific community. Early solidification models, and therefore shrinkage formation mechanisms, were proposed in instances when there was not a thorough knowledge of the morphology of the solid phases during solidification. Nowadays, defects formation mechanisms can be described with higher accuracy. Therefore, an updated classification of shrinkage porosity for spheroidal graphite iron is proposed.