Aggregation of preferences in recommender systems

This chapter gives an overview of aggregation functions toward their use in recommender systems. Simple aggregation functions such as the arithmetic mean are often employed to aggregate user features, item ratings, measures of similarity, etc., however many other aggregation functions exist which could deliver increased accuracy and flexibility to many systems. We provide definitions of some important families and properties, sophisticated methods of construction, and various examples of aggregation functions in the domain of recommender systems.

Vector valued similarity measures for Atanassov's intuitionistic fuzzy sets

We present a new approach for defining similarity measures for Atanassov's intuitionistic fuzzy sets (AIFS), in which a similarity measure has two components indicating the similarity and hesitancy aspects. We justify that there are at least two facets of uncertainty of an AIFS, one of which is related to fuzziness while other is related to lack of knowledge or non-specificity. We propose a set of axioms and build families of similarity measures that avoid counterintuitive examples that are used to justify one similarity measure over another. We also investigate a relation to entropies of AIFS, and outline possible application of our method in decision making and image segmentation. © 2014 Elsevier Inc. All rights reserved.

The diet of Pacific gulls (Larus pacificus) breeding at Seal Island in northern Bass Strait

The endemic Pacific gull (Larus pacificus) is Australia's largest larid, and though little is currently known of its foraging ecology, its size and wide distribution suggest that it may play an important role within the marine environment. In the present study, regurgitate pellets collected from Seal Island in northern Bass Strait were used to compare intra- and interannual trends in diet composition. The main taxa identified in pellets were the common diving-petrel (Pelecanoides urinatrix), leatherjacket species (Family Monacanthidae), short-tailed shearwater (Puffinus tenuirostris) and mirror bush (Coprosma repens). Analysis of similarity (ANOSIM) identified no significant differences in numerical abundance of the dominant prey species between years, suggesting that the prey base in this region is temporally consistent or that the gulls consume low enough numbers to be unaffected by fluctuation in prey populations. Diving-petrels were consumed in consistently high numbers, suggesting the gulls may be an important predator of this species, or that the gulls are particularly skilled at foraging for them. © CSIRO 2014.

Characterization of antibody V segment diversity in the Tasmanian devil (Sarcophilus harrisii)

The Tasmanian devil (Sarcophilus harrisii) immune system has recently been under scrutiny because of the emergence of a contagious cancer, which has decimated devil numbers. Here we provide a comprehensive description of the Tasmanian devil immunoglobulin variable regions. We show that heavy chain variable (VH) and light chain variable (VL) repertoires are similar to those described in other marsupial taxa: VL diversity is high, but VH diversity is restricted and belongs only to clan III. As in other mammals, one VH and one Vλ germline family and multiple incomplete Vκ germline sequences were identified in the genome. High Vκ variation was observed in transcripts and we predict that it may have arisen by gene conversion and/or somatic mutations, as it does not appear to have originated from germline variation. Phylogenetic analyses revealed that devil VL gene segments are highly complex and ancient, with some lineages predating the separation of marsupials and eutherians. These results indicate that although the evolutionary history of immune genes lead to the expansions and contractions of immune gene families between different mammalian lineages, some of the ancestral immune gene variants are still maintained in extant species. A high degree of similarity was found between devil and other marsupial VH segments, demonstrating that they originated from a common clade of closely related sequences. The VL families had a higher variation than VH both between and within species. We suggest that, similar to other studied marsupial species, the complex VL segment repertoire compensates for the limited VH diversity in Tasmanian devils.

Biased experts and similarity based weights in preferences aggregation

In a group decision making setting, we consider the potential impact an expert can have on the overall ranking by providing a biased assessment of the alternatives that differs substantially from the majority opinion. In the framework of similarity based averaging functions, we show that some alternative approaches to weighting the experts' inputs during the aggregation process can minimize the influence the biased expert is able to exert.

Strains of Xylella fastidiosa rapidly distinguished by arbitrarily primed-PCR

Genomic DNAs isolated from strains of Xylella fastidiosa that caused citrus variegated chlorosis, coffee leaf scorch, Pierce's Disease of grapevine, and plum leaf scorch were analyzed by arbitrarily primed polymerase chain reaction. Purified DNA was amplified under nonstringent conditions with single primers 21 nucleotides (nt) long. Thirty-nine amplification products were observed that were useful to distinguish among the strains and to derive a similarity matrix and construct a phenogram showing possible relationships among the strains. Strains isolated from diseased coffee and citrus in Brazil were closely related to each other (coefficient of similarity of 0.872), but only distantly related to a strain isolated from diseased grapevine in the USA (coefficient of similarity of 0.650). Strains of Xylella fastidiosa isolated from diseased plums in the USA and Brazil clustered with strains from different hosts isolated from their respective countries of origin. Thus, there may be two quite dissimilar clusters of strains of Xylella fastidiosa, one in North America and the other in South America. Each cluster contains strains that can cause disease in plum. The methods described provide a convenient and rapid method to distinguish between strains of Xylella fastidiosa that cause diseases of coffee and citrus in the same region of Brazil. This has not been possible previously. This will potentially enable the two strains to be distinguished in alternate hosts or in insect vectors.

Strain differentiation of Trichophyton rubrum by randomly amplified polymorphic DNA and analysis of rDNA nontranscribed spacer

Trichophyton rubrum is the most common pathogen causing dermatophytosis. Molecular strain-typing methods have recently been developed to tackle epidemiological questions and the problem of relapse following treatment. A total of 67 strains of T rubrum were screened for genetic variation by randomly amplified polymorphic DNA (RAPD) analysis, with two primers, 5'-d[GGTGCGGGAA]-3' and 5'-d[CCCGTCAGCA]-3', as well as by subrepeat element analysis of the nontranscribed spacer of rDNA, using the repetitive subelements TRS-1 and TRS-2. A total of 12 individual patterns were recognized with the first primer and 11 with the second. Phylogenetic analysis of the RAPID products showed a high degree of similarity (> 90 %) among the epidemiologically related clinical isolates, while the other strains possessed 60% similarity. Specific amplification of TRS-1 produced three strain-characteristic banding patterns (PCR types); simple patterns representing one copy of TRS-1 and two copies of TRS-2 accounted for around 85 % of all isolates. It is concluded that molecular analysis has important implications for epidemiological studies, and RAPID analysis is especially suitable for molecular typing in T. rubrum.

Comparison of the genomes of two Xanthomonas pathogens with differing host specificities

The genus Xanthomonas is a diverse and economically important group of bacterial phytopathogens, belonging to the gamma-subdivision of the Proteobacteria. Xanthomonas axonopodis pv. citri (Xac) causes citrus canker, which affects most commercial citrus cultivars, resulting in significant losses worldwide. Symptoms include canker lesions, leading to abscission of fruit and leaves and general tree decline(1). Xanthomonas campestris pv. campestris (Xcc) causes black rot, which affects crucifers such as Brassica and Arabidopsis. Symptoms include marginal leaf chlorosis and darkening of vascular tissue, accompanied by extensive wilting and necrosis(2). Xanthomonas campestris pv. campestris is grown commercially to produce the exopolysaccharide xanthan gum, which is used as a viscosifying and stabilizing agent in many industries(3). Here we report and compare the complete genome sequences of Xac and Xcc. Their distinct disease phenotypes and host ranges belie a high degree of similarity at the genomic level. More than 80% of genes are shared, and gene order is conserved along most of their respective chromosomes. We identified several groups of strain-specific genes, and on the basis of these groups we propose mechanisms that may explain the differing host specificities and pathogenic processes.

Assessment of the ichthyofauna in stretches under the influence of Salto Grande Reservoir (Middle Paranapanema River, SP/PR, Brazil).

Aim: To analyze the composition, species richness and spatial distribution of the fish fauna in the area under influence of the Salto Grande Reservoir; Methods: Fish were caught every two months from November/05 to October/06, using gill nets, seining nets and sieve; Results: It was registered 67 fish species, 1,964 individuals and 146.2 kg, representing CPUEn of 1,964 and CPUEb of 278 kg. The most representative orders were Characiformes, with 29 species, Siluriformes (21 species) and Perciformes (11 species) Nine non-native species were registered. The most abundant species in the reservoir were Astyanax altiparanae, Steindachnerina insculpta and Acestrorhynchus lacustris, indicating the predominance of medium and small fishes in this reservoir. The dendrogram of similarity separated the Dam and Pedra Branca stretches from Pardo River Mouth, indicating differences in the assemblages. The species turnover among the stretches was demonstrated by the beta diversity, which may be related to the diversity of habitat; Conclusions: Although being a small reservoir, it displays great habitat diversity, reflecting in the composition and structure of fish assemblages.

Bathymetric distribution of brachyuran crab (Crustacea, Decapoda) communities on coastal soft bottoms off southeastern Brazil

The communities of brachyuran crabs living on soft bottoms off Ubatuba in SE Brazil were studied with respect to their structure, bathymetric distribution, composition, diversity and indices of similarity. The data were analyzed using multivariate techniques of classification and ordination. Most of the individuals caught during summer were the swimming crab Portunas spinicarpus at the 35 m isobath, which contributed to the much-decreased diversity in this season and site. Multivariate analysis indicated that the species were distributed according to depth and also in relation to environmental gradients.

Molecular and ultrastructural profiles of the symbionts in Cephalotes ants

The molecular and ultrastructural profiles of the symbionts found in the midgut and ileum of Cephalotes atratus, Cephalotes clypeatus, and Cephalotes pusillus were determined using the V3 region of the bacterial 16S rDNA gene and transmission electron microscopy (T.E.M.). Two samples of C atratus, three of C clypeatus, and six of C. pusillus were analyzed. The coefficients of similarity ranged from 80% to 94% for the samples of symbionts from C. clypeatus and C. atratus, despite being collected in geographically distant sites. The variability within symbionts found in the samples of C. pusillus varied from 29% to 55%, in samples geographically close as well as distant. PCR-DGGE was effective for the purpose of this study and can be considered a versatile tool to analyze gut microbiota. Details of the ultrastructural aspect of these bacteria are presented. (C) 2010 Elsevier Ltd. All rights reserved.

Food resource utilization of the skates Rioraja agassizii (Müller & Henle, 1841) and Psammobatis extenta (Garman, 1913) on the continental shelf off Ubatuba, South-eastern Brazil

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Identification of key microhabitats for fish assemblages in tropical Brazilian savanna streams

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Genetic variability in natural populations of Zeyheria montana mart. from the Brazilian Cerrado

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

PHYLOGENETIC STUDIES OF SOME SPECIES OF THE GENUS COFFEA .1. NUMERICAL-ANALYSIS OF FLAVONOID COMPOUNDS

Flavonoid compounds were analyzed in ripe fruit pulp of ten species of Coffea, including two cultivars of C. arabica and two of C. canephora. Three coefficients of similarity: Simple-Matching, Jaccard and Ochiai and three different clustering methods, Single Linkage, Complete Linkage and Unweighted Pair Group, Using Arithmetic Averages (UPGMA), were used to analyze the data.Jaccard and Ochiai's coefficients of association showed a more coherent result, when compared with taxonomic and hybridization studies. Inclusion of Psilanthopsis kapakata in the genus Coffea, as C. kapakata, is justified by the similarity of this species with other studied species, and clusters clearly approximate the species C. arabica and C. eugenioides. The latter is one of the possible parents of the allotetraploid species C. arabica, C. congensis is the only species whose position remains ambiguous, probably due to the fact that the plants of this species that were introduced into the Campinas collections, were hybrids and not typical of C. congensis.