Improving multivariate data streams clustering.

Clustering data streams is an important task in data mining research. Recently, some algorithms have been proposed to cluster data streams as a whole, but just few of them deal with multivariate data streams. Even so, these algorithms merely aggregate the attributes without touching upon the correlation among them. In order to overcome this issue, we propose a new framework to cluster multivariate data streams based on their evolving behavior over time, exploring the correlations among their attributes by computing the fractal dimension. Experimental results with climate data streams show that the clusters' quality and compactness can be improved compared to the competing method, leading to the thoughtfulness that attributes correlations cannot be put aside. In fact, the clusters' compactness are 7 to 25 times better using our method. Our framework also proves to be an useful tool to assist meteorologists in understanding the climate behavior along a period of time.

Wavelet analysis and hierarchical clustering of ofm phenotypic population's variations originated from southern america and europe and attempts of a first relation to molecular divergence.

In this work we compare Grapholita molesta Busck (Lepidoptera: Tortricidae) populations originated from Brazil, Chile, Spain, Italy and Greece using power spectral density and phylogenetic analysis to detect any similarities between the population macro- and the molecular micro-level. Log-transformed population data were normalized and AR(p) models were developed to generate for each case population time series of equal lengths. The time-frequency/scale properties of the population data were further analyzed using wavelet analysis to detect any population dynamics frequency changes and cluster the populations. Based on the power spectral of each population time series and the hierarchical clustering schemes, populations originated from Southern America (Brazil and Chile) exhibit similar rhythmic properties and are both closer related with populations originated from Greece. Populations from Spain and especially Italy, have higher distance by terms of periodic changes on their population dynamics. Moreover, the members within the same cluster share similar spectral information, therefore they are supposed to participate in the same temporally regulated population process. On the contrary, the phylogenetic approach revealed a less structured pattern that bears indications of panmixia, as the two clusters contain individuals from both Europe and South America. This preliminary outcome will be further assessed by incorporating more individuals and likely employed a second molecular marker.

Synchronizing the in vitro germination of Psidium guineense Sw. seeds by means of osmotic priming.

The Brazilian guava (Psidium guineense Swartz) is seed-propagated and, being native to the Caatinga biome, may frequently have uneven germination.Thus, we aimed to evaluate the synchronization of the in vitro seed germination of three accessions of the Brazilian guava, using water, polyethyleneglycol (PEG 6000), and potassium nitrate (KNO3) at different potentials and times of osmotic priming. Seeds from three accessions of the Brazilian guava (Y85, Y93,and Y97) from the UNEB/BA Germplasm Active Bank were subjected to the following pretreatments: -0.6, -1.0, -1.4, and -1,8 MPa PEG 6000; 10 and 20% KNO3 for 24h; 10 and 20% KNO3 for 48h; water for 24 and 48h; and non-primed seeds as the control. The experimental design was therefore a 10x3+1 factorial scheme. We assessed the germination percentage (G), mean germination time (MGT), germination speed (GS), and germination speed index (GSI). Data was subjected to analysis of variance followed by a means test (Duncan at 5% probability) and regression. There was interaction between the priming treatments and accessions for all evaluated features, except G. PEG 6000 decreased the MGT (from 6 to 8 days) and increased GS and GSI of seeds from all three accessions at potentials -1.0 to -1.5 MPa.Water-priming had a positive effect on MGT, GS, and GSI of accession Y85 seeds. KNO3 negatively affected germination of seeds from all three accessions. Thereby, we could synchronize seed germination of accessions Y85 and Y97 with PEG 6000.