Development of Hierarchical Clustering Techniques for Gridded Data from Mixed Data Sequences

Knowledge discovery in databases is the non-trivial process of identifying valid, novel potentially useful and ultimately understandable patterns from data. The term Data mining refers to the process which does the exploratory analysis on the data and builds some model on the data. To infer patterns from data, data mining involves different approaches like association rule mining, classification techniques or clustering techniques. Among the many data mining techniques, clustering plays a major role, since it helps to group the related data for assessing properties and drawing conclusions. Most of the clustering algorithms act on a dataset with uniform format, since the similarity or dissimilarity between the data points is a significant factor in finding out the clusters. If a dataset consists of mixed attributes, i.e. a combination of numerical and categorical variables, a preferred approach is to convert different formats into a uniform format. The research study explores the various techniques to convert the mixed data sets to a numerical equivalent, so as to make it equipped for applying the statistical and similar algorithms. The results of clustering mixed category data after conversion to numeric data type have been demonstrated using a crime data set. The thesis also proposes an extension to the well known algorithm for handling mixed data types, to deal with data sets having only categorical data. The proposed conversion has been validated on a data set corresponding to breast cancer. Moreover, another issue with the clustering process is the visualization of output. Different geometric techniques like scatter plot, or projection plots are available, but none of the techniques display the result projecting the whole database but rather demonstrate attribute-pair wise analysis

The C-peptide response to a standard mixed meal in a group of Brazilian type 1 diabetic patients

In order to analyze the different parameters used in the interpretation of C-peptide response in a functional test, we compared a group of 26 type 1 diabetics aged 21.1 ± 8.2 years, with a diabetes duration of 7.9 ± 6.7 months, with a group of 24 non-diabetic subjects aged 25.0 ± 4.4 years. A standard mixed meal of 317 kcal was used as a stimulus. Blood sampling for C-peptide determinations was performed at regular intervals. Although all the studied C-peptide variables were significantly lower in the diabetic group (P<0.0001), some overlapping of parameters was observed between the two groups. The highest degree of overlapping was found for basal value (BV) (30.8%) and percent increase (42.31%), and the lowest for incremental area, absolute increase, peak value (PV) (3.8%), and total area (7.7%) (c2 = 31.6, P<0.0001). We did not observe a definite pattern in the time of maximum response among the 21 diabetics who showed an increase in C-peptide levels after the stimulus. In this group, however, there was a highly significant number of late responses (120 min) (c2 = 5.7, P<0.002). Although BV showed a significant correlation with PV (rS = 0.95, P<0.0001), the basal levels of C-peptide did not differentiate the groups with and without response to the stimulus. We conclude that the diabetic group studied showed delayed and reduced C-peptide responses, and that the functional test can be an important tool for the evaluation of residual ß cell function.

Magnetic suspension of the rotor of a ventricular assist device of mixed flow type

This work presents results of preliminary studies concerning application of magnetic bearing in a ventricular assist device (VAD) being developed by Dante Pazzanese Institute of Cardiology-IDPC (Sao Paulo, Brazil). The VAD-IDPC has a novel architecture that distinguishes from other known VADs. In this, the rotor has a conical geometry with spiral impellers, showing characteristics that are intermediate between a centrifugal VAD and an axial VAD. The effectiveness of this new type of blood pumping principle was showed by tests and by using it in heart surgery for external blood circulation. However, the developed VAD uses a combination of ball bearings and mechanical seals, limiting the life for some 10 h, making impossible its long-term use or its use as an implantable VAD. As a part of development of an implantable VAD, this work aims at the replacement of ball bearings by a magnetic bearing. The most important magnetic bearing principles are studied and the magnetic bearing developed by Escola Politecnica of Sao Paulo University (EPUSP-MB) is elected because of its very simple architecture. Besides presenting the principle of the EPUSP-MB, this work presents one possible alternative for applying the EPUSP-MB in the IDPC-VAD.

The influence of environmental variables on capybara (Hydrochoerus hydrochaeris: Rodentia, Hydrochoeridae) detectability in anthropogenic environments of southeastern Brazil

Capybaras were monitored weekly from 1998 to 2006 by counting individuals in three anthropogenic environments (mixed agricultural fields, forest and open areas) of southeastern Brazil in order to examine the possible influence of environmental variables (temperature, humidity, wind speed, precipitation and global radiation) on the detectability of this species. There was consistent seasonality in the number of capybaras in the study area, with a specific seasonal pattern in each area. Log-linear models were fitted to the sample counts of adult capybaras separately for each sampled area, with an allowance for monthly effects, time trends and the effects of environmental variables. Log-linear models containing effects for the months of the year and a quartic time trend were highly significant. The effects of environmental variables on sample counts were different in each type of environment. As environmental variables affect capybara detectability, they should be considered in future species survey/monitoring programs.

Incidence rate and spatio-temporal clustering of type 1 diabetes in Santiago, Chile, from 1997 to 1998

OBJECTIVE: To estimate the incidence rate of type 1 diabetes in the urban area of Santiago, Chile, from March 21, 1997 to March 20, 1998, and to assess the spatio-temporal clustering of cases during that period. METHODS: All sixty-one incident cases were located temporally (day of diagnosis) and spatially (place of residence) in the area of study. Knox's method was used to assess spatio-temporal clustering of incident cases. RESULTS: The overall incidence rate of type 1 diabetes was 4.11 cases per 100,000 children aged less than 15 years per year (95% confidence interval: 3.06--5.14). The incidence rate seems to have increased since the last estimate of the incidence calculated for the years 1986--1992 in the metropolitan region of Santiago. Different combinations of space-time intervals have been evaluated to assess spatio-temporal clustering. The smallest p-value was found for the combination of critical distances of 750 meters and 60 days (uncorrected p-value = 0.048). CONCLUSIONS: Although these are preliminary results regarding space-time clustering in Santiago, exploratory analysis of the data method would suggest a possible aggregation of incident cases in space-time coordinates.

Clustering of variables with a three-way approach for health sciences

TPM Vol. 21, No. 4, December 2014, 435-447 – Special Issue © 2014 Cises.

How can clusters help in the product type choice? The case of a specialized retail store

This work project (WP) is a study about a clustering strategy for Sport Zone. The general cluster study’s objective is to create groups such that within each group the individuals are similar to each other, but should be different among groups. The clusters creation is a mix of common sense, trial and error and some statistical supporting techniques. Our particular objective is to support category managers to better define the product type to be displayed in the stores’ shelves by doing store clusters. This research was carried out for Sport Zone, and comprises an objective definition, a literature review, the clustering activity itself, some factor analysis and a discriminant analysis to better frame our work. Together with this quantitative part, a survey addressed to category managers to better understand their key drivers, for choosing the type of product of each store, was carried out. Based in a non-random sample of 65 stores with data referring to 2013, the final result was the choice of 6 store clusters (Figure 1) which were individually characterized as the main outcome of this work. In what relates to our selected variables, all were important for the distinction between clusters, which proves the adequacy of their choice. The interpretation of the results gives category managers a tool to understand which products best fit the clustered stores. Furthermore, as a side finding thanks to the clusterization, a STP (Segmentation, Targeting and Positioning) was initiated, being this WP the first steps of a continuous process.

Clustering barotrauma patients in ICU–A data mining based approach using ventilator variables

Lecture Notes in Computer Science, 9273

Vibration training elicits a mixed aerobic and resistance-type exercise in sedentary subjects.

Objective. Vibration training (VT) is a new exercise method, with good acceptance among sedentary subjects, due to its passive principle: the machine moves the subject, not the opposite. We hypothesize that untrained subjects can benefit from a greater cardiovascular and metabolic stimulation than trained athletes, resembling classical aerobic-type activity, in addition of eliciting strength gains shown in diverse studies. Methods. 3 group of male subjects, inactive (SED), endurance trained athletes (END) and strength trained athletes (STR) underwent fitness (VO2max) and lower-body strength tests (isokinetic). Subjects were submitted to a session of oscillating VT, composed of 3 exercises (isometric half-squat, dynamic squat, dynamic squat with added load), each of 3 minutes duration, and repeated at 3 frequencies. VO2, heart rate and Borg scale were monitored. Results. 27 healthy subjects (10 SED, 9 END and 8 STR), mean age 24.5 (SED), 25.0 (STR) and 29.8 (END) were included. VO2max was significantly different as expected (47.9 vs. 52.9 vs. 63.9 ml/kg/min, resp. for SED, STR and END). Isokinetic dominant leg extensors strength was higher in STR (3.32 Nm/kg vs. 2.60 and 2.74 in SED and END). During VT, peak oxygen consumption (% of VO2max) attained was 59.3 in SED, 50.8 in STR and 48.0 in END (P<0.001 between SED and other subjects). Peak heart rate (% of heart rate max) was 82.7 in SED, 80.4 in STR and 72.4 in END. In SED, dynamic exercises without extra load elicited 51.0% of VO2max and 72.1% of heart rate max, and perceived effort reached 15.1/20. Conclusions. VT is an unconventional type of exercise, which has been shown to enhance strength, bone density, balance and flexibility. Users are attracted by the relative passivity. In SED, we show that VT elicits sufficient cardiovascular response to benefit overall fitness in addition to the known strength effects. VT's higher acceptance as an exercise in sedentary people, compared to jogging or cycling for example, can lead to better adherence to physical activity. Although long-term effects of VT on health are not avalaible, we believe this type of combination of aerobic and resistance-type exercise can be beneficial on multiple health parameters, especially cardiovascular health.

Molecular characterization of G and P-types bovine rotavirus strains from Goiás, Brazil: high frequency of mixed P-type infections

In this study, 331 samples from calves less than one month old from a dairy herd in the district of Piracanjuba, state of Goiás, Brazil were tested for rotavirus. Thirty-three samples (9.9%) tested positive for rotavirus. Out of those, 31 were submitted to G and P characterization by reverse transcription followed by semi-nested polymerase chain reaction. Two samples were characterized as G6P[1], three as G10P[11] and five as G6P[11]. The majority of the samples (51.6%) displayed multiple P genotypes (P-genotype mixtures), including typical human genotypes P[4] and P[6M], suggesting the occurrence of co-infections and genetic reassortment. Also, the detection of human genotypes in bovine samples may be considered evidence of the zoonotic potential of rotaviruses. To our knowledge, this is the first report of such a high frequency of P genotype mixtures in bovine rotavirus samples. It also increases data on G and P rotavirus genotypes circulating in dairy herds in Brazil and can help in the development of more efficient immunization approaches, thereby controlling infection and reducing economical losses.

Omega 3 fatty acids induce a marked reduction of apolipoprotein B48 when added to fluvastatin in patients with type 2 diabetes and mixed hyperlipidemia: a preliminary report

BACKGROUND Mixed hyperlipidemia is common in patients with diabetes. Statins, the choice drugs, are effective at reducing lipoproteins that contain apolipoprotein B100, but they fail to exert good control over intestinal lipoproteins, which have an atherogenic potential. We describe the effect of prescription omega 3 fatty acids on the intestinal lipoproteins in patients with type 2 diabetes who were already receiving fluvastatin 80 mg per day. METHODS Patients with type 2 diabetes and mixed hyperlipidemia were recruited. Fasting lipid profile was taken when patients were treated with diet, diet plus 80 mg of fluvastatin and diet plus fluvastatin 80 mg and 4 g of prescription omega 3 fatty acids. The intestinal lipoproteins were quantified by the fasting concentration of apolipoprotein B48 using a commercial ELISA. RESULTS The addition of 4 g of prescription omega 3 was followed by significant reductions in the levels of triglycerides, VLDL triglycerides and the triglyceride/HDL cholesterol ratio, and an increase in HDL cholesterol (P < 0.05). Fluvastatin induced a reduction of 26% in B100 (P < 0.05) and 14% in B48 (NS). However, the addition of omega 3 fatty acids enhanced this reduction to 32% in B100 (NS) and up to 36% in B48 (P < 0.05). CONCLUSION Our preliminary findings therefore suggest an additional benefit on postprandial atherogenic particles when omega 3 fatty acids are added to standard treatment with fluvastatin.

Discourse Type Clustering using POS n-gram Profiles and High-Dimensional Embeddings

Abstract: To cluster textual sequence types (discourse types/modes) in French texts, K-means algorithm with high-dimensional embeddings and fuzzy clustering algorithm were applied on clauses whose POS (part-ofspeech) n-gram profiles were previously extracted. Uni-, bi- and trigrams were used on four 19th century French short stories by Maupassant. For high-dimensional embeddings, power transformations on the chi-squared distances between clauses were explored. Preliminary results show that highdimensional embeddings improve the quality of clustering, contrasting the use of bi and trigrams whose performance is disappointing, possibly because of feature space sparsity.

PhD Students’ Research Group Social Capital in Two Countries: A Clustering Approach with Duocentred Network Measures

The article examines the structure of the collaboration networks of research groups where Slovenian and Spanish PhD students are pursuing their doctorate. The units of analysis are student-supervisor dyads. We use duocentred networks, a novel network structure appropriate for networks which are centred around a dyad. A cluster analysis reveals three typical clusters of research groups. Those which are large and belong to several institutions are labelled under a bridging social capital label. Those which are small, centred in a single institution but have high cohesion are labelled as bonding social capital. Those which are small and with low cohesion are called weak social capital groups. Academic performance of both PhD students and supervisors are highest in bridging groups and lowest in weak groups. Other variables are also found to differ according to the type of research group. At the end, some recommendations regarding academic and research policy are drawn

Spin-glass behaviour in mixed metal oxides with a rutile-type structure

We report here on the magnetic properties of compounds of composition Fe1−xCrxSbO4 and Fe1−xGaxSbO4. The introduction of paramagnetic Cr3+ and diamagnetic Ga3+ into the rutile‐related iron antimonate lattice does not destroy the antisite atomic ordering which exists in iron antimonate of composition FeSbO4. The initial slope of the Curie temperature dependence on x is similar in both series, indicating that Fe3+‐Cr3+ interactions are very small. The magnetic susceptibility measurements recorded from the compounds of composition Fe1−xCrxSbO4, x<0.4, and Fe0.9Ga0.1SbO4 show them to behave as spin glasses at low temperatures. The inhibition of compounds of the type Fe1−xCrxSbO4, x>0.4, and Fe1−xGaxSbO4, x>0.1 to undergo a spin‐glass transition above 4.2 K is associated with a dilution effect.