Search

With the proliferation of smart wearables, motion wristbands provide a wealth of data essential for comprehending the dynamic nature of health. However, outlier detection is typically necessary due to the presence of unknown outliers in their multidimensional activity data. Conventional approaches frequently result in incorrect object identification due to the curse of dimensionality. Using the Gaussian Mixture Generative Model (GMGM), we provide a method to identify outliers and address this problem. Training on raw data is done using a VariationalAutoencoder (VAE). While avoiding rebuilding mistakes, we want to achieve as many brief features as possible. To predict the likelihood that examples contain many types of data, a DBN will utilise feature extractions and latent distributions in the future. The model's robustness is enhanced by enhancing the VAE, deep learning components, and the GMM overall. When densities surpass the training level, the Gaussian Mixture Model identifies outliers. To achieve this, it makes educated guesses about the densities of each data point. Compared to the deep learning Autoencoding Gaussian Mixture Model (DAGMM), GMGM achieves a 5.5 % higher area under the curve (AUC) on the ODDS standard dataset. Experiments conducted on real datasets further demonstrate the efficacy of this strategy.

This study compared yield, nutrient compositions and forage quality of Napier grass (Pennisetum purpureum Schumach.) and pearl millet (Pennisetum glaucum L.) to determine their preference utilization as forage crops. Four accessions each of Napier grass and pearl millet were grown in the field with three replications in a Randomized Complete Block Design. Data were collected and analyzed on vegetative growth and yield parameters at 10 weeks after planting. Also nutrient composition of the shoot was determined. Napier grass grew more vigorously with significantly higher number of leaf/plant, leaf length and width. Meanwhile, plant height were similar (91.02 -111.95 cm) among the accessions bur for pearl millet IP17862 which had the least (82.82 cm). Napier grass accessions had significantly higher stem girth than the pearl millet. Mean tiller/plant was less than one in pearl millet and between 18.21-31.55 in Napier grass. Total soluble sugar and carbohydrate of the millet accessions were comparable with COM-CO3 and COM-CO4 Napier grass. Pearl millet NGB00551 was the richest in calcium (3.79 mg/g), potassium (0.36 mg/g), iron (0.52 ppm) and Zinc (0.60 ppm), however, IP17862 and COM-CO4 contained highest phosphorous and magnesium respectively. Dry matter percentage (DM), acid detergent fibre (ADF) and neutral detergent fibre (NDF) were higher in Napier grass than pearl millet accessions, but pearl millet accessions had more crude protein than the Napier grass. In overall, forage quality was higher in pearl millet accessions compared to Napier grass. [ABSTRACT FROM AUTHOR]

Aim: The objective of the present study was to evaluate in vitro, the influence of different ligatures, archwire material and archwire sizes on static frictional resistance. Methods: The frictional resistance was evaluated by using different ligatures Gray, Purple, Alastic, Teflon coated stainless steel ligature and 0.009"Stainless steel ligatures on 0.017"×0.025" and 0.019"×0.025" TMAand SS archwires using standard edgewise brackets with zero degree tip and torque mounted on Perspex blocks using an universal testing machine with a cross head speed of 5mm/min with unstimulated saliva as a lubricant. Results: Mean and SD values were calculated. Data was statistically analyzed using three way analysis of variance (ANOVA), pair wise comparison by Tukeys multiple post hoc test and multiple regression analysis to calculate coefficient of static friction. The results showed that irrespective of wire size and wire material combination Teflon coated SS ligatures produced the least frictional resistance while Gray elastomeric modules produced the highest with Alastic, SS ligature and purple falling in between. Multiple linear regression analysis of static friction revealed that only arch wire size has significant influence on static friction with groups and archwire composition showing negligible influence. Interpretation and conclusion: (a)Teflon coated stainless steel ligatures produced lowest frictional forces with all wire, material contributions followed by Alastic, Stainless steel, Purple and gray elastomeric modules. (b)Increase in arch wire dimension has increased the frictional resistance, whereas archwire composition has minimal effect on frictional resistance. [ABSTRACT FROM AUTHOR]

In this paper, we have presented the symbolic processing of MATLAB to compute the high order Newton Cotes closed formulas. The weight coefficients are found as fractions in exact form by use of the symbolic computation up to 100th order. These computed values of weight coefficients are used in the program for Newton Cotes integration as symbolic constants to test an example integral. The result of the integral confirms that high order Newton Cotes formulas will also be useful. The results are more accurate than those obtained by a similar MATLAB program under double precision arithmetic. The MATLAB programs under double precision arithmetic diverge for rules greater than 20. Where as under symbolic computation the results converge up to order 70 and surprisingly some results of order 80, 90, 95 and 100 also show convergence. The MATLAB programs and the results of the example integral are given for future applications. [ABSTRACT FROM AUTHOR]

In this paper we have presented the symbolic processing of MATLAB to compute the abscissas and weight coefficients of the Gauss Legendre quadrature formulas. There are at least four formulas to compute the Legendre polynomials which are fundamental to the accuracy of points (zeros/abscissas) and about eight formulas to compute the weight coefficients which depend on these points. We have chosen the most appropriate pair of formulas from among the available set of formulas to implement the MATLAB symbolic code. The output of this code for abscissas and weights is in agreement with the standard table of values which are reported in the literature, specially the one which is available in the internet by Pavel Holoborodko for up to 100th order, where we have used 32-digit variable precision arithmetic (vpa) command of MATLAB. The paper also test these formulas on an integration example which converges very slowly by using the high order Gauss Legendre quadrature rules .We have constructed a useful computer codes which gives abscissas and weight coefficients for Gauss Legendre quadrature rules up to the 150 th order. [ABSTRACT FROM AUTHOR]