Subsampling and Aggregation: A Solution to the Scalability Problem in Distance-Based Prediction for Mixed-Type Data

The distance-based linear model (DB-LM) extends the classical linear regression to the framework of mixed-type predictors or when the only available information is a distance matrix between regressors (as it sometimes happens with big data). The main drawback of these DB methods is their computational cost, particularly due to the eigendecomposition of the Gram matrix. In this context, ensemble regression techniques provide a useful alternative to fitting the model to the whole sample. This work analyzes the performance of three subsampling and aggregation techniques in DB regression on two specific large, real datasets. We also analyze, via simulations, the performance of bagging and DB logistic regression in the classification problem with mixed-type features and large sample sizes.

Drone RGB Images as a Reliable Information Source to Determine Legumes Establishment Success

The use of drones in agriculture is becoming a valuable tool for crop monitoring. There are some critical moments for crop success; the establishment is one of those. In this paper, we present an initial approximation of a methodology that uses RGB images gathered from drones to evaluate the establishment success in legumes based on matrixes operations. Our aim is to provide a method that can be implemented in low-cost nodes with relatively low computational capacity. An index (B1/B2) is used for estimating the percentage of green biomass to evaluate the establishment success. In the study, we include three zones with different establishment success (high, regular, and low) and two species (chickpea and lentils). We evaluate data usability after applying aggregation techniques, which reduces the picture’s size to improve long-term storage. We test cell sizes from 1 to 10 pixels. This technique is tested with images gathered in production fields with intercropping at 4, 8, and 12 m relative height to find the optimal aggregation for each flying height. Our results indicate that images captured at 4 m with a cell size of 5, at 8 m with a cell size of 3, and 12 m without aggregation can be used to determine the establishment success. Comparing the storage requirements, the combination that minimises the data size while maintaining its usability is the image at 8 m with a cell size of 3. Finally, we show the use of generated information with an artificial neural network to classify the data. The dataset was split into a training dataset and a verification dataset. The classification of the verification dataset offered 83% of the cases as well classified. The proposed tool can be used in the future to compare the establishment success of different legume varieties or species.

Water Quality Models and their Use in Monitoring the Quality of Surface Water Bodies

Water quality has been considered as one of the major challenges in water resource management. The main reason of degradation of water quality over the years is anthropogenic activities. Also, the monitoring of surface water bodies is a tedious as well as expensive process. For the depiction of water quality in simple and easy to understand terminology Water Quality Index (WQI) is found to be one of the widely used tool. It provides a transparent picture of the status of the pollution of a water body that is why it has been widely accepted by policy makers as well as other concerned authorities. Many WQI models have been developed throughout the world, using various water quality parameters, different techniques to generate subindices and also involving various mathematical techniques for aggregation of subindices. This paper deals with the comparison of various water quality models-based om number of parameters used, methods to generate subindices, aggregation techniques as well as their application and uses.

Bayesian mixture model for clustering rare-variant effects in human genetic studies

Rare-variant aggregate analysis from exome and whole genome sequencing data typically summarizes with a single statistic the signal for a gene or the unit that is being aggre- gated. However, when doing so, the effect profile within the unit may not be easily characterized across one or multiple phenotypes. Here, we present an approach we call Multiple Rare-Variants and Phenotypes Mixture Model (MRPMM), which clusters rare variants into groups based on their effects on the multivariate phenotype and makes statistical inferences about the properties of the underlying mixture of genetic effects. Using summary statistic data from a meta-analysis of exome sequencing data of 184,698 individuals in the UK Biobank across 6 populations, we demonstrate that our mixture model can identify clusters of variants responsible for significantly disparate effects across a multivariate phenotype; we study three lipid and three renal traits separately. The method is able to estimate (1) the proportion of non-null variants, (2) whether variants with the same predicted consequence in one gene behave similarly, (3) whether variants across genes share effect profiles across the multivariate phenotype, and (4) whether different annotations differ in the magnitude of their effects. As rare-variant data and aggregation techniques become more common, this method can be used to ascribe further meaning to association results.

High Levels IPF as Possible Predictor Heparin-Induced Thrombocytopenia

Background: Heparin-Induced Thrombocytopenia (HIT) represents a serious complication of heparin treatment. IgG antibodies binding Platelet Factor 4 (PF4) and heparin trigger the clinical manifestations of HIT. A 4T score is used to stratify the selection of patients suitable for examination. However, the selection of suitable patients remains at the discretion of the clinician, who is confronted with determining the cause of thrombocytopenia. The inclusion of the evaluation of the Immature platelet fraction result seems to be a suitable complement to the stratification of patients because we do not climb elevated IPF values when consuming platelets due to their immunization. Materials and Methods: In a group of 432 thrombocytopenic samples IPF was detected and analyzed in 45 patients with suspected HIT, a 4T score was determined; IPF and HIT functional tests were examined. IPF was determined by oxazine fluorescent dyeing structures of nucleic acid-containing platelets and fluorescence detection on a Sysmex XN 1000 analyser. To determine HIT, impedance aggregometry using the Multiplate® analyser (MEA) as heparin-induced aggregation techniques. The MEA method uses sensitization of donor platelets with patient plasma in the presence of heparin at a concentration of 0.5IU/mL. Results: From the results of the test, it is evident that 10 patients from our group of 45 examined showed positivity of HIT, which is a significant number due to the proven occurrence of HIT in patients treated with LMWH and showing thrombocytopenia. If we evaluate these 10 patients in terms of IPF value, it is evident that 6 of them have an increased value of IPF >10%, which is a 33% positive predictive value and 4 have IPF >30%, when the positive predictive value is even 100%. Conclusions: Diagnosis of HIT remains a complicated clinical laboratory issue. However, new diagnostic options provide considerable potential for solving this problem. The implementation of IPF assays helps us in the diagnosis of HIT on two levels. On the one hand, it provides us with information on platelet consumption in hospitalized patients and thus draws our attention to HIT as one of the options for congestive thrombocytopenia, unless, of course, disseminated intravascular coagulation or thrombotic microangiopathy. Secondly, its implementation will increase the predictive value of the 4T score in patients at medium risk, which is, however, the vast majority indicated for HIT examination.

Full Content-based Web Page Classification Methods by using Deep Neural Networks

The quality of the web page classification process has a huge impact on information retrieval systems. In this paper, we proposed to combine the results of text and image data classifiers to get an accurate representation of the web pages. To get and analyse the data we created the complicated classifier system with data miner, text classifier, and aggregator. The process of image and text data classification has been achieved by the deep learning models. In order to represent the common view onto the web pages, we proposed three aggregation techniques that combine the data from the classifiers.

Low-Complexity Aggregation Techniques for DOA Estimation over Wide-RF Bandwidths

Accurate direction of arrival (DOA) estimation of wideband, low-power nonstationary signals is important in many radio frequency (RF) applications. This article analyses the performance of two incoherent aggregation techniques for the DOA estimation of high chirp-rate linear frequency modulated (LFM) signals used in modern radar and electronic warfare (EW) applications. The aim is to determine suitable aggregation techniques for blind DOA estimation for real-time implementation with a frequency channelised signal. The first technique calculates a single pseudospectrum by directly combining the spatial covariance matrices from each of the frequency bins. The second technique first calculates the spatial pseudospectra from the spatial covariance matrix (SCM) from each frequency bin and then combines the spatial pseudospectra into one single estimate. Firstly, for single and multiple signal emitters, we compare the DOA estimation performance of incoherent SCM-based aggregation with that of the incoherent spatial pseudospectra-based aggregation using the root mean-squared error (RMSE). Secondly, we determine the types of signals and conditions for which these incoherent aggregation techniques are more suited. We demonstrate that the low-complexity SCM-based aggregation technique can achieve relatively good estimation performance compared to the pseudospectra-based aggregation technique for multiple narrowband signal detection. However, pseudospectra aggregation is better suited for single wideband emitter detection. Both the incoherent aggregation techniques presented in this article offer a computational advantage over the coherent processing techniques and hence are better suited for real-time implementation.