Prediction of Corn and Sugar Prices Using Machine Learning, Econometrics, and Ensemble Models

This paper explores the use of several state-of-the-art machine learning models for predicting the daily prices of corn and sugar in Brazil in relation to the use of traditional econometrics models. The following models were implemented and compared: ARIMA, SARIMA, support vector regression (SVR), AdaBoost, and long short-term memory networks (LSTM). It was observed that, even though the prices time series for both products differ considerably, the models that presented the best results were obtained by: SVR, an ensemble of the SVR and LSTM models, an ensemble of the AdaBoost and SVR models, and an ensemble of the AdaBoost and LSTM models. The econometrics models presented the worst results for both products for all metrics considered. All models presented better results for predicting corn prices in relation to the sugar prices, which can be related mainly to its lower variation during the training and test sets. The methodology used can be implemented for other products.

Seasonal variation in the essential oil from Varronia curassavica Jacq. accessions

Varronia curassavica has anti-inflammatory properties because of the terpenes, α-humulene and β-caryophyllene, present in the essential oil. The objective of this study was to evaluate the influence of seasonality on the essential oil of V. curassavica accessions. Leaves from six accessions were collected from the Germplasm Bank of the Federal University of Minas Gerais over 12 months. Correlations between the essential oil content and meteorological factors were determined. Gas chromatography analysis coupled with mass spectrometry was conducted to determine the chemical composition of the essential oil. The content and chemical composition of the oil varied throughout the year. Relative humidity was correlated with accessions ICA-VC2 (-0.64) and ICA-VC4 (0.68). β-bourbonene, β-elemene, spathulenol, germacrene, caryophyllene oxide, α-humulene, and β-caryophyllene occurred in all accessions. Accession ICA-VC3 exhibited lower variation (22.17%), higher average (0.97%) essential oil, and maintained an average abundance of α-humulene greater than 2.6%, which is the amount necessary for phytotherapeutics.

Identification of Reference Genes for Expression Studies in the Whole-Blood from Three Cattle Breeds under Two States of Livestock Weather Safety

Real-time PCR is widely used to study the relative abundance of mRNA due to its specificity, sensitivity, and repeatability quantification. However, relative quantification requires a reference gene, which should be stable in its expression, showing lower variation by experimental conditions or tissues. The aim of this study was to evaluate the stability of the expression of five commonly used reference genes (actb, ywhaz, b2m, sdha, and 18s rRNA) at different physiological stages (alert and emergency) in three different cattle breeds. In this study, five genes (actb, ywhaz, b2m, sdha, and 18s rRNA) were selected as candidate reference genes for expression studies in the whole blood from three cattle breeds (Romosinuano, Gyr, and Brahman) under heat stress conditions. The transcription stability of the candidate reference genes was evaluated using geNorm and NormFinder. The results showed that actb, 18SrRNA, and b2m expression were the most stable reference genes for whole blood of Gyr and Brahman breeds under two states of livestock weather safety (alert and emergency). Meanwhile, actb, b2m, and ywhaz were the most stable reference genes for the Romosinuano breed.

Fabrication and Characterization of Nanonet-Channel LTPS TFTs Using a Nanosphere-Assisted Patterning Technique

We present the fabrication and electrical characteristics of nanonet-channel (NET) low-temperature polysilicon channel (LTPS) thin-film transistors (TFTs) using a nanosphere-assisted patterning (NAP) technique. The NAP technique is introduced to form a nanonet-channel instead of the electron beam lithography (EBL) or conventional photolithography method. The size and space of the holes in the nanonet structure are well controlled by oxygen plasma treatment and a metal lift-off process. The nanonet-channel TFTs show improved electrical characteristics in terms of the ION/IOFF, threshold voltage, and subthreshold swing compared with conventional planar devices. The nanonet-channel devices also show a high immunity to hot-carrier injection and a lower variation of electrical characteristics. The standard deviation of VTH (σVTH) is reduced by 33% for a nanonet-channel device with a gate length of 3 μm, which is mainly attributed to the reduction of the grain boundary traps and enhanced gate controllability. These results suggest that the cost-effective NAP technique is promising for manufacturing high-performance nanonet-channel LTPS TFTs with lower electrical variations.

Comparison of linear combination modeling strategies for GABA-edited MRS at 3T

Purpose J-difference-edited spectroscopy is a valuable approach for the in vivo detection of γ-aminobutyric-acid (GABA) with MRS. A recent expert consensus article recommends linear combination model-ing (LCM) of edited MRS, but does not give specific details of implementation. This study explores different modeling strategies to adapt LCM for GABA-edited MRS. Methods 62 medial parietal lobe GABA-edited MEGA-PRESS spectra from a recent 3T multi-site study were modeled using 102 different strategies combining six different approaches to account for co-edited macromolecules, three modeling ranges, three baseline knot spacings, and using basis sets with or without homocarnosine. The resulting GABA and GABA+ estimates (quantified relative to total creatine), the residuals at different ranges, SDs and CVs, and Akaike information criteria, were used to evaluate the models' performance. Results Significantly different GABA+ and GABA estimates were found when a well-parameterized MM3co basis function was included in the model. The mean GABA estimates were significantly lower when modeling MM, while the CVs were similar. A sparser spline knot spacing led to lower variation in the GABA and GABA+ estimates and a narrower modeling range - only including the signals of interest - did not substantially improve or degrade modeling performance. Additionally, results suggest that LCM can separate GABA and the underlying co-edited MM3co. Incorporating homocarnosine into the modeling did not significantly improve variance in GABA+ estimates. Conclusion GABA-edited MRS is best quantified by LCM with a well-parameterized co-edited MM3co basis function with a constraint to the non-overlapped MM0.93 in combination with a sparse spline knot spacing and a modeling range between 0.5 and 4 ppm.

Development of a modified cross-over hop test to reduce measurement errors in return-to-competition testing

Context Hop tests play an important role in the rehabilitation process after injuries. A comparison of the jumping distances of both extremities allows for an evaluation of the injured limb. In the conventional cross-over hop test for distance, the jump width (medial vs. lateral) that the athlete has to cross during the jump is not standardised and therefore highly variable. This affects the absolute jump length in each jump series. Hypothesis Modifying the test may reduce the jump length variance between test series of an athlete as well as the test-dependent variations in the cross-over hop for distance. Methods N = 47 athletes from the German and French national Judo youth teams were included in the study (age: 15.3 years ± 13–17). A modified version of the cross-over hop for distance was developed with a cross-over width of 50 cm and a fixed landing zone of 10 cm. The jump lengths of the conventional test and the modified test were documented. The change in jump length variations of the two sexes were compared. Results The mean value of the coefficient of variation decreased significantly from 4.09 % to 2.83 % (p < 0.01) due to the test modification. This resulted in an absolute improvement in accuracy of 1.26 % and a relative improvement of 30.8 %. A comparison of the limb symmetry index between the conventional and the modified cross-over hop for distance revealed no significant differences. Conclusion The modified cross-over hop for distance showed a significantly lower variation in jump lengths compared with the conventional cross-over hop for distance. As a result, more accurate statements can be made regarding the patient’s return-to-competition progress.

Effect of Cu Ions Implantation on Structural, Electronic, Optical and Dielectric Properties of Polymethyl Methacrylate (PMMA)

In this work, the effect of ion bombardment on the optical properties of Polymethylmethacrylate (PMMA) was studied. Polymer samples were implanted with 500 keV Cu+ ions with a fluence ranging from 1 × 1012 to 1 × 1014 ions/cm2. X-ray Diffractometer (XRD) study indicated a relatively lower variation with a higher dose of ions. Fourier Transform Infrared (FTIR) spectra exhibited that with the implantation of Cu ions the intensity of existing bands decreases, while the result confirms the existence of a C=C group. The pristine and ion-implanted samples were also investigated using photoluminescence (PL) and Ultra Violet-Visible (UV-VIS) spectra. The optical band gap (Eg) was observed up to 3.05 eV for the implanted samples, while the pristine sample exhibited a wide energy-gap up to ~3.9 eV. The change in the optical gap indicated the presence of a gradual phase transition for the polymer blends. The dielectric measurements of the pristine and Cu-implanted PMMA were investigated in the 10 Hz to 2 GHz frequency range. It was found that the implanted samples showed a significant decrease in the value of the dielectric constant. The value of the dielectric constant and dielectric loss of the PMMA and Cu-implanted samples at a 1-kHz frequency were found to be ~300 and 29, respectively. The modification of the PMMA energy bandgap in the current research suggested the potential use of Cu implanted PMMA in the field of optical communications and flexible electronic devices.

The Use of UAVs for the Characterization and Analysis of Rocky Coasts

Rocky coasts represent three quarters of all coastlines worldwide. These areas are part of ecosystems of great ecological value, but their steep configuration and their elevation make field surveys difficult. This fact, together with their lower variation rates, explains the lower numbers of publications about cliffs and rocky coasts in general compared with those about beach-dune systems. The introduction of UAVs in research, has enormously expanded the possibilities for the study of rocky coasts. Their relative low costs allow for the generation of information with a high level of detail. This information, combined with GIS tools, enables coastal analysis based on Digital Models and high spatial resolution images. This investigation summarizes the main results obtained with the help of UAVs between 2012 and the present day in rocky coastline sections in the northwest of the Iberian Peninsula. These investigations have particularly focused on monitoring the dynamics of boulder beaches, cliffs, and shore platforms, as well as the structure and function of ecosystems. This work demonstrates the importance of unmanned aerial vehicles (UAVs) for coastal studies and their usefulness for improving coastal management. The Galician case was used to explain their importance and the advances in the UAVs’ techniques.

Comparison of Hysteresis Based PWM Schemes ΔΣ-PWM and Direct Torque Control

This paper presents the differences and similarities of ΔΣ-PWM as a hysteresis-based PWM scheme with direct torque control (DTC) using simulation models. The variable switching frequency caused by the hysteresis element is examined with regard to its instantaneous values. The comparison is based on an equal maximum switching frequency as a design criterion. With this first assumption, the variation of the instantaneous switching frequency is higher when using DTC because of the temporary prioritization of one inverter leg. Besides the lower variation, ΔΣ-PWM shows a higher average switching frequency. Because the switching frequency is related to the torque ripple, the usage of ΔΣ-PWM results in a smaller torque ripple. Due to the dependence of torque ripple on switching frequency, a second comparison is carried out based on the same average switching frequency. In this comparison the ΔΣ-PWM shows higher torque ripple than DTC.

Reference-based QUantification Of gene Dispensability (QUOD)

Background Dispensability of genes in a phylogenetic lineage, e.g. a species, genus, or higher-level clade, is gaining relevance as most genome sequencing projects move to a pangenome level. Most analyses classify genes as core genes, which are present in all investigated individual genomes, and dispensable genes, which only occur in a single or a few investigated genomes. The binary classification as ‘core’ or ‘dispensable’ is often based on arbitrary cutoffs of presence/absence in the analysed genomes. Even when extended to ‘conditionally dispensable’, this concept still requires the assignment of genes to distinct groups. Results Here, we present a new method which overcomes this distinct classification by quantifying gene dispensability and present a dedicated tool for reference-based QUantification Of gene Dispensability (QUOD). As a proof of concept, sequence data of 966 Arabidopsis thaliana accessions (Ath-966) were processed to calculate a gene-specific dispensability score for each gene based on normalised coverage in read mappings. We validated this score by comparison of highly conserved Benchmarking Universal Single Copy Orthologs (BUSCOs) to all other genes. The average scores of BUSCOs were significantly lower than the scores of non-BUSCOs. Analysis of variation demonstrated lower variation values between replicates of a single accession than between iteratively, randomly selected accessions from the whole dataset Ath-966. Functional investigations revealed defense and antimicrobial response genes among the genes with high-dispensability scores. Conclusions Instead of classifying a gene as core or dispensable, QUOD assigns a dispensability score to each gene. Hence, QUOD facilitates the identification of candidate dispensable genes, associated with high dispensability scores, which often underlie lineage-specific adaptation to varying environmental conditions.