Machine Learning-based USD/PKR Exchange Rate Forecasting Using Sentiment Analysis of Twitter Data

This study proposes an approach based on machine learning to forecast currency exchange rates by applying sentiment analysis to messages on Twitter (called tweets). A dataset of the exchange rates between the United States Dollar (USD) and the Pakistani Rupee (PKR) was formed by collecting information from a forex website as well as a collection of tweets from the business community in Pakistan containing finance-related words. The dataset was collected in raw form, and was subjected to natural language processing by way of data preprocessing. Response variable labeling was then applied to the standardized dataset, where the response variables were divided into two classes: “1” indicated an increase in the exchange rate and “ −1” indicated a decrease in it. To better represent the dataset, we used linear discriminant analysis and principal component analysis to visualize the data in three-dimensional vector space. Clusters that were obtained using a sampling approach were then used for data optimization. Five machine learning classifiers—the simple logistic classifier, the random forest, bagging, naïve Bayes, and the support vector machine—were applied to the optimized dataset. The results show that the simple logistic classifier yielded the highest accuracy of 82.14% for the USD and the PKR exchange rates forecasting.     

Some factors affecting the ambient ozone concentrations measured at Ottawa

The concentration of ozone was measured from June 1972 to February 1973 at a location just outside the centre of Ottawa. Average monthly ozone levels were found to fall on a smooth curve with a maximum concentration in early summer and a minimum concentration in November. The daily ozone concentration profiles showed a consistent pattern, with a significant difference, however, between weekdays and weekends. The strong morning minimum found on weekdays could be attributed to anthropogenic effects. Measurements of NO and NO₂ concentrations appear to confirm the role of NO as the pollutant primarily responsible for the destruction of ozone.     

Investigation on the dispersion of carbon nanotubes in nitrile butadiene rubber: Role of polymer-to-filler grafting reaction

This contribution reports on the dispersion by simple melt blending of tiny amounts of carbon nanotubes (CNT) in nitrile butadiene rubbers (NBR). Acrylonitrile (ACN) units of NBR are known to generate free radicals upon heating and/or shearing. This paper highlights elements evidencing a possibility for NBR polymer chains to react by a free-radical mechanism and to graft onto CNT surface all along the process of mechanical blending of NBR with CNTs. More precisely and since the formation of the free-radicals takes place on the ACN units, the influence of the ACN relative content in NBR on the grafted CNT amount has been studied. It comes out that the polymer-grafting rate onto the CNT surface increases with the ACN content in NBR. Interestingly, the nanotubes proved more finely dispersed in NBR containing higher relative ACN content as evidenced by morphological observations as well as electrical measurements.     

Partial spatial coherence effects in digital holographic microscopy with a laser source

We investigate a digital holographic microscope that permits us to modify the spatial coherence state of the sample illumination by changing the spot size of a laser beam on a rotating ground glass. Out-of-focus planes are refocused by digital holographic reconstruction with numerical implementation of the Kirchhoff-Fresnel integral. The partial coherence nature of the illumination reduces the coherent artifact noise with respect to fully coherent illumination. The investigated configuration allows the spatial coherence state to be changed without modifying the illumination level of the sample. The effect of the coherence state on the digital holographic reconstruction is theoretically and experimentally evaluated. We also show how multiple reflection interferences are limited by the use of reduced spatial coherent illumination.     

Identification of a gene encoding a homocitrate synthase isoenzyme of Saccharomyces cerevisiae

In Saccharomyces cerevisiae, most of the LYS structural genes have been identified except the genes encoding homocitrate synthase and α-aminoadipate aminotransferase. Expression of several LYS genes responds to an induction mechanism mediated by the product of LYS14 and an intermediate of the pathway, α-aminoadipate semialdehyde (αAASA) as an inducer. This activation is modulated by the presence of lysine in the growth medium leading to an apparent repression. Since the first enzyme of the pathway, homocitrate synthase, is feedback inhibited by lysine, it could be a major element in the control of αAASA supply. During the sequencing of chromosome IV of S. cerevisiae, the sequence of ORF D1298 showing a significant similarity with the nifV gene of Azotobacter vinelandii was reported. Disruption and overexpression of ORF D1298 demonstrate that this gene, named LYS20, encodes a homocitrate synthase. The disrupted segregants are able to grow on minimal medium and exhibit reduced but significant homocitrate synthase indicating that this activity is catalysed by at least two isoenzymes. We have also shown that the product of LYS20 is responsible for the greater part of the lysine production. The different isoforms are sensitive to inhibition by lysine but only the expression of LYS20 is strongly repressed by lysine. The N-terminal end of homocitrate synthase isoform coded by LYS20 contains no typical mitochondrial targeting sequence, suggesting that this enzyme is not located in the mitochondria.     

Variation of cardinal growth parameters and growth limits according to phylogenetic affiliation in the Bacillus cereus Group. Consequences for risk assessment

The growth rates of strains covering the seven major phylogenetic groups of Bacillus cereus sensu lato (as defined by Guinebretiere et al., 2008) at a range of temperature (7 °C–55 °C), pH (4.6–7.5) and a_w (0.929–0.996, with 0.5%–10% NaCl as humectant) were determined. Growth rates were fitted by non-linear regression to determine the cardinal parameters T_min, T_opt, T_max, pH_min, pH_opt, a_wmin and μ_opt. We showed that cardinal parameters reflected the differences in the temperature adaptation observed between B. cereus phylogenetic groups I to VII. The ability of growing at low pH (up to 4.3) or low a_w (from a_w 0.929 and up to 10% NaCl) varied among strains. The strains of groups III and VII, the most tolerant to heat, were also the most adapted to high NaCl (all strains growing at 8% NaCl) and the ones of groups I and VI the least adapted (no growth at 7% NaCl). All strains of groups II and VII were able to grow at pH 4.6, and only a few strains of group VI. Phenotypic differences between the two psychrotrophic groups II and VI were revealed by contrasted acid and salt tolerance. The cardinal values determined in this work were validated by comparing with cardinal parameters of a panel of strains published elsewhere and with predictions of growth in a range of foods.     

Synthesis and flash sintering of zirconium nitride powder

Zirconium nitride (ZrN) is a transition metal nitride of great interest due to its excellent physical and chemical properties. This study aims to synthesize ZrN fine powders by a facile and low-cost urea route that avoids the use of any solvent. ZrCl₄ and urea mixtures were heat-treated at up to 1600˚C in nitrogen gas. The products were characterized by X-ray diffraction, scanning electron microscopy, transmission electron microscopy, energy-dispersive X-ray spectroscopy, and thermogravimetric analysis. The effects of different processing parameters such as metal to urea molar ratio, heat treatment temperature, and dwelling time on the product phase and stoichiometry were studied to understand the synthesis method. In addition, synthesized ZrN powder was consolidated into near fully dense single-phase bulk ceramic with a homemade flash sintering setup. A constant DC electrical field of ∼80 V/cm and pressure of ∼14 MPa at room temperature triggered flash sintering without pre-heating, and the entire process finished in 200 s. The composition, microstructure, density, hardness, and oxidation properties of the sintered pellet were also characterized.     

Platooning Control for Sailboats Using a Tack Strategy

International Journal of Control, Automation and Systems - This paper addresses the problem of platooning control for a fleet of sailboats. A quadrilateral path is proposed to avoid going into the...     

Do additive manufactured parts deserve better?

Additive manufacturing of metallic components is regarded as one of the more exciting developments in engineering. The combined attractions of near net shape, tailored composition, and geometry optimisation have led to much interest in the various processes used and a drive to improve the mechanical properties to match those of wrought parts. In this paper, we reflect on the apparent lack of ambition in optimising the structural integrity of parts made using these new manufacturing processes. The current research focus seems to be either on largely irrelevant static properties, or on quantifying the fatigue response in a way that would be familiar to engineers in the 19^th Century. Given the work on the role of microstructure and fatigue, which dates back to Ewing and Humphrey in 1903 reaching its zenith in the 1980s and 90s with Keith Miller in the vanguard, and recent developments in both imaging technologies and sophisticated numerical modelling, all the elements are in place for a much more rigorous, and ultimately more fruitful, approach to understand the structural integrity of additive manufactured components.