The features of radio wave scattering in multifractal media associated with the thickness and structure of a turbulent layer with inhomogeneity of the electronic medium of the mid-latitude ionosphere are considered. It is shown that the method of multidimensional structure functions is applicable under certain conditions for a thick layer with small-scale inhomogeneities

A total of 50 grain rice (Oryzae sativa) samples were tested to establish their mycological contamination and their aflatoxigenic potential. Rice is the most extensively consumed cereal grain by a substantial portion of the world\'s society, and in Asia predominantly. Under certain conditions, a variety of fungi may develop within rice grains; some of which have the capacity to synthesize mycotoxins. Thus, rice consumers are considered to be a high-risk population specially since this toxin has been linked to health problems and is also highly associated with liver cancer today. When compared to non-local samples, samples from Iraqi markets (of various origins), particularly imported ones, exhibited high quantities of fungus. From samples, three species of Aspergillus section Flavi (A. flavus, A. parasiticus, and A. tamarii) have been isolated and identified. Culture-based and ELISA approaches were used to detect aflatoxigenic A.strains. Fluorescence in response to UV long-wavelength (365 nm) light and pigment synthesis in response to ammonium hydroxide were used. By both methods, the ratio of aflatoxigenic A. flavus isolates to non-aflatoxigenic strains was higher. All the tested strains of A. parasiticus showed aflatoxigenic potential. Aflatoxigenic potential of selected strains by ELISA technique for A.parasiticus isolates ranged from 181.0 to 360 ppb, whilest, levels of aflatoxins production for A.flavus isolates ranged from 183 to 300 ppb. 9p

Transformer models with multi-head attentions outperformed all existing translation models. Nevertheless, some features of traditional statistical models, such as prior alignments between source and target words, have proved useful in training the state-of-the-art Transformer models. It has been reported that lightweight prior alignments can effectively guide a head in the multi-head cross-attention sublayer responsible for the alignment of Transformer models. In this work, we make a step further by applying heavyweight prior alignments to guide all heads. Specifically, we use the weight of 0.5 for the alignment cost added to the token cost in formulating the overall cost of training a Transformer model. The alignment cost is defined as the deviation of the attention probabilities from the prior alignments. The attention probabilities are computed by averaging all heads of the multi-head attention sublayer within the penultimate layer of Transformer models.\nExperimental results on a English-Vietnamese translation task show that our proposed approach helps train superior Transformer-based translation models. Our Transformer model outperforms the baseline model by the huge 4.37 BLEU. Case studies by human on some translation results validate the machine judgment. \nThe results so far encourage the use of heavyweight prior alignments to improve Transformer-based translation models.

A Collection H = (G1, G2, ..., Gn) of a graph G is said to be Gamma c decomposition if satisfies the condition γ_c(Gi) = i, 1 ≤ i ≤ n. Moreover, it must satisfy the condition \n⋃_(i=1)^n▒〖E(G_i)〗 = E(G). If each G_i is a path in Gamma c Decomposition, then G is said to be Gamma c Path Decomposition. In this paper, we characterize the aspects of Gamma c Decomposition. Extend this Gamma c Path Decomposition as Gamma c Path Decomposition number. In addition, we gathered some basic aspects of Gamma c Path Decomposition Number.

In the computed tomography (CT) images, at different times the lung nodule growth acquires primary malignant-benign feature. Since then, hybrid machines have been operating in this area for a long time. The main concern created by such a system is the design and adoption of these features, and the complexity of these approaches is limited in their applications. To overcome the above problems, an early prediction and classification of lung nodule diagnosis have been proposed on CT images based on hybrid machine learning (EPC-HML) techniques. The main contributions are summarized as follows: First, an adaptive crow swarm optimization with a mixture model (ACS-ML) for lung nodule segmentation using statistical information has been introduced. Second, a higher-order beam search optimization (HBS) algorithm for the feature extraction and selection process is illustrated. Third, a hybrid classifier i.e. semi module-convolutional neural network (SM-CNN) for nodule prediction and classification has been developed. The proposed EPC-HML attribute scheme provides more quantitative assessments for decisions support during the diagnosis process. The research pursuits and different experiments illustrate the strength of the novel-built model. The method could detection the defect and segment accuracy of 96.39%, a sensitivity of 95.25%, a specificity of 96.12%, an area under the curve (AUC) of 96.05%. Therefore, the technique is very helpful for the clinical detection of malignant benign cancer by accessing the nodule size shape at an early stage.

The COVID-19 pandemic has been an important\nrole in the world for one year. It has much more the negative\nimpact on the people. So the scientists have been working on the\nvaccines to get rid of COVID-19. The Chinese Scientists explored\nthe CoronaVac. In this paper we examine the CoronaVac in terms\nof Cryptography.

In this paper, we present two public-key cryptosystems\nover finite fields. First of them is based on polynomials. The\npresented system also considers a digital signature algorithm. Its\nsecurity is based on the difficulty of finding discrete logarithms\nover GF(qd+1) with sufficiently large q and d. Is is also examined\nalong with comparison with other polynomial based public-key\nsystems. The other public-key cryptosystem is based on linear\ncodes. McEliece studied the first code-based public-key cryptosystem.\nWe are inspired by McEliece system in the construction of\nthe new system. We examine its security using linear algebra and\ncompare it with the other code-based cryptosystems. Our new\ncryptosystems are too reliable in terms of security.