结合句法依存信息的方面级情感分类

引入句法依存信息到原方面术语,提出一种新的方面术语表示方法,利用Glove词向量表示单词以及单词与单词之间的依存关系,构造出包含句法依存信息的依存关系邻接矩阵和依存关系表示矩阵,利用图卷积神经网络和多头注意力机制将句法依存信息融入到方面术语中,使得方面术语表达与上下文结构高度相关。将改进后的方面词术语表示替换到现有模型后,模型泛化能力得到有效提升。对比试验和分析结果表明:该方法具有有效性和泛化性。     

Genetic and genomic analysis of age at first insemination in Israeli dairy cattle

We performed a genetic analysis of age at first insemination, including estimation of the heritability and genetic correlations with other economic traits, and the consequences of including this trait in the Israeli selection index. The genetic factors affecting age at first insemination were determined via GWAS. Five data sets were analyzed. Data sets 1, 2, and 3 were used to compute variance components among age at first insemination, first calving age, days from first insemination to calving, and the 9 traits included in the Israel breeding index. Heritabilities for age at first insemination, calving age, and days from first insemination to calving in Israeli Holsteins as computed by REML individual animal model analyses of 273,239 Israeli Holstein cows were 0.072, 0.042, and 0.014. The estimated genetic correlation between the first 2 traits was 0.88. In addition to the fact that heritability of age at first insemination is 1.7 times the heritability for calving, the former trait has the advantage that the number of records is greater, and the records are generated earlier. Absolute values of the genetic and residual correlations between age at first insemination and the 9 traits included in the Israeli index were all less than 0.2. Data set 4 included first insemination dates of 1,181,600 calves born from 1985 through 2018. Genetic evaluations were computed by a single trait animal model. Annual phenotypic and genetic trends for age at first calving for calves born since 1985 were “positive,” that is, economically negative, at 0.320 ± 0.003 and 0.169 ± 0.005 d, respectively. Applying the GCTA-GREML software, 54% of variance in the transmitting ability of 1,585 sires could be explained by considering all 40,498 markers included in the GWAS analysis. The significant markers were mainly associated with milk production genes. The SNP UA-IFASA-8854 on chromosome 11 had the lowest probability value, 1.2 × 10^?24. This marker is located between the genes RETSAT and ELMOD3, both of which are overexpressed in human mammary glands. The gene RETSAT is reported to be essential for lipid accumulation and adipogenesis promotion. Gene enrichment analysis found that genes in the genomic region flanking significant markers are associated with vasopressin receptor activity, which was shown to mediate puberty in humans. If age at first insemination is included in the index with a weighting to account for 9% of the index, reductions of 2.8 and 2.6 d for age at first insemination and first calving age after 10 yr of selection are predicted, as compared with reductions of 1.4 and 1.1 d with the current index. Gains for the other index traits are only marginally affected. We suggest selection on age at first insemination as an alternative to selection for early calving.     

Improving accuracy of automatic optical inspection with machine learning

Electronic devices require the printed circuit board(PCB)to support the whole structure,but the assembly of PCBs suffers from welding problem of the electronic components such as surface mounted devices(SMDs)resistors.The automated optical inspection(AOI)machine,widely used in industrial production,can take the image of PCBs and examine the welding issue.However,the AOI machine could commit false negative errors and dedicated technicians have to be employed to pick out those misjudged PCBs.This paper proposes a machine learning based method to improve the accuracy of AOI.In particular,we propose an adjacent pixel RGB value based method to pre-process the image from the AOI machine and build a customized deep learning model to classify the image.We present a practical scheme including two machine learning procedures to mitigate AOI errors.We conduct experiments with the real dataset from a production line for three months,the experimental results show that our method can reduce the rate of misjudgment from 0.3%–0.5%to 0.02%–0.03%,which is meaningful for thousands of PCBs each containing thousands of electronic components in practice.     

Dilated and soft attention-guided convolutional neural network for breast cancer histology images classification

Breast cancer is one of the most common types of cancer in women, and histopathological imaging is considered the gold standard for its diagnosis. However, the great complexity of histopathological images and the considerable workload make this work extremely time-consuming, and the results may be affected by the subjectivity of the pathologist. Therefore, the development of an accurate, automated method for analysis of histopathological images is critical to this field. In this article, we propose a deep learning method guided by the attention mechanism for fast and effective classification of haematoxylin and eosin-stained breast biopsy images. First, this method takes advantage of DenseNet and uses the feature map's information. Second, we introduce dilated convolution to produce a larger receptive field. Finally, spatial attention and channel attention are used to guide the extraction of the most useful visual features. With the use of fivefold cross-validation, the best model obtained an accuracy of 96.47% on the BACH2018 dataset. We also evaluated our method on other datasets, and the experimental results demonstrated that our model has reliable performance. This study indicates that our histopathological image classifier with a soft attention-guided deep learning model for breast cancer shows significantly better results than the latest methods. It has great potential as an effective tool for automatic evaluation of digital histopathological microscopic images for computer-aided diagnosis.     

The International Society for the Study of Vascular Anomalies (ISSVA) ontology

The International Society for the Study of Vascular Anomalies (ISSVA) provides a classification for vascular anomalies that enables specialists to unambiguously classify diagnoses. This classification is only available in PDF format and is not machine-readable, nor does it provide unique identifiers that allow for structured registration. In this paper, we describe the process of transforming the ISSVA classification into an ontology. We also describe the structure of this ontology, as well as two applications of the ontology using examples from the domain of rare disease research. We used the expertise of an ontology expert and clinician during the development process. We semi-automatically added mappings to relevant external ontologies using automated ontology matching systems and manual assessment by experts. The ISSVA ontology should contribute to making data for vascular anomaly research more Findable, Accessible, Interoperable, and Reusable (FAIR). The ontology is available at https://bioportal.bioontology.org/ontologies/ISSVA.     

Practical Application of the Probabilistic‐Statistical Model of the Suspension Separation in Hydrocyclones

An effective practical approach that allows not only a significant reduction in the scope of practical experiments in the course of studying suspension separation processes in hydrocyclones, but also makes it possible to assess the intensity of random components of the processes and define the interrelation between such components and hydrodynamics of flows in a hydrocyclone is presented. Within the frames of the developed probabilistic‐statistical model of suspension separation in hydrocyclones on the basis of statistical self‐similarity properties, a relationship was found between determined and random components of the processes. This allowed transitioning from three‐parameter probability density functions for suspension particles in hydrocyclones to two‐parameter functions; thus significantly improving the efficiency of practical application of the developed model.     

Sequential alignment attention model for scene text recognition

Scene text recognition has been a hot research topic in computer vision due to its various applications. The state-of-the-art solutions usually depend on the attention-based encoder-decoder framework that learns the mapping between input images and output sequences in a purely data-driven way. Unfortunately, there often exists severe misalignment between feature areas and text labels in real-world scenarios. To address this problem, this paper proposes a sequential alignment attention model to enhance the alignment between input images and output character sequences. In this model, an attention gated recurrent unit (AGRU) is first devised to distinguish the text and background regions, and further extract the localized features focusing on sequential text regions. Furthermore, CTC guided decoding strategy is integrated into the popular attention-based decoder, which not only helps to boost the convergence of the training but also enhances the well-aligned sequence recognition. Extensive experiments on various benchmarks, including the IIIT5k, SVT, and ICDAR datasets, show that our method substantially outperforms the state-of-the-art methods.     

明代盔甲种类及制作工艺研究

本文通过盔甲的制作及材料,区分了非金属材料的皮甲、瓠子作甲、藤甲、绵甲、纸甲以及金属材料组成的齐腰甲、齐腰明甲、缉甲、布甲、铁甲、钢甲、铜甲、锡甲等类型,并对其制作方式等进行分析。结合不同材质盔甲的使用地区、战争方式进行研究,以对明代盔甲有一个整体性的认知。