FIDIC条款的文体特征

FIDIC即“Conditions of Contract For Works of Civil Engineering Construction”(国际土木工程施工合同条款),是编制招标文件、投标以及合同管理的必不可少的部分,是土木工程的“Bible”(圣经)。本文较全面深入地分析总结了FIDIC条款在词法、句法及篇章结构三个方面的文体特征,以期对这类合同的理解、商定以及教学等有所帮助。     

An improved strategy for skin lesion detection and classification using uniform segmentation and feature selection based approach

Melanoma is the deadliest type of skin cancer with highest mortality rate. However, the annihilation in early stage implies a high survival rate therefore, it demands early diagnosis. The accustomed diagnosis methods are costly and cumbersome due to the involvement of experienced experts as well as the requirements for highly equipped environment. The recent advancements in computerized solutions for these diagnoses are highly promising with improved accuracy and efficiency. In this article, we proposed a method for the classification of melanoma and benign skin lesions. Our approach integrates preprocessing, lesion segmentation, features extraction, features selection, and classification. Preprocessing is executed in the context of hair removal by DullRazor, whereas lesion texture and color information are utilized to enhance the lesion contrast. In lesion segmentation, a hybrid technique has been implemented and results are fused using additive law of probability. Serial based method is applied subsequently that extracts and fuses the traits such as color, texture, and HOG (shape). The fused features are selected afterwards by implementing a novel Boltzman Entropy method. Finally, the selected features are classified by Support Vector Machine. The proposed method is evaluated on publically available data set PH2. Our approach has provided promising results of sensitivity 97.7%, specificity 96.7%, accuracy 97.5%, and F‐score 97.5%, which are significantly better than the results of existing methods available on the same data set. The proposed method detects and classifies melanoma significantly good as compared to existing methods.     

基于归一化特征判别的日志模板挖掘算法

针对传统日志模板挖掘时需要日志聚类数目作为先验信息的问题,提出了一种基于归一化特征判别的日志模板挖掘算法.首先,对日志数据进行压缩,以提高后续处理效率;其次,进行日志聚类过程,使用归一化的日志统计特征判断聚类是否满足要求,若满足,则聚类成功;若不满足,则采用二分搜索的方式调整日志聚类的数目,重新进行聚类;最后,从聚类结果中提取日志模板,设计了一种衡量模板挖掘效果的评价指标.在真实数据集上的实验结果表明,算法的模板挖掘匹配度优于基准方法,并且具有良好的泛化性能.     

“九厅十八井”围屋建筑特征及保护策略研究——以清远清代钟氏大屋村为例

钟氏大屋村是清远地区保存较为完整的清代“九厅十八井”围屋建筑实例,在建筑形制与营造技艺上具有典型的客家围屋特色,对研究清远地区人文历史具有重要意义。通过实地调研,在分析造成建筑损坏的社会历史因素,运用测绘工具、摄影技术和计算机技术建立建筑数字化模型的基础上,解析清远钟氏大屋村围屋建筑的布局特征与防御特性,以及记录了建筑的细部特征,为该建筑的保护和可持续发展提供合适的策略,以激活建筑并延续其生命力。     

基于域对抗网络和BERT的跨领域文本情感分析

跨领域文本情感分析时,为了使抽取的共享情感特征能够捕获更多的句子语义信息特征,提出域对抗和BERT(bidirectional encoder representations from transformers)的深度网络模型。利用BERT结构抽取句子语义表示向量,通过卷积神经网络抽取句子的局部特征。通过使用域对抗神经网络使得不同领域抽取的特征表示尽量不可判别,即源领域和目标领域抽取的特征具有更多的相似性;通过在有情感标签的源领域数据集上训练情感分类器,期望该分类器在源领域和目标领域均能达到较好的情感分类效果。在亚马逊产品评论数据集上的试验结果表明,该方法具有良好的性能,能够更好地实现跨领域文本情感分类。     

基于预测数据特征的空气质量预测方法

采用LightGBM预测模型对空气质量预测问题进行研究,提出并设计一种基于预测性特征的空气质量预测方法,有效地预测北京市区内未来24 h核心表征空气质量的PM_2.5质量浓度。在构建预测方案过程中,分析训练数据集特性开展数据清洗,利用随机森林与线性插值相结合的方法,解决数据大量缺失以及噪声干扰问题;提出使用预测性数据特征方法,同时设计相关统计特征,提高预测结果的准确性;采用滑窗机制挖掘高维时间特征,增加数据特征数量级;对预测模型的工作性能和结果进行详细分析,并结合基线模型进行对比评价。试验结果表明,基于预测性特征结合采用LightGBM预测模型的方案具有更高的预测精度。     

基于BERT的复合网络模型的中文文本分类

针对自然语言在语句结构上有着较强的前后依赖关系,提出一种基于BERT的复合网络模型进行中文新闻分类。首先利用BERT的基于注意力机制的多层双向transformer特征提取器获得字词以及句子之间更加全局的特征关系表达。然后将所得数据输入门结构更加简单的双向门控循环神经网络层将特征增强的同时减少时间代价,加强数据特征的选取精确度。最后将不同权重的文本特征信息输入softmax函数层进行新闻分类。通过在cnews新浪新闻数据集上进行实验,获得97.21%的F1值,结果表明所提特征融合模型较其他模型分类效果更好。     

Surface modifications of activated carbon by gamma irradiation

Four commercial activated carbons with different chemical and textural characteristics were modified by gamma irradiation under five different conditions: irradiated in absence of water, in presence of ultrapure water, in ultrapure water at pH = 1.0 and 1000 mg L⁻¹ Cl⁻, in ultrapure water at pH = 7.5 and 1000 mg L⁻¹ Br⁻, and in ultrapure water at pH = 12.5 and 1000 mg L⁻¹ NO₃⁻. Changes in surface chemistry were studied by X-ray photoelectron spectroscopy; pH of point of zero charge, total acidic groups and total basic groups, which were determined by assessment with HCl and NaOH; and textural changes were determined by obtaining the corresponding adsorption isotherms of N₂ and CO₂. Outcomes show that the activated carbon surface chemistry can be modified by gamma irradiation and that the changes depend on the irradiation conditions. Modifications in the sp² hybridization of the surface carbons suggest that the irradiated carbons undergo graphitization. Measurements of structural parameters indicate that the irradiation treatment does not modify the textural properties of the carbons. Finally, studies of pristine and irradiated activated carbons using diffuse reflectance spectroscopy with the Kubelka–Munk function revealed a reduction in band gap energy in the irradiated carbons associated with an increase in sp² hybridization of the carbon atoms.     

Role of activated carbon on micropollutans degradation by ionizing radiation

The objective of this study was to analyze the influence of the presence of activated carbon on the degradation of the triiodinated contrast medium diatrizoate (DTZ) by the simultaneous use of gamma radiation and activated carbon. Four commercial activated carbons (Ceca, Witco, Sorbo, and Merck) with different textural and chemical characteristics were used for this purpose. The percentage DTZ removal obtained was considerably higher with the gamma radiation/activated carbon (GM/AC) system than with radiolysis in the absence of activated carbon, and it depended on the specific activated carbon employed. First, we optimized the amount of activated carbon required to maximize the amount of DTZ removed by the GM/AC system (0.06 g). The degradation constants were higher with the GM/AC system than with radiolysis alone, evidencing a synergic effect that favors pollutant removal. This synergic effect is independent of the textural but not the chemical characteristics of the activated carbon, observing a higher synergic activity for carbons with a higher surface content of oxygen, specifically quinone groups. We also highlight that the synergic effect of the activated carbon requires adsorbent–adsorbate electrostatic interaction and is absent when this interaction is hindered.