Risk-based water quality decision-making under small data using Bayesian network

A knowledge-based network for Section Yidong Bridge,Dongyang River,one tributary of Qiantang River,Zhejiang Province,China,is established in order to model water quality in areas under small data.Then,based on normal transformation of variables with routine monitoring data and normal assumption of variables without routine monitoring data,a conditional linear Gaussian Bayesian network is constructed.A "two-constraint selection" procedure is proposed to estimate potential parameter values under small data.Among all potential parameter values,the ones that are most probable are selected as the "representatives".Finally,the risks of pollutant concentration exceeding national water quality standards are calculated and pollution reduction decisions for decision-making reference are proposed.The final results show that conditional linear Gaussian Bayesian network and "two-constraint selection" procedure are very useful in evaluating risks when there is limited data and can help managers to make sound decisions under small data.

A Review of Cybersecurity Challenges in Small Business: The Imperative for a Future Governance Framework

Technological shifts—coupled with infrastructure, techniques, and applications for big data—have created many new opportunities, business models, and industry expansion that benefit entrepreneurs. At the same time, however, entrepreneurs are often unprepared for cybersecurity needs—and the policymakers, industry, and nonprofit groups that support them also face technological and knowledge constraints in keeping up with their needs. To improve the ability of entrepreneurship research to understand, identify, and ultimately help address cybersecurity challenges, we conduct a literature review on the state of cybersecurity. The research highlights the necessity for additional investigation to aid small businesses in securing their confidential data and client information from cyber threats, thereby preventing the potential shutdown of the business.

A LoRa-based protocol for connecting IoT edge computing nodes to provide small-data-based services

Data is becoming increasingly personal.Individuals regularly interact with a variety of structured data,ranging from SQLite databases on the phone to personal sensors and open government data.The“digital traces left by individuals through these interactions”are sometimes referred to as“small data”.Examples of“small data”include driving records,biometric measurements,search histories,weather forecasts and usage alerts.In this paper,we present a flexible protocol called LoRaCTP,which is based on LoRa technology that allows data“chunks”to be transferred over large distances with very low energy expenditure.LoRaCTP provides all the mechanisms necessary to make LoRa transfer reliable by introducing a lightweight connection setup and allowing the ideal sending of an as-long-as necessary data message.We designed this protocol as communication support for small-data edge-based IoT solutions,given its stability,low power usage,and the possibility to cover long distances.We evaluated our protocol using various data content sizes and communication distances to demonstrate its performance and reliability.

Data science for oceanography:from small data to big data

The rapid development of ocean observation technology has resulted in the accumulation of a large amount of data and this is pushing ocean science towards being data-driven.Based on the types and distribution of oceanographic data,this paper analyzes the present and makes predictions for the future regarding the use of big and small data in ocean science.The ocean science has not fully entered the era of big data.There are two ways to expand the amount of oceanographic data to better understanding and man-agement of the ocean.On the data level,fully exploit the potential value of big and small ocean data,and transform the limited,small data into rich,big data,will help to achieve this.On the application level,oceanographic data are of great value if realize the federation of the core data owners and the consumers.The oceanographic data will provide not only a reliable scientific basis for climate,ecological,disaster and other scientific research,but also provide an unprecedented rich source of information that can be used to make predictions of the future.

An analytical model for estimating rock strength parameters from small-scale drilling data

The small-scale drilling technique can be a fast and reliable method to estimate rock strength parameters. It needs to link the operational drilling parameters and strength properties of rock. The parameters such as bit geometry, bit movement, contact frictions and crushed zone affect the estimated parameters.An analytical model considering operational drilling data and effective parameters can be used for these purposes. In this research, an analytical model was developed based on limit equilibrium of forces in a Tshaped drag bit considering the effective parameters such as bit geometry, crushed zone and contact frictions in drilling process. Based on the model, a method was used to estimate rock strength parameters such as cohesion, internal friction angle and uniaxial compressive strength of different rock types from operational drilling data. Some drilling tests were conducted by a portable and powerful drilling machine which was developed for this work. The obtained results for strength properties of different rock types from the drilling experiments based on the proposed model are in good agreement with the results of standard tests. Experimental results show that the contact friction between the cutting face and rock is close to that between bit end wearing face and rock due to the same bit material. In this case,the strength parameters, especially internal friction angle and cohesion, are estimated only by using a blunt bit drilling data and the bit bluntness does not affect the estimated results.

A new package:MySAS for small angle scattering data analysis

In this paper,A MySAS package,which is verified on Windows XP,can easily convert two-dimensional data in small angle neutron and X-ray scattering analysis,operate individually and execute one particular operation as numerical data reduction or analysis,and graphical visualization.This MySAS package can implement the input and output routines via scanning certain properties,thus recalling completely sets of repetition input and selecting the input files.On starting from the two-dimensional files,the MySAS package can correct the anisotropic or isotropic data for physical interpretation and select the relevant pixels.Over 50 model functions are fitted by the POWELL code using x^2 as the figure of merit function.

Model-data-driven seismic inversion method based on small sample data

As sandstone layers in thin interbedded section are difficult to identify,conventional model-driven seismic inversion and data-driven seismic prediction methods have low precision in predicting them.To solve this problem,a model-data-driven seismic AVO(amplitude variation with offset)inversion method based on a space-variant objective function has been worked out.In this method,zero delay cross-correlation function and F norm are used to establish objective function.Based on inverse distance weighting theory,change of the objective function is controlled according to the location of the target CDP(common depth point),to change the constraint weights of training samples,initial low-frequency models,and seismic data on the inversion.Hence,the proposed method can get high resolution and high-accuracy velocity and density from inversion of small sample data,and is suitable for identifying thin interbedded sand bodies.Tests with thin interbedded geological models show that the proposed method has high inversion accuracy and resolution for small sample data,and can identify sandstone and mudstone layers of about one-30th of the dominant wavelength thick.Tests on the field data of Lishui sag show that the inversion results of the proposed method have small relative error with well-log data,and can identify thin interbedded sandstone layers of about one-15th of the dominant wavelength thick with small sample data.

Consensus of heterogeneous multi-agent systems based on sampled data with a small sampling delay

In this paper, consensus problems of heterogeneous multi-agent systems based on sampled data with a small sampling delay are considered. First, a consensus protocol based on sampled data with a small sampling delay for heterogeneous multi-agent systems is proposed. Then, the algebra graph theory, the matrix method, the stability theory of linear systems, and some other techniques are employed to derive the necessary and sufficient conditions guaranteeing heterogeneous multi-agent systems to asymptotically achieve the stationary consensus. Finally, simulations are performed to demonstrate the correctness of the theoretical results.

Data processing of small samples based on grey distance information approach

Data processing of small samples is an important and valuable research problem in the electronic equipment test. Because it is difficult and complex to determine the probability distribution of small samples, it is difficult to use the traditional probability theory to process the samples and assess the degree of uncertainty. Using the grey relational theory and the norm theory, the grey distance information approach, which is based on the grey distance information quantity of a sample and the average grey distance information quantity of the samples, is proposed in this article. The definitions of the grey distance information quantity of a sample and the average grey distance information quantity of the samples, with their characteristics and algorithms, are introduced. The correlative problems, including the algorithm of estimated value, the standard deviation, and the acceptance and rejection criteria of the samples and estimated results, are also proposed. Moreover, the information whitening ratio is introduced to select the weight algorithm and to compare the different samples. Several examples are given to demonstrate the application of the proposed approach. The examples show that the proposed approach, which has no demand for the probability distribution of small samples, is feasible and effective.

基于NVAE和OB-Mix的小样本数据增强方法

由于深度学习模型对海量标注数据的依赖性较高,导致目前许多前沿性目标检测理论难以适用于工业检测领域。为此,提出一种基于NVAE图像生成和OB-Mix数据增强的小样本数据扩充方法。具体方法是通过NVAE构建检测目标的数据分布模型,再通过采样潜变量的方式生成与真实目标图像属于同一分布的全新目标图像。在得到生成目标图像后,提出了OB-Mix数据增强策略,将生成目标图像与背景图像进行随机位置融合以构建出新的图像数据,从而提高网络的定位能力及泛化能力。方法在仅使用474张标注图像以及400张无检测目标的背景图像情况下,使YOLOv5的检测精确率达到95.86%,相比于不使用该方法的结果提高了17.60个百分点。

暗数据视角下高校图书馆数据治理策略研究

研究探讨高校图书馆的数据管理问题可为图书馆的数据应用开拓新思路。文章运用网络调查法与文献分析法回顾了暗数据研究进展,归纳总结出高校图书馆暗数据的内涵、成因、价值与构成,从暗数据视角研究探讨高校图书馆数据治理策略。根据高校图书馆数据的特点构建高校图书馆暗数据治理框架,以建立系统化的数据中心为核心,以专业化的数据馆员、智能化的数据提取和规范化的数据标准为支撑,通过严格的数据质量和数据安全管控,从数据层、保障层、治理层到应用层四个层面,实现对图书馆各类数据全生命周期的统一管理与揭示。

基于多元统计分析的小样本数据预测模型设计

若小样本数据预测误差较大,会直接影响数据应用效果,为提升小样本数据预测精度,提出基于多元统计分析的小样本数据预测模型设计方法。将小样本数据放入SPSS软件中,结合自助法完成小样本数据的经验分布分析。基于样本数据经验分布特征,结合具备学习能力的Fisherface算法对小样本上数据实施预分类,建立测试样本类别标签,实现小样本数据的特征提取。通过多元统计分析数据特征的主元成分,确定模型回归函数,结合支持向量机构建数据预测模型,通过上述模型完成小样本数据的精准预测。实验结果表明,使用上述方法开展小样本数据预测时,预测误差较低,效率较高,说明其预测效果较好。

基于生成对抗网络数据增强的舰炮可靠性分析

舰炮故障数据不足会导致其可靠性分析变得极其困难。为解决故障数据不足这一问题,采用生成式对抗网络(generative adversarial network,GAN)对故障样本进行数据增强。建立GAN数据增强深度神经网络的舰炮可靠性分析模型,并与原始数据进行可靠性分析获得的指标进行了对比。结果表明,利用GAN数据增强后的扩充样本得到的指数分布拟合精度及威布尔分布拟合精度分别提高了5.40%和11.90%,相较于原始数据有了显著提升。为实现舰炮故障数据可靠性分析提供了方法和思路。

Data Mining for Small and Medium Enterprises: A Conceptual Model for Adaptation

The following paper explored data mining issues in Small and Medium Enterprises’ (SMEs), firstly exploring the relationship between data mining and economic development. With SME’s contributing most employment prospects and output within any emerging economy such as the Kingdom of Saudi Arabia. Adopting technology will improve SME’s potential for effective decision making and efficient operations. Hence, it is important that SMEs have access to data mining techniques and implement the most suited into their business to improve their business intelligence (BI). The paper is aimed to critically review the existing literature on data mining in the field of SME to provide a theoretical underpinning for any future work. It has been found data mining to be complicated and fragmented with a multitude of options available for businesses from quite basic systems implemented within Excel or Access to more sophisticated cloud-based systems. For any business, data mining is trade-off between the need for data analysis, and intelligence against the cost and resource-use of the system put in place. Multiple challenges have been identified to data mining, most notably the resource-intensive nature of systems (both in terms of labor and capital) and the security issues of data collection, analysis and storage;with General Data Protection Regulation (GDPR) a key focus for Kingdom of Saudi Arabia businesses. With these challenges the paper suggests that any SME starts small with an internal data mining exercise to digitalize and analyze their customer data, scaling up over time as the business grows and acquires the resources needed to properly manage any system.

基于小扰动频率量测数据的电网节点惯量分布评估方法

惯量在电力系统抵抗扰动中起到关键作用。随着新能源渗透率的提升,电力系统中低惯量与惯量时空分布不均的特征日益凸显,准确评估电网节点惯量分布具有重要意义。为此,文中提出一种基于小扰动频率量测数据的节点惯量分布评估方法。首先,分析了频率与惯量的数学耦合机理,揭示了新型电力系统惯量分布特征。然后,根据单次小扰动下的频率量测数据与系统电源参数,分别提出了考虑各节点频率相较于系统中心频率变化水平偏差的节点等效惯量定义方法,以及基于自适应阶数多项式拟合的节点频率变化率计算方法,两层方法相结合即可计算出单次小扰动下的节点等效惯量。进一步,鉴于单次小扰动下求出的节点等效惯量具有一定的随机误差,提出考虑多次扰动事件评估结果的动态聚合策略,形成适用于新型电力系统的节点等效惯量分布评估方法。最后,分别以IEEE 39节点系统、中国某省级电网为例验证了所提方法的有效性。结果表明,所提方法所需数据量少,仅需各节点同步相量测量单元采集的频率量测数据即可实现对电网惯量分布的有效评估,评估结果准确性高且时效性强。

Kenyan Counties Geospatial Data Knowledge to Monitor Crop Production

Climate change effects have had negative effects on most farmers, both small and large-scale, with weather patterns increasingly becoming unpredictable, such that farmers are unable to plan well for their farming, resulting in reduced harvests and sometimes losses for farmers. Better availability of information such as weather patterns, suitable crops, nutrient requirements based on soil types and conditions would greatly alleviate these challenges. While geospatial information is being developed and improved continuously by researchers, its accessibility and use by the counties has not been established and cannot be identified as contributing to better crop production outcomes. The aim of this study, therefore, was to assess the awareness and status of geospatial data availability and use for crop production, and the level of the relevant capacities, both human and infrastructural, in selected Counties of Kenya. A survey was conducted in the four counties of Vihiga, Kilifi, Wajir and Nyeri and key informant interviews were conducted with both management and technical County Agricultural Officers, as well as sub-county agricultural extension officers. From the results of the survey, out of the four counties, only one has adequate infrastructure in terms of hard-ware, software and connectivity to conduct useful geospatial data acquisition and processing. While most indicated awareness of the existence of geospatial data, limited resources, low skills and knowledge have restricted any meaningful sourcing of and access to data, with only 38% moderately or highly skilled in acquisition, 48% in processing and 57% in interpretation and use of geospatial data. The study concludes that moderate skills and capacities available within the counties have considerable potential to make use the available geospatial data to inform farmers accordingly and improve their farming outcomes.

基于CNN和XgBoost的香蕉成熟度判别

目的:提高香蕉成熟度的判别准确率。方法:基于卷积神经网络和极限梯度提升算法建立香蕉成熟度的判别方法。先通过卷积神经网络提取香蕉图像特征,并采用全连接层网络和线性判别分析方法精简香蕉图像特征;通过贝叶斯优化算法优化极限梯度提升算法超参数;将简化后的香蕉图像特征输入极限梯度提升算法,通过极限梯度提升算法对香蕉成熟度进行判别。结果:所提方法对香蕉成熟度的判别准确度为91.25%;与已有方法相比,所提方法对小数据量香蕉的成熟度判别准确率明显提高。结论:该方法可实现被测香蕉成熟度的准确判别,有助于仓库经理、出口商实时监测香蕉的成熟度状况。

小数据视角下知识服务的体系建构与伦理反思

知识服务智慧化是国家教育和文化数字化战略的重要组成部分,在近年来的大数据技术热潮下,研究者亟须冷静思考小数据的价值。文章首先分析了潜在期、成长期、成熟期和衰退期四类小数据用户,阐释了其个体独特性、实时动态性和适度可控性三方面特征,归纳了包括静态数据、动态数据以及第三方数据在内的三类小数据来源及结构。其次在数据与知识双轮驱动的第三代人工智能思想指导下,以知识图谱为基石,以用户画像为核心,构建了小数据视角下的知识服务体系,提出新型的知识服务需融合多模态数据提高用户画像的精准性,以知识图谱为引擎促进知识管理的认知性,坚持应用为王增加知识服务的多样性。最后,从价值观、正义观、自由观三方面审视小数据视角下知识服务面临的伦理挑战。

小样本下基于改进ACGAN数据增强的X射线矿石图像分类方法

针对工业领域利用深度学习模型对矿石进行在线分类时,矿石样本稀少导致的模型过拟合、分类准确率低的问题,提出一种结合X射线透射成像技术的矿石数据增强分类方法。该方法基于改进辅助生成对抗网络(Enhance-based Classification ACGAN-gp, EC-ACGAN-gp),采用卷积和连续残差块构建判别器和生成器,引入注意力机制捕捉矿石细节特征,生成高质量样本扩充原始数据集,同时使用带梯度惩罚的Wasserstein距离重构判别器的损失函数提高对抗训练的稳定性,避免模式崩溃。通过增加辅助分类器重建样本标签信息,最终实现矿石样本的类别预测。结果表明,该方法能实现矿石品位分类的精准预测,准确率可达89.62%,比现有传统方法提高3.98%。该模型生成的矿石样本泛化性良好,能够显著提高小样本数据集的泛化性,在SVM、LeNet5、VGGNet、ResNet上测试,精度分别提升了2.83%、2.36%、1.89%和3.74%,可进一步用于提升其他分类模型在矿石品位预测方面的性能。