A PID-incorporated Latent Factorization of Tensors Approach to Dynamically Weighted Directed Network Analysis

A large-scale dynamically weighted directed network (DWDN) involving numerous entities and massive dynamic interaction is an essential data source in many big-data-related applications, like in a terminal interaction pattern analysis system (TIPAS). It can be represented by a high-dimensional and incomplete (HDI) tensor whose entries are mostly unknown. Yet such an HDI tensor contains a wealth knowledge regarding various desired patterns like potential links in a DWDN. A latent factorization-of-tensors (LFT) model proves to be highly efficient in extracting such knowledge from an HDI tensor, which is commonly achieved via a stochastic gradient descent (SGD) solver. However, an SGD-based LFT model suffers from slow convergence that impairs its efficiency on large-scale DWDNs. To address this issue, this work proposes a proportional-integral-derivative (PID)-incorporated LFT model. It constructs an adjusted instance error based on the PID control principle, and then substitutes it into an SGD solver to improve the convergence rate. Empirical studies on two DWDNs generated by a real TIPAS show that compared with state-of-the-art models, the proposed model achieves significant efficiency gain as well as highly competitive prediction accuracy when handling the task of missing link prediction for a given DWDN.      

Parameter identification for Wiener-finite impulse response system with output data of missing completely at random mechanism and time delay

In this paper, the parameter estimation issue of Wiener system with random time delay and missing output data is studied. The linear part of Wiener system is described by Finite Impulse Response (FIR) model. The mathematical formula of the Expectation Maximum algorithm to identify Wiener-FIR system that contains the random time delay and the nonlinear output data in missing completely at random mechanism is derived, which is never considered before. To obtain the unmeasurable intermediate variable in Wiener-FIR system, the idea of auxiliary model is adopted. The time delay and system parameters can be estimated simultaneously by this method. Numerical example and the identification of water tank system example are carried out, the effectiveness of the algorithm is proved.     

基于运行协变量分析的机床主轴备件预测

机床的故障率取决于固有可靠性,运行环境影响因素虽然不会改变其故障规律,但是一定程度上会推进或者延迟故障发生的周期,因此在进行机床备件预测时考虑运行环境影响要素显得尤为重要。通过对某型号机床主轴运行环境分析,筛选主要影响协变量,基于数据独立同分布的假设检验,采用PHM回归模型建立圆柱滚子轴承可靠性模型。基于此,利用更新模型估计故障发生点,参考备件保有率,建立主轴备件的需求预测模型。计算结果表明:考虑环境要素进行备件预测更为精确。该研究成果对机床不可修复备件预测在理论研究方面具有参考价值。     

浅谈清单实体工程量对工程造价的影响

从合同管理的角度对清单实体工程量变化的原因进行了分析，列出了清单实体工程量存在的调整情况，论述了招标阶段和施工阶段对清单实体工程量的调整原则，以保证工程量的准确性和全面性，从而合理控制工程造价。     

弓形缺失水平井井孔声场数值模拟

针对水平井中可能存在的水泥环上层弓形缺失,通过建立不同弓形缺失高度、弓形缺失部分为混浆等模型,利用交错网格有限差分法求解应力—速度的一阶偏微分方程组,实现水平井弓形缺失复杂井孔声场的数值模拟。模拟结果显示,弓形缺失的弓高影响声场首波幅度和旅行时间,弓高低于水泥环厚度30mm首波为第Ⅱ界面特征波,高于30mm首波为套管波;弓形缺失部分为混浆对声场波幅和首波旅行时间影响不是特别明显,但是与内部为水的声场相比,首波幅度明显减小;当水泥环存在弓形缺失时,靠近缺失部位的接收器接收到的首波幅度明显减小,首波旅行时间则明显增大,依此可以判断缺失方位。     

Data‐driven model for river flood forecasting based on a Bayesian network approach

Uncertainty analysis of hydrological models often requires a large number of model runs, which can be time consuming and computationally intensive. In order to reduce the number of runs required for uncertainty prediction, Bayesian networks (BNs) are used to graphically represent conditional probability dependence between the set of variables characterizing a flood event. Bayesian networks (BNs) are relevant due to their capacity to handle uncertainty, combine statistical data and expertise and introduce evidences in real‐time flood forecasting. In the present study, a runoff–runoff model is considered. The discharge at a gauging station located is estimated at the outlet of a basin catchment based on discharge measurements at the gauging stations upstream. The BN model shows good performances in estimating the discharges at the basin outlet. Another application of the BN model is to be used as a reverse method. Knowing discharges values at the outlet of the basin, we can propagate back these values through the model to estimate discharges at upstream stations. This turns out to be a practical method to fill the missing data in streamflow records which are critical to the sustainable management of water and the development of hydrological models.     

Candidate Gene Discovery in Hereditary Colorectal Cancer and Polyposis Syndromes–Considerations for Future Studies

To discover novel high-penetrant risk loci for hereditary colorectal cancer (hCRC) and polyposis syndromes many whole-exome and whole-genome sequencing (WES/WGS) studies have been performed. Remarkably, these studies resulted in only a few novel high-penetrant risk genes. Given this observation, the possibility and strategy to identify high-penetrant risk genes for hCRC and polyposis needs reconsideration. Therefore, we reviewed the study design of WES/WGS-based hCRC and polyposis gene discovery studies (n = 37) and provide recommendations to optimize discovery and validation strategies. The group of genetically unresolved patients is phenotypically heterogeneous, and likely composed of distinct molecular subtypes. This knowledge advocates for the screening of a homogeneous, stringently preselected discovery cohort and obtaining multi-level evidence for variant pathogenicity. This evidence can be collected by characterizing the molecular landscape of tumors from individuals with the same affected gene or by functional validation in cell-based models. Together, the combined approach of a phenotype-driven, tumor-based candidate gene search might elucidate the potential contribution of novel genetic predispositions in genetically unresolved hCRC and polyposis.     

College students’ academic motivation,media engagement and fear of missing out

The concerns about the consequences of mental problems related to use of social media among university students have recently raised consciousness about a relatively new phenomenon termed Fear of Missing Out (FoMO). Drawing on the self-determination theory and on the assumption that low levels of basic need satisfaction may relate to FoMO and social media engagement, the aim of the present research was to examine for the first time possible links between FoMO, social media engagement, and three motivational constructs: Intrinsic, extrinsic and amotivation for learning. Data were gathered from 296 undergraduate students by using the following scales: Social Media Engagement (SME), Fear of Missing Out (FoMOs) and Academic Motivation. The SME is a new scale, specifically designed for this study to measure the extent to which students used social media in the classroom. This scale includes three categories: Social engagement, news information engagement and commercial information engagement. Path analysis results indicated that the positive links between social media engagement and two motivational factors: Extrinsic and amotivation for learning are more likely to be mediated by FoMO. Interpretation of these results, their congruence within the context of the theoretical frameworks and practical implications are discussed.     

基于人体关节点数据的攻击性行为识别

针对人体攻击性行为识别问题，提出一种基于人体关节点数据的攻击性行为识别方法。首先，利用OpenPose获得单帧图像中的人体关节点数据，并通过最近邻帧特征加权法和分段多项式回归完成由人体自遮挡和环境因素所导致缺失值的补全；然后，对每个人体定义动态"安全距离"阈值，如果两人真实距离小于阈值，则构建行为特征矢量，其中包括帧间人体重心位移、人体关节旋转角角速度和发生交互时的最小攻击距离等；最后，提出改进的LightGBM算法w-LightGBM，并对攻击性行为进行识别。采用公共数据集UT-interaction对所提出的攻击性行为分类识别方法进行测试实验，准确率达到95.45%。实验结果表明，所提方法能够有效识别各种角度的攻击性行为。     

基于CART的社区矫正人员危险性评估

目前对社区矫正人员危险性评估的研究普遍停留在理论研究阶段，建立的模型大多缺乏实际数据的支撑，为了方便掌握社区矫正实际情况，基于CART（分类回归树）算法对某市司法局的社区矫正人员数据（不包含敏感信息）进行建模分析，通过服刑人员的基础特征值得到该服刑人员的危险度评估结果，与实际结果相比，准确率达到85%~90%，此外对社区矫正中实际存在的数据缺失问题进行讨论并提出解决办法。