融合节点状态信息的跨社交网络用户对齐

提出一种融合节点状态信息的跨社交网络用户对齐方法, 通过网络表示捕获节点的局部特征和节点状态信息得到每个账户的嵌入向量, 计算不同账户对应表示之间的相似性发现对齐用户。在2个真实数据集上的试验结果表明, 提出的方法相对于其他方法可以对齐更多的用户。在预测不同尺度的top-k时, 提出的方法在网络结构较稠密的Twitter-Foursquare数据集上能够在top-9时对齐准确率达到50%且在稀疏且大网络数据集DM-ML上相比其他方法对齐准确率提高12.06%~36.62%;在分析F1-score时, 提出的方法能够有效提高用户对齐的性能。     

一种新的知觉分割模型*

根据知觉的最新理论研究和实验结果,提出一种新的知觉模型,并用于考查计算机视觉中的分割问题.分割问题实质上是知觉表达问题,现在分割的困难性可能是知觉上局部到整体的还原论假设造成的;分割可能不是一次完成的,而是经过整体到局部的过程才最终完成.新的知觉模型将对建立更好的分割模型提供理论基础.     

基于用户行为特征的P2P代理缓存的研究*

通过对BitTorrent用户一段时间内的行为统计分析,把握用户的行为特征.在此特征的基础上,提出一种BitTorrent缓存模式,更大程度地利用缓存空间,提高缓存命中率,减轻网络运营商的出口压力.     

基于非侵入式负荷监测的用户行为精细化分析

随着非侵入式电力负荷监测的发展,如何利用用户负荷用电细节信息,分析用户用电行为特征成为重要的研究方向。本文提出了基于非侵入式负荷监测的用户行为精细化分析方法,根据历史数据建立用户行为特性指标,提出基于因子分析与支持向量机的用户聚类方法。在此基础上,建立基于精细化负荷数据的用户需求响应潜力评估指标。算例结果表明,所研究内容可以准确分析用户行为,为电力公司施行需求侧管理提供科学指导。     

基于Stacking模型融合的专变用户电费回收风险识别方法

针对当前电力公司面临的专变用户电费回收风险,提出一种基于Stacking模型融合的专变用户电费回收风险识别方法。对专变用户数据进行特征处理、特征构造与特征筛选,从样本分布和特征属性上优化模型的泛化性能;利用Stacking模型融合多个基学习器,构建专变用户电费回收风险识别模型。实验结果表明,相较于其他常用的分类算法,所提方法具有更优的精确率、召回率、P-R调和均值、AUC值以及模型泛化性能,对专变风险用户的识别率也更高。     

Dynamics of fluid circulation in coupled free and heterogeneous porous domains

Fluid flow in coupled free and porous domain, particularly when the porous medium is heterogeneous, is encountered in many hydro-environmental conditions, e.g., leakage from underground pipe, combined groundwater lake-subsurface interactions. One of the most difficult problems in the study of coupled flow behaviour has been the development of a universally applicable modelling scheme for combining the flow regimes. This is because the free/porous interfacial properties (e.g., shear-stress; velocity slip) that govern the coupled flow behaviour are difficult to determine experimentally under hydro-environmental conditions. On the other hand, the implications of various forms of heterogeneity in the porous media properties can be very different on the fluid-flow behaviour. Difficulties may also arise in direct coupling of the model equations that govern the fluid flow in the individual regions (e.g., Navier-Stokes for free-flow region and the Darcy's equation for the porous flow region). Consequently, models of coupled free and porous flow for hydro-environmental conditions are not very well developed at the moment. While there are some indications that fluids in coupled free and porous domains may circulate (i.e., development of flow cells), there is a lack of appropriate 3D analysis on how heterogeneities in porous media may affect such flow patterns. In this paper, we aim to analyse how porous media heterogeneity affects the dynamics of flow circulation in the porous side of a coupled free and porous domain. For this purpose, we analyse flow patterns in several model domains made up of two porous layers with differing permeabilities. The governing model equations are discretised and solved using the standard finite volume method on a staggered cell-centred mesh. The temporal discretisation is done using the explicit method. An in-house graphical user interface (GUI) has been created specifically to aid in the visualisation of otherwise complex flow patterns. The GUI contains many post-processing options and provides a comprehensive tool for the analysis of hydrodynamics and contaminant motion (not discussed in this paper) in coupled free and porous flow domains. This GUI is described in this paper briefly. The effects of altering the aspect ratio (i.e., multi-scale) of the domain on the coupled flow pattern have also been discussed.     

Augment low-field intra-operative MRI with preoperative MRI using a hybrid non-rigid registration method

Background

Preoperatively acquired diffusion tensor image (DTI) and blood oxygen level dependent (BOLD) have been proved to be effective in providing more anatomical and functional information; however, the brain deformation induced by brain shift and tumor resection severely impairs the correspondence between the image space and the patient space in image-guided neurosurgery.

Method

To address the brain deformation, we developed a hybrid non-rigid registration method to register high-field preoperative MRI with low-field intra-operative MRI in order to recover the deformation induced by brain shift and tumor resection. The registered DTI and BOLD are fused with low-field intra-operative MRI for image-guided neurosurgery.

Results

The proposed hybrid registration method was evaluated by comparing the landmarks predicted by the hybrid registration method with the landmarks identified in the low-field intra-operative MRI for 10 patients. The prediction error of the hybrid method is 1.92 ± 0.54 mm, and the compensation accuracy is 74.3 ± 5.0%. Compared to the landmarks far from the resection region, those near the resection region demonstrated a higher compensation accuracy (P-value = .003) although these landmarks had larger initial displacements.

Conclusions

The proposed hybrid registration method is able to bring preoperatively acquired BOLD and DTI into the operating room and compensate for the deformation to augment low-field intra-operative MRI with rich anatomical and functional information.     

基于海量搜索历史数据的用户兴趣模型

针对互联网搜索引擎环境中,基于海量搜索历史数据分析用户兴趣的问题,提出一种改进的用户兴趣模型。该模型根据用户搜索的历史数据,结合向量空间模型（ VSM ）和TF-IDF算法,递归地回溯出用户兴趣权重列表。为解决用户兴趣变化和时间性能的问题,该模型引入时间遗忘机制进行动态更新,并在Hadoop分布式系统架构下利用MapReduce分布式编程模型进行实现。实验结果表明,改进的用户兴趣模型的查准率和召回率都能达到50',具有较好的可行性和可用性。     

基于遗忘曲线的微博用户兴趣模型

为解决微博用户兴趣漂移问题,以人类记忆学中遗忘曲线为基础,提出一种微博用户兴趣模型,利用用户历史信息预测当前兴趣。在预测过程中,用户关注某信息的时间距离当前时间越远,该信息越容易被遗忘,其对用户当前兴趣的影响越小;用户关注某一领域的信息越多,印象越深刻,对该领域的兴趣度越高。这两点与人类对知识逐渐遗忘和重复学习的过程具有高度相似性,因此该模型预测准确性更高。实验结果表明,该模型能较好地预测微博用户兴趣,召回率可达85.3％,实用性较强。     

红细胞瞬间磁化试剂的制备与应用

目的研制用于ABO、RHD血型检测试剂的红细胞瞬间磁化试剂。方法采用共沉淀法制备磁性粒子,并检测其主要成分及粒径大小,用此磁性粒子制备红细胞瞬间磁化试剂,并分别进行各项检测。结果磁性粒子的主要成分为四氧化三铁,粒径100nm左右。该试剂的外观及特异性均合格,磁响应时间不超过30s,重复性较好,敏感性为98.20%,与试管法检测结果比较,差异无显著意义。结论红细胞瞬间磁化试剂可用于红细胞系统的血型检测。