结合语义知识的汉语词义消歧

词义消歧一直是自然语言处理领域中的关键性问题。为了提高词义消歧的准确率,从目标歧义词汇出发,挖掘左右词单元的语义知识。以贝叶斯模型为基础,结合左右词单元的语义信息,提出了一种新的词义消歧方法。以SemEval-2007：Task#5作为训练语料和测试语料,对词义消歧分类器进行优化,并对优化后的分类器进行测试。实验结果表明：词义消歧的准确率有所提高。     

城市轨道交通设备故障聚类与贝叶斯网络预警

针对城市轨道交通的海量监控报警数据分析难度大的问题,结合设备故障特征,应用数据挖掘方法定义聚类特征树,建立基于故障集的层次聚类分析模型,取得故障要素的高频聚类并识别关联规则。通过构建故障贝叶斯层次网络,提出故障预警模型,计算取值概率推断预警值,进一步探讨了关联性预警的识别和应用策略。采用某城市轨道交通的阶段报警数据实例验证,表明识别出的故障关联规则与实际较好地符合,利用该方法推断故障预警能快速定位风险隐患,为安全管理提供有效的决策支持。     

基于滑动窗口均值先验的非同构动态贝叶斯网络转换点检测算法*

为了放宽动态贝叶斯网络中的同构假设,提出非同构贝叶斯网络.基于此种情况,文中提出结合先验知识的可逆跳转的马尔可夫链蒙特卡洛算法(APK-RJ-MCMC).算法基本假设为如果一个时间点左右窗口中数据均值间的欧氏距离越大,则这个时间点作为转换点的可能性越高.基于上述假设,可得到关于每个时间点作为转换点可能性的粗略估计,将其作为先验知识调控可逆跳转的马尔可夫蒙特卡洛采样技术(RJ-MCMC)采样转换点时的生成、消除、转换动作的提议概率之比,进而调节状态跳转时的接受概率.在人工数据集和基因数据集上的实验表明,相比其它算法,APK-RJ-MCMC在转换点检测上具有更高的检测后验概率.     

基于分期组合预报模型的水电站优调度及其模糊风险分析

针对年内不同时期径流成因与变化规律的差异,本文建立了月径流的分期组合预报模型,对不同时期径流成因特点选用相应的组合预报模型.针对过渡期与丰水期径流预报不确定件较大的特点,采用贝叶斯平均模型进行组合预报,能够发布概率预报.为量化预报误差对水电站发电调度造成的影响,建立了水电站预报优化调度的模糊风险分析方法,从而为科学决策提供参考依据.实例应用表明,分期组合预报模型优于单一模型,且发布的概率预报结合模糊风险分析对制定水电站的月测度计划具有一定的参考价值.     

基于稀疏贝叶斯学习的电力系统暂态稳定评估

介绍了基于稀疏贝叶斯学习理论的模式识别技术相关向量机及其分类器,在此基础上构建了电力系统暂态稳定评估模型.以EPRI36电力系统暂态稳定仿真数据为例,在相同的数据输入和相同的仿真环境下同时构建相关向量机和支持向量机2种暂态稳定评估模型.仿真预测计算显示,作为一种全新的概率学习模型,相关向量机不仅得到了比支持向量机更高的预测精确度,而且还能得到支持向量机无法完成的概率性预测和更高的稀疏性计算.     

电力竞价上网的博弈模型与算法

假设每个发电企业对上网电价进行密封报价、统一时间开标,报价低的一方中标,提出2个发电企业竞价上网的静态贝叶斯博弈模型与算法,当每个发电企业均采用线性策略时,分析了模型的收敛性和贝叶斯纳什均衡解的存在性;然后对2个发电企业竞价上网模型进行扩展,研究了多个发电企业竞价的贝叶斯纳什均衡,建立了多个发电企业竞价上网博弈模型.最后进行了计算机系统仿真,结果表明:所建立的模型随着企业数的增加,发电企业的报价更加趋于理性,逐渐趋近于预测值.     

基于贝叶斯网络的电容型设备故障诊断方法

提出了基于贝叶斯网络的电容型设备故障诊断方法,阐述了贝叶斯网络的构建过程.通过广泛收集有关电容型设备的故障资料,综合其各种检测数据和故障征兆,获得了较为全面的故障集和征兆集.经过对数据的统计分析获得了各故障类型下各征兆量有明显体现的条件概率,在此基础上建立了基于贝叶斯网络的电容型设备故障诊断模型,并根据电容型设备故障诊断的特点改进了贝叶斯网络的推理过程,采用连概率计算过程进行故障类型的概率信息计算,根据概率信息进行故障分类,提高了该方法的实用性.通过电容型设备故障实例分析,诊断结果验证了该方法的正确性和有效性.     

基于贝叶斯网络的发动机曲柄连杆机构可靠性分析

曲柄连杆机构是发动机中重要的组成部分,其工作状态直接影响发动机的输出性能.为评估其可靠性,识别系统中的薄弱环节,基于贝叶斯网络提出了一种曲柄连杆机构可靠性评估方法.在运用故障树进行分析的基础上进一步将故障树映射为贝叶斯网络,给出了故障树向贝叶斯网络的映射方法,融合各验前信息计算各底事件和顶事件的发生概率,分析各个底事件对系统整体可靠性的影响,并计算各事件的重要度指标,通过分析查找系统的薄弱环节,以应用实例对提出方法的有效可行性进行了验证.     

Feature extraction based on sparse representation with application to epileptic EEG classification

Epilepsy seizure detection in electroencephalogram (EEG) is a major issue in the diagnosis of epilepsy, and it can be considered as a classification problem. Considering the particular property of EEG, which is sparse in Garbor dictionary, a feature extraction method based on sparse representation has been applied to epilepsy detection. To improve classification accuracy, in this article, a novel feature vector is developed, which not only can reflect the main structure, but also can give expression to the relation between main structure and residual information. Classification accuracy, efficiency, and robustness to noise of the new feature are explored and analyzed with publicly available data set. It is demonstrated by experiments that the classification accuracy and the efficiency are simultaneously enhanced with this new feature extraction method, and that the novel classification feature proposed in this work greatly improves the classification performance of epilepsy detection. © 2013 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 23, 104–113, 2013     

Sequential Monte Carlo-based fidelity selection in dynamic-data-driven adaptive multi-scale simulations

In a simulation-based planning and control framework, timely monitoring, analysis, and control is important not to disrupt a dynamically changing system. To meet this temporal requirement, a dynamic-data-driven adaptive multi-scale simulation (DDDAMS) paradigm was proposed earlier, where the fidelity of a complex simulation model adapts to available computational resources by incorporating dynamic data into the executing model, which then steers the measurement process for selective data update. In this work, a sequential Monte Carlo method (sequential Bayesian inference technique) is proposed and embedded into the simulation to enable its ideal fidelity selection given massive datasets under the DDDAMS framework. As dynamic information becomes available, the proposed method makes efficient inferences to determine the sources of abnormality in the system (a shop floor in this paper). A parallelisation framework is also discussed to further reduce the number of data accesses while maintaining the accuracy of parameter estimates. A prototype DDDAMS involving the proposed algorithm has been implemented successfully for preventive maintenance scheduling and part routing scheduling in a semiconductor manufacturing supply chain, reducing the average waiting time of batches and increasing the machine utilisation significantly.