NL句法分析中超语法符合现象的处理

本文回顾并总结了句法层次上六种不同类型的超语法符合(extra grammaticality,简称EG)现象及它们的恢复原则, 描述了如何用扩充的GLR分析器(Extended GLR Parser简称EGLR)完整解决其中四种、部分解决其余两种现象的技术。在处理后两种现象时, 要涉及语义问题。     

基于GLR算法的维吾尔语句法分析研究

采用GLR算法对维吾尔语句子进行句法分析,并且与平行LR算法进行比较,比较它们的分析过程,针对于维吾尔语在单词集上进行句法分析。分析结果采用最优规则,选取最优的句法分析树,对下一步的句法分析研究提供很大的帮助。     

一种3G中基于三层数据库的呼叫建立机制研究

基于3G网络三层数据库结构中GLR (Gateway Location Register)服务区远大于VLR (Visitor Location Register)服务区的情况,考虑被呼业务本地化(即本地移动用户呼叫本地移动用户的比例占总呼叫数的比例大),通过在GLR中设置缓存的方法,给出了一种新的呼叫建立机制。建立分析模型得到新机制的总代价函数和总时间延迟函数,并同基本呼叫建立机制以及2G网络基于VLR缓存的呼叫建立机制进行了对比分析,验证了新呼叫建立机制的优越性。     

多分辨率特征在UWB SAR目标检测中的应用

本文借鉴高频合成孔径雷达(SAR)目标检测中多分辨率特征分析方法,结合超宽带合成孔径雷达(UWB SAR)工作体制和树干杂波的特殊性,详细分析了UWB SAR目标检测中树干杂波和目标多分辨率特性的区别;构造了多分辨率差分序列,并采用一阶AR模型实现了对多分辨率差分序列的建模;给出了基于多分辨率的广义似然比(GLR)定义,并提出了利用感兴趣区域(ROI)内GLR之和来确定ROI真假属性的判别方法;针对树干杂波和目标的模型训练分别提出了预先训练和现场训练的方法,大大提高了多分辨率特征提取的稳健性;基于实际UWB SAR图像数据的试验结果证明了利用多分辨率特征降低ROI虚警率的有效性。     

3G中增强的3层动态位置管理建模与分析

基于3G网络中由HLR(home location register)、GLR(gateway location register)、VLR(visitor location register)构成的3层数据库结构,提出了基于移动的增强动态位置管理策略,重点分析了该策略下位置更新和寻呼的代价模型。并使用该代价模型对3G网络中基于移动的增强动态位置管理策略、基于移动的基本动态位置管理策略和静态位置管理策略进行了评价,计算结果显示增强动态位置管理策略要优于基于移动的基本位置管理策略和静态位置管理策略。     

Effect of ultrasound on bubble breakup within the mixing chamber of an effervescent atomizer

Gas bubbles introduced into a liquid in the mixing chamber help to break up the liquid into fine droplets on being expanded to the ambient pressure. The passage of gas bubbles through the orifice of the nozzle requires that the size of the bubbles be much smaller than the diameter of the orifice. In the present work, the effectiveness of 20 kHz ultrasound to increase number density of fine bubbles within the mixing chamber of an effervescent atomizer by breaking up bubbles introduced in it by an aerator was investigated. Bubbles of initial size in the range of 5-10 mm were shown to get disintegrated into clusters of micron and sub-micron sized bubbles. A fine spray was produced in the presence of ultrasound at a gas-to-liquid mass flowrate ratio (GLR) of 0.063%. The half-cone angle of spray was in the range of 6-10°, which compares favorably with conventional atomizers. The experimental findings of bubble breakup were theoretically modeled by the Rayleigh-Plesset equation. The results of the model indicate that bubbles having initial radius less than 3 mm undergo growth and subsequent disintegration at 20 kHz for the given acoustic pressure of 0.3 MPa.     

A Kullback-Leibler information control chart for linear profiles monitoring

Profile monitoring is referred to as monitoring the functional relationship between the response and explanatory variables. Traditionally, process control charts monitor the mean and/or the variance of a univariate quality characteristic. For several correlated quality characteristics, multivariate process control charts monitor the mean vector and/or the covariance matrix. However, monitoring the functional relationship between variables is sometimes more beneficial. One example is the power output of a Diesel engine and the amount of fuel injected should be related. In this paper, we propose a Kullback-Leibler information (KLI) control chart for linear profiles monitoring in Phase II. The plotted statistics of the KLI chart are calculated based on a backward procedure. The functional relationship is described by linear regression. The performance of the proposed KLI control chart is compared with those of other existing control charts, especially the Generalized Likelihood Ratio (GLR) chart for both KLI and GLR charts do not require design parameters. The results show that (1) the KLI control chart is better than the GLR control chart in detecting the shift of variance in terms of Average Time to Signal, and (2) if the shift of the regression coefficient is small, the GLR chart outperforms the KLI chart, but if the size of shift is large, the KLI chart outperforms the GLR chart. The plotted statistics of KLI are compared to that of GLR. Their similarity is discussed.     

基于改进GLR算法的智能识别英语翻译模型设计

GLR算法模型翻译识别结果存在数据点重合的情况,精确度无法得到有效保障。为了准确的识别短语,设计了基于改进GLR算法的短语智能识别算法,该算法构建标记规模约74万个英汉单词的短语语料库,使短语具备可搜索功能,通过短语中心点构建短语结构,可获得词性识别结果,依据解析线性表的句法功能校正词性识别结果中的英汉结构歧义,最终获得识别的内容。实际测评结果显示,该算法克服了GLR的弊端,相对统计算法和动态记忆算法提高了运算速度和处理性能,更加适合机器翻译任务,为在智能机器翻译领域提供了新的思路。     

基于GLR算法的新型概率句法分析器

为了提高句法分析器的分歧能力和分析准确率,结合上下文无关概率模型PCFG和头驱动概率模型HDSM各自的优点,提出了一种新型的概率模型PCFG_HDSM,并基于GLR算法,实现了一个新型的汉语句法分析器。在词性标注阶段,通过对助词的详细标注使部分歧义在规则阶段就被去除掉,提高了系统消歧的能力。经过开放测试,准确率和回归率分别达到82.8%、74.7%,与其他分析器分析结果比较有了较大提高,证明新模型PCFG_HDSM确实提高了分析器的分歧能力。     

一种改进的基于说话者的语音分割算法

语音分割是语音识别和语音文档检索等众多语音应用的基础.提出一种改进的基于说话者的语音分割算法,对GLR和BIC相结合的算法作进一步的改进:(1) 基于GLR距离方差的自适应阈值调整算法改进了不同声学特征下基于距离的语音分割算法中的阈值选取方法;(2) 引入BIC可测度概念来度量其适用范围;(3) BIC信息准则校准非冗余的候选分割点的偏差.实验结果表明,此改进算法优于原算法.