中文非投射语义依存现象分析研究

汉语是一种语序灵活的语言,句子变式很多,基于传统依存树的投射现象还不能很好解决某些句式的语义理解问题。文章以10000个句子的汉语语义依存图库为基础,验证并明确了汉语非投射现象的客观存在性,考察了汉语句子中存在的非投射现象,并从语言学和句子深层语义理解的角度对非投射现象进行了归纳和解释。文章总结了7类出现非投射现象的情况,包括小句宾语句、比较句、主谓谓语句、紧缩复句、代词、动补谓语句以及注释短语或复句。这对于自动语义依存标注有重要的指导作用。     

基于WINDOWS环境下色斑图制作方法

根据气象要素地理平面分析的需求,在WINDOWS操作系统环境下采用B/S架构的方式,独立绘制了宁夏区域气象要素地理分布色斑图。程序采用模块化设计,便于开发、修改和扩充。目前程序已投入业务运行,运行状态良好。     

Bidirectional Label Propagation over Graphs

Graph-Based label propagation algorithms are popular in the state-of-the-art semi-supervised learning research. The key idea underlying this algorithmic family is to enforce labeling consistency between any two examples with a positive similarity. However, negative similarities or dissimilarities are equivalently valuable in practice. To this end, we simultaneously leverage similarities and dissimilarities in our proposed semi-supervised learning algorithm which we term Bidirectional Label Propagation (BLP). Different from previous label propagation mechanisms that proceed along a single direction of graph edges, the BLP algorithm can propagate labels along not only positive but also negative edge directions. By using an initial neighborhood graph and class assignment constraints inherent among the labeled examples, a set of class-specific graphs are learned, which include both positive and negative edges and thus reveal discriminative cues. Over the learned graphs, a convex propagation criterion is carried out to ensure consistent labelings along the positive edges and inconsistent labelings along the negative edges. Experimental evidence discovered in synthetic and real-world datasets validates excellent performance of the proposed BLP algorithm.     

高维数据挖掘中基于中位数回归的特征提取新方法

为降低噪声对数据特征提取(变量选择)效果的不利影响,基于中位数回归分析方法,利用变量选择降维技术(正则化估计),提出了一种稳健、有效的特征提取(变量选择)新方法,并具体给出了估计算法,该算法具有快速计算的特点.实验结果表明,新方法能够有效地对高维数据集进行估计和变量选择,且具有较高的准确性,即使数据中的信噪比很低时,该方法仍具有较好的效果.因此,该方法为高维数据挖掘特征提取提供了稳健且有效的方法.     

基于线性CCD的寻线智能车设计

本设计以第八届"飞思卡尔"杯全国大学生智能车竞赛为背景,提出了一种根据线性CCD采集图像引导直立小车循迹行驶的方案。本文介绍了这一方案的基本思想,所依据的物理原理,并重点介绍在双速度控制算法下对转向及障碍问题的优化处理,并根据方案实际制作了小车。实践证明该方案是可行的,并且效果较好。     

Dipole moments of copolymers in relation to the distribution of polar monomer unit sequences

A method for evaluating the dipole moments of copolymers in relation to the distribution of polar unit sequences is proposed. Experimental data on the synthesis and dipole moments of styrene-p-chlorostyrene and styrene-p-methoxystyrene copolymers were used to calculate the effective dipole moment of a polar unit in a copolymer. For this purpose, we assumed the following effective moment values for the corresponding polar unit sequences along the copolymer chains: when both of the nearest neighbours are polar, μ_PP, when one of the nearest neighbours is polar and the other is non-polar, μ_PS, and when both of the nearest neighbours are non-polar, μ_SS. The relative magnitudes of μ_SS, μ_PS and μ_PP calculated for both copolymers reflect the influence of dipole-dipole interactions on the effective dipole moment of a polar unit in a copolymer molecule.     

四元相图在复分解法制硝酸钾中的应用

在硝酸铵和氯化钾复分解反应制硝酸钾和氯化铵多温相图中，由于氯化铵的结晶区随温度变化而变化不大，导致欲获得较纯的氯化铵与硝酸钾十分困难。本实验通过加入大量与本盐互体系无关的酸、碱、盐，或有机物质，使上述反应的共饱和曲线发生较大漂移，使氯化铵的结晶区随温度化而变化明显增大，结晶曲线增长，从而在相同条件下，能顺利地分离出较纯的氯化铵与硝酸钾粗品。硝酸钾粗品经过重结晶一次可达到或超过GB／T1918—1998标准的优级品。     

网络爬行虫智能化研究分析

Web爬行虫是当今搜索引擎的主要组成部分,也是信息检索领域研究的热点问题。本文综述了具有一定智能性的网络爬行虫的研究历史与现状,主要包括两个方面:传统的人工智能方法如神经网络、遗传算法、蚁群算法等在网络爬行虫的应用,以及借助这些方法发展起来的主题爬行虫;多网络爬行虫系统中爬行虫的协调的Agent技术。在此基础上,提出了一个语义概念背景图的网络爬行的基本思路。     

图割综述

综述了图割的基本理论框架及利用图割解决问题时的主要流程和图割方法存在的难点问题.介绍了图割的发展状况及图割的应用领域,指出了其中的公开问题,并对图割未来的研究方向做出展望.     

构建本质图的改进算法*

利用图论相关知识,提出了一种构建本质图的改进算法。数值实验表明,对于V-结构较多的非循环有向图,改进算法能相对较快地确定出其本质图,从而能更好地学习贝叶斯网络的等价类。