描述逻辑系统εLN中的概念包含推理算法研究

首先我们在F Baader给出的εL概念描述树基础上给出了εLN中概念描述树的定义,然后定义了εLN中概念描述树之间的同态关系,最后通过计算两棵概念描述树之间的同态给出了εLN中概念之间的相互包含关系算法。     

二次插值对称点算法求解一维搜索问题

插值法是一类用插值多项式来逼近未知或复杂函数的方法。本文基于二次插值,将插值多项式的极小点和其对称点作为搜索区间的两个探索点,通过不断缩小搜索区间,求解一维搜索问题的最优解。本文给出了二次插值对称点的算法,并用0.618法进行了数值比较。结果表明,新算法比0.618法效果好。     

基于蚁群算法在交通领域的应用综述

蚁群算法是受现实蚂蚁群体行为启发而得出的一类仿生算法。从蚁群算法与实际交通问题的求解有很强直接对应特性出发,对蚁群算法在交通领域的应用进行全面综述。蚁群算法也可应用于交通过程建模、交通过程优化及导航、交通运输规划等交通领域。     

基于改进小生境遗传算法的自适应组卷系统

针对传统基于简单遗传算法的组卷系统收敛速度慢,组卷质量较差等缺陷,设计一种基于改进小生境遗传算法的自适应组卷系统。首先依照组卷约束权重比,生成适应度值较高的初始种群．其次对选择策略进行优化。引入小生境预选机制以维持种群多样性,防止过早收敛。再次根据组卷特点改进交叉算子和变异算子,根据种群进化情况,自动调整交叉概率和变异概率。最后经过若干迭代后达到终止条件,以自适应方式生成目标试卷。实验结果证明,与简单遗传算法相比,改进的遗传算法在收敛速度和组卷质量上均有显著的提高。     

Associations of neighborhood disorganization and maternal spanking with children’s aggression: A fixed-effects regression analysis

This study employed fixed effects regression that controls for selection bias, omitted variables bias, and all time-invariant aspects of parent and child characteristics to examine the simultaneous associations between neighborhood disorganization, maternal spanking, and aggressive behavior in early childhood using data from the Fragile Families and Child Wellbeing Study (FFCWS). Analysis was based on 2,472 children and their mothers who participated in Wave 3 (2001–2003; child age 3) and Wave 4 (2003–2006; child age 5) of the FFCWS. Results indicated that higher rates of neighborhood crime and violence predicted higher levels of child aggression. Maternal spanking in the past year, whether frequent or infrequent, was also associated with increases in aggressive behavior. This study contributes statistically rigorous evidence that exposure to violence in the neighborhood as well as the family context are predictors of child aggression. We conclude with a discussion for the need for multilevel prevention and intervention approaches that target both community and parenting factors.     

Understanding trends in neighborhood child maltreatment rates: A three-wave panel study 1990–2010

This study examines how changes in the social and economic structure of neighborhoods relate to changes in child maltreatment report rates over an extended period. The panel study design allows us to partition the changes in child maltreatment report rates into a portion associated with how the levels of socio-economic risk factors have changed over time, and a portion related to how the relative importance of those factors in explaining maltreatment report rates has changed over time. Through the application of fixed effects panel models, the analysis is also able to control for unmeasured time-invariant characteristics of neighborhoods that may be a source of bias in cross-sectional studies. The study finds that increases in vacant housing, single parent families and unemployment rates are strongly associated with increases in child maltreatment report rates. Changes in racial/ethnic composition did not produce changes in maltreatment report rates except when they reached extreme levels of segregation. Although poverty rates were predictive of cross-sectional variation in child maltreatment, increases in neighborhood poverty became less associated with increases in child maltreatment report rates over time.     

《数据结构》的教学探讨

<数据结构>是计算机及相关专业的一门专业基础课程,在众多的计算机系统软件和应用软件中都要用到各种数据结构,要想有效地使用计算机,必须学习数据结构的有关知识.然而,该课程理论性和实践性强,内容抽象,是学生学习的一大难题.作为教师,如何改进教学方法,充分调动学生的学习主动性,把<数据结构>教活、教好是十分重要的,也是十分必要的.     

通用题库数据结构与组卷算法设计

题库的数据结构和组卷算法决定了题库的适用范围,该文根据不同课程试卷及其试题属性的构成,提出一种新的题库数据结构,并在分析组卷控制模型的基础上,对组卷算法进行详细设计。     

一种基于遗传算法的连续属性离散化方法

连续属性离散化是基于粗糙集的数据挖掘中的关键问题。针对该问题提出了一种遗传算法,在保持不可分辨关系的前提下,以最小断点集为目标,对交叉算子、变异算子进行了改进。最后,使用UCI机器学习数据对算法进行了验证,并与其它的离散化算法进行了比较,结果表明算法是有效的。     

基于免疫算法和人工势场的移动机器人路径规划

分析了人工势场模型存在的两个问题:目标不可到达问题和由于局部最优解的存在而产生的死锁问题。在人工势场路径规划的基础上,提出了基于免疫算法和人工势场法的最优路径规划方法。通过仿真算例验证了该模型的有效性。