一种分布式的模糊聚类方法

由于FCM算法中的初始值需要随机的设定,这种随机性不能保证每次都能达到全局最优,也就是说如果初始聚类中心的设置具有全局的特点,那么聚类的结果才能达到全局最优。因此主要针对模糊c-均值（FCM）聚类算法对初始值很敏感,而且容易陷入局部最优解的这一特点,提出了一种分布式的模糊聚类方法。首先用分治法得到模糊聚类的全局的聚类中心值,然后再用FCM进行聚类,从而克服FCM算法对初始值敏感和容易陷入局部最优解的缺陷,达到全局最优。经仿真实验证明结果是很理想的。     

一种改进的基于加权属性的SLOF离群点挖掘算法

SLOF算法采用了空间对象的空间属性和空间关系确定空间邻域,并结合非空间属性的权值来计算对象在其邻域内的离群度,但在计算属性权值时,仍然由邻域专家决定,存在人为因素。文中采用计算每个对象的每个非空间属性的去一划分信息熵增量,并通过这个值来反映各个属性对对象离群的贡献程度,给出一种改进的SLOF算法。实验结果表明,算法具有计算效率高和对用户依赖性小的优点。     

The integration of information in a digital,multi-modal learning environment

The aim of the reported eye-tracking study was to investigate whether learners integrate information presented on several pages within a digital learning environment and whether the underlying processes differ between a purely text-based and a multi-modal digital learning environment. Participants (N = 97) learned about the development of tornados. Picture presentation (yes vs. no) and presentation of inconsistent information between pages (yes vs. no) was varied between-subjects. Results showed that inconsistent information led to more intense gaze behavior. This indicates that learners re-activated information presented on previous pages to integrate it with the actual information, enabling them to detect the inconsistencies. Furthermore, the data pattern for text-related gaze behavior was not influenced by picture presentation, indicating that cognitive processes in purely text based or multi-modal learning environments are much the same. Comprehension outcomes did not differ between groups. The implications of these results are discussed.     

纵向数据线性混合模型中M估计的渐近正态性

对纵向数据的线性混合模型yk=Xkβ＋Ckτk＋ek用Fisher得分迭代法得到了参数的M估计（稳健估计）,并在一系列的正则条件下,证明了参数M估计的渐近性质．     

转型时期地方高师院校专业建设的问题与思考

专业建设是地方高师院校在转型过程中亟待解决的问题。地方高师院校必须找准定位,既要确保教师教育特色,又要立足于地方经济社会发展需要,适度发展非师专业,才能顺应当前趋势,在转型中得到更好的发展。     

论转型期基层统一战线在扩大公民有序政治参与中的作用

随着我国经济社会结构的转型不断深化,基层统一战线内部构成发生深刻变化,统战对象和领域都大量增加。基层统一战线是扩大公民有序政治参与的重要体现,但当前我国公民在政治参与方面还存在一些问题。转型期要利用基层统一战线的特点和优势,在进一步扩大公民有序政治参与主体、提供公民有序政治参与制度保障、营造公民有序政治参与的良好环境、取得民主监督成效等方面发挥作用。     

多种因素交互影响农村儿童阅读能力的数据挖掘研究

[目的/意义]通过数据挖掘算法,揭示各种因素的内在交互性对农村儿童阅读能力的影响。[方法/过程]利用基于Apriori算法的关联规则挖掘方法和可视化技术,对广东省粤西方言区农村儿童的阅读现状进行了定量化分析。[结果/结论]网络流量费高、受方言影响、家长不鼓励、缺少亲子阅读机会等因素交互作用是导致农村儿童对阅读不感兴趣的主要原因。留守儿童在受家长鼓励、亲子阅读、阅读兴趣以及信息获取能力等方面的支持度显著低于非留守儿童,并且阅读兴趣更容易受方言和农忙因素的影响。农村儿童阅读能力低下是网络信息的获取能力、留守及农忙状态、方言影响等多种因素共同作用的结果。     

电子书包环境下小学智慧课堂教学资源的有效推送

智慧课堂教学环境中,教师与学生之间的活动、交流、学习依赖多媒体信息技术平台推送的相关资源,学生借助网络媒体平台学习所形成的有效及时的数据,对于教师的课中教学环节起着极为重要的作用。本文将从教师教学、学生学习的数据资源推送着手,构建电子书包支撑环境下个性化师生互动的高效课堂。     

Predicting future dynamics from short-term time series using an Anticipated Learning Machine

Predicting time series has significant practical applications over different disciplines. Here, we propose an Anticipated Learning Machine (ALM) to achieve precise future-state predictions based on short-term but high-dimensional data. From non-linear dynamical systems theory, we show that ALM can transform recent correlation/spatial information of high-dimensional variables into future dynamical/temporal information of any target variable, thereby overcoming the small-sample problem and achieving multistep-ahead predictions. Since the training samples generated from high-dimensional data also include information of the unknown future values of the target variable, it is called anticipated learning. Extensive experiments on real-world data demonstrate significantly superior performances of ALM over all of the existing 12 methods. In contrast to traditional statistics-based machine learning, ALM is based on non-linear dynamics, thus opening a new way for dynamics-based machine learning.     

从大数据特征看档案界对于档案大数据认知的误区

本文认为档案数据不具备多维度、完整性和实时性,将大量档案数据称之为档案大数据是不准确的,甚至是错误的.大数据时代,档案数据不能缺位,要使档案数据成为真正意义上的大数据,就应改变不能适应时代的固有模式与方法,积极主动融合到社会大数据之中.