多媒体查询语言及其评价准则

随着多媒体技术和多媒体相关应用的发展,对有效检索多媒体信息的要求越来越迫切.查询语言作为信息检索的有效工具,其研究也越来越受到关注.对现有的多媒体查询语言进行了全面的综述,将它们分为两类:专用语言和通用语言.尽管查询语言对多媒体信息系统提供有效的查询服务至关重要,但目前仍没有较好的多媒体查询语言评价准则.因此,针对多媒体查询语言的查询表达能力设计准则共16条,并根据这些准则对多媒体查询语言进行了评价.评价结果表明,这些语言能满足用户的基本查询要求,但在高级语义查询和不确定查询等方面还有欠缺.最后展望了多媒体查询语言的未来研究方向.     

基于图像处理的自动聚焦技术及应用

自动聚焦技术是提高压痕直径测量系统测量精度、智能化和自动化的重要手段.介绍了采用图像处理法实现压痕直径测量系统的自动聚焦技术,核心就是选择一个合适的图像清晰度评价函数.在研究了众多图像清晰度评价函数的基础上,提出了基于向量模型和改进DCT变换的图像清晰度评价函数,实验表明,提出的算法具有良好的单峰性、准确性、稳定性、可靠性和快速性.最后通过基于COM组件技术的Matlab与VB混合编程来保证算法实现和软件设计.     

融合威胁动态检测的网络安全评估探究

首先对网络系统常规的安全评估做简要分析后,指出静态风险评估存在的不足之处,进而提出通过融入动态威胁检测,增加网络系统的动态安全评估,并给出相应的动态检测系统设计和部署,对改进网络系统的安全评估进行探究。     

Evaluation of a type 1 diabetes serum cohort by SELDI-TOF MS protein profiling

Proteomics analysis of serum from patients with type 1 diabetes (T1D) may lead to novel biomarkers for prediction of disease and for patient monitoring. However, the serum proteome is highly sensitive to sample processing and before proteomics biomarker research serum cohorts should preferably be examined for potential bias between sample groups. SELDI‐TOF MS protein profiling was used for preliminary evaluation of a biological‐bank with 766 serum samples from 270 patients with T1D, collected at 18 different paediatric centers representing 15 countries in Europe and Japan over 2 years (2000–2002). Samples collected 1 (n = 270), 6 (n = 248), and 12 (n = 248) months after T1D diagnosis were grouped across centers and compared. The serum protein profiles varied with collection site and day of analysis; however, markers of sample processing were not systematically different between samples collected at different times after diagnosis. Three members of the apolipoprotein family increased with time in patient serum collected 1, 6, and 12 months after diagnosis (ANOVA, p<0.001). These results support the use of this serum cohort for further proteomic studies and illustrate the potential of high‐throughput MALDI/SELDI‐TOF MS protein profiling for evaluation of serum cohorts before proteomics biomarker research.     

基于遗传算法对奥赛罗问题的研究与改进

遗传算法是一种高效的最佳化的搜寻方法,具有简单、通用、鲁棒性强的特点,适合处理传统的搜索方法难以解决的复杂和非线性问题.利用遗传算法的思想,根据先验知识设计遗传算法初始群体产生方式和适应度函数,对传统的奥赛罗问题进行求解,并在传统遗传算法中加入自我调适的方法,来提高遗传算法的总体性能.在对这些问题进行研究的基础上,给出了大量的对比实验证明该算法是有效的,得出改进的遗传算法比传统遗传算法有更好的演化结果,并克服了传统遗传算法的早熟现象.     

导弹仿真系统试验结果的可信性评估

对导弹武器的性能与效能分析而言,导弹仿真系统的试验结果有重要的参考价值.为了评估和保证导弹仿真系统试验结果的可信性,首先从时域和频域入手,通过对假设检验法和频谱分析法的深入分析,提出了导弹仿真系统试验结果的可信度检验准则;然后分析了仿真结果的可信性评估过程,包括飞行试验结果的预处理、飞行试验结果的一致性检验、飞行试验结果与仿真试验结果的一致性检验等阶段;最后根据某型导弹仿真系统运行得到的数据,评估出该试验结果的可信度为95%.可以证明,上述评估方法和过程确实实用且可行.     

基于深度学习的电网安全态势感知方法

为了提高电力网络的安全性,实现电力网络的可持续运行,引入深度学习神经网络,开展对电力网络安全态势感知方法的设计研究,以此提出一种全新的安全态势感知方法。本文采用电力网络安全态势评估指标,结合各类电力网络环境因素,对未来可能发生的电力网络变化趋势进行预测;明确电力网络安全态势评估指标及其相关表述含义后,对电力网络安全态势风险进行综合量化,通过划分电力网络安全态势风险量化及等级,构建基于深度学习的电力网络安全态势预测模型,验证模拟安全态势感知预测结果。通过真实电力网络算例的方式,得出新的安全态势感知方法应用在现实电力网络运行环境中时,能够实现对其安全等级的精准预测,可以为电力网络的可持续运行提供安全保障条件,具有一定的实用性。     

基于BSC-AHP的绩效评估模型研究与实现

为了解决企业战略目标执行难的问题,提升绩效评估合理性。笔者在深入研究现存理论和方法的基础上,采用平衡记分卡理论从企业战略目标中提取关键评估指标,并提出了用分解矩阵的方法将评估指标分解到各评估对象。最后引用层次分析法确定各指标权重。建立了引导、鼓励员工工作向有利于企业战略的执行与控制的绩效评估指标体系和评价模型。开发的系统在某印刷企业得到应用并取得良好效果,有效证明了模型的可行性和科学性。     

Investigating acceptance toward mobile learning to assist individual knowledge management: Based on activity theory approach

Mobile devices could facilitate human interaction and access to knowledge resources anytime and anywhere. With respect to wide application possibilities of mobile learning, investigating learners’ acceptance towards it is an essential issue. Based on activity theory approach, this research explores positive factors for the acceptance of m-learning systems. In the research, we developed an m-learning system for learners’ knowledge management and invited 152 participants who knew how to use the m-learning system then report on their experience. The results show that enhancing learners’ satisfaction, encouraging learners’ autonomy, empowering system functions, and enriching interaction and communication activities have a significant positive influence on the acceptance of m-learning systems.     

A quantitative multimodal discourse analysis of teaching and learning in a web-conferencing environment – The efficacy of student-centred learning designs

This paper presents a quantitative approach to multimodal discourse analysis for analyzing online collaborative learning. The coding framework draws together the fields of systemic functional linguistics and Activity Theory to analyze interactions between collaborative-, content- and technology-related discourse. The approach is used to examine how the task subject matter, the activity design, and the choice of interface affected interaction and collaboration for a computing course conducted in a web-conferencing environment. The analysis revealed the critical impact of activity design on the amount and type of discourse that transpired. Student-centred designs resulted in over six times more student discourse as compared to teacher-centred designs and created a learning environment where students took greater ownership over the tasks and contributed more to the content-based discussion. The paper also incorporates a rationale for the approach to coding and a reflection on its efficacy for discourse analysis in technology-based learning environments.