基于角度优化的鲁棒极端学习机算法

针对数据集样本中带有噪声和离群点问题,提出了一种基于角度优化的鲁棒极端学习机算法。该方法利用鲁棒激活函数角度优化的原则,首先降低了离群点对分类算法的影响,从而保持数据样本的全局结构信息,达到更好的去噪效果。其次,有效的避免隐层节点输出矩阵求解不准的问题,进一步增强极端学习机的泛化性能。通过应用在普遍图像数据库上的实验结果表明,这种提出的算法与其他算法相比具有更强的鲁棒性和较高的识别率。     

一种新的双足机器人模型设计与相关研究

针对双足机器人最简模型在行走过程中出现摆动腿足部擦地的问题,提出了一种通过摆动腿膝关节弯曲达到摆动腿缩短的新模型。当摆动腿开始摆动时,摆动腿膝关节弯曲锁定,摆动腿缩短;当摆动腿摆动到最大位置时,膝关节解锁,摆动腿伸直再锁定,此后摆动腿回摆,系统变为直腿模型。采用脚后跟冲击控制,在摆动腿落地前,拖后的支撑腿与地面接触处施加一指向髋关节的瞬时冲击力,冲击力可以减小摆动腿着地时能量的损耗,同时驱动被动机器人向前行走。设计了迭代学习控制算法,找到极限环与不动点,实现不同给定期望步长跟踪的冲击力的计算。仿真结果表明,迭代学习控制可以有效的实现不同期望步长的跟踪,可以很快的找到机器人系统的不动点,通过收敛的相平面,得到稳定的极限环,保证了机器人行走过程稳定。     

基于函数调用图的Android恶意代码检测方法研究

提出了一种基于函数调用图的 Android 程序特征提取及检测方法。该方法通过对 Android 程序进行反汇编得到函数调用图,在图谱理论基础上,结合函数调用图变换后提取出的图结构和提取算法,获取出具有一定抗干扰能力的程序行为特征。由于 Android 函数调用图能够较好地体现 Android 程序的功能模块、结构特征和语义。在此基础上,实现检测原型系统,通过对多个恶意 Android 程序分析和检测,完成了对该系统的实验验证。实验结果表明,利用该方法提取的特征能够有效对抗各类 Android 程序中的混淆变形技术,具有抗干扰能力强等特点,基于此特征的检测对恶意代码具有较好地识别能力。     

深度学习视域下的物质单元复习教学研究

从深度学习视域下的化学学科本质出发思考初中化学复习教学改革策略,以初中化学总复习重要内容之一——物质为例,呈现一例关于物质的深度学习的单元教学案例研究。从认知目标与内容的匹配、学生课堂学习参与的程度、深度学习的特征3个方面反思了教学实践。     

概念转变理论的发展及面临的挑战

经过30年的发展,经典概念转变理论的内涵已发生了变化。通过对概念转变理论发展历程的分析,总结了概念转变研究的视角、研究的共识和研究的争议,指出当前概念转变研究面临3方面的挑战:寻找概念转变有效证据的研究方法;如何将成功的概念转变策略引入课堂教学实践;要进一步明确"概念转变"的含义。     

研究性学习在精细化工课程教学中的实践与探索

在精细化工课程中以食品添加剂与食品安全问题为专题开展了研究性学习。以学生兴趣为导向,采用主题离散式的选题方法,开展小组合作研究。重视学习过程和学习能力培养,采用多种方式相结合的方法进行评价,并对研究性学习在实施过程中存在的问题进行了探讨。实践表明,研究性学习可有效地激发学生的学习兴趣,培养学生的探究能力和综合素质。     

基于思维导图的先行组织者支架在化学教学中的应用

教学中教师为了完成教学任务会忽视学生的高阶思维发展的培养,这样做虽然教学目标可以及时完成,但是教学效果会大打折扣,更重要的是直接阻碍了创新及批判性思维、获得隐性知识、自我管理等高阶能力的发展。高阶思维需要高阶学习支架支持,以“基于思维导图的先行组织者支架”在不同课型中的运用,帮助学生参与复杂的认知活动,进而建立高阶思维模型和提升元认知水平。     

Exploring the relationship between fractal features and bacterial essential genes

Essential genes are indispensable for the survival of an organism in optimal conditions.Rapid and accurate identifications of new essential genes are of great theoretical and practical significance.Exploring features with predictive power is fundamental for this.Here,we calculate six fractal features from primary gene and protein sequences and then explore their relationship with gene essentiality by statistical analysis and machine learning-based methods.The models are applied to all the currently available identified genes in 27 bacteria from the database of essential genes(DEG).It is found that the fractal features of essential genes generally differ from those of non-essential genes.The fractal features are used to ascertain the parameters of two machine learning classifiers:Na¨?ve Bayes and Random Forest.The area under the curve(AUC) of both classifiers show that each fractal feature is satisfactorily discriminative between essential genes and non-essential genes individually.And,although significant correlations exist among fractal features,gene essentiality can also be reliably predicted by various combinations of them.Thus,the fractal features analyzed in our study can be used not only to construct a good essentiality classifier alone,but also to be significant contributors for computational tools identifying essential genes.     

Recent Advances in In Silico Target Fishing

In silico target fishing, whose aim is to identify possible protein targets for a query molecule, is an emerging approach used in drug discovery due its wide variety of applications. This strategy allows the clarification of mechanism of action and biological activities of compounds whose target is still unknown. Moreover, target fishing can be employed for the identification of off targets of drug candidates, thus recognizing and preventing their possible adverse effects. For these reasons, target fishing has increasingly become a key approach for polypharmacology, drug repurposing, and the identification of new drug targets. While experimental target fishing can be lengthy and difficult to implement, due to the plethora of interactions that may occur for a single small-molecule with different protein targets, an in silico approach can be quicker, less expensive, more efficient for specific protein structures, and thus easier to employ. Moreover, the possibility to use it in combination with docking and virtual screening studies, as well as the increasing number of web-based tools that have been recently developed, make target fishing a more appealing method for drug discovery. It is especially worth underlining the increasing implementation of machine learning in this field, both as a main target fishing approach and as a further development of already applied strategies. This review reports on the main in silico target fishing strategies, belonging to both ligand-based and receptor-based approaches, developed and applied in the last years, with a particular attention to the different web tools freely accessible by the scientific community for performing target fishing studies.     

基于极端学习机背景预测的红外小目标检测算法

为自适应检测复杂环境中的红外小目标,提出了基于极端学习机背景预测的红外小目标检测算法。首先,依据灰度值分布设计局部边缘敏感平滑滤波器,在相近的灰度范围内,使中心像素的灰度值等于邻域内多数灰度值的融合,对红外图像进行滤波,能够去除大量噪声并突出图像主要结构;其次,利用极端学习机对滤波后的图像建立回归模型,以邻域像素值为输入,以中心像素值为输出训练模型,并对背景进行预测,得到的图像与滤波后的图像做差,得到小目标显著图;最后,利用图像块对比特性对显著区域处理,使小目标区域均匀突出,抑制背景区域,并经过简单阈值操作,实现对红外小目标的检测。实验结果表明:与其他检测算法相比,在复杂背景下,本文算法检测结果的局部信噪比增益最高,单帧检测时间为0.18 s。本文算法对背景进行学习,发掘背景与目标的差异,提高了算法的适应能力,并且能够有效检测小目标。