音乐物联网的设计与研究

音乐物联网(Internet of musical things,IoMusT)是一个涉及多学科、多专业的综合工程,以往的研究都是单一的从某一个方面着手,缺乏一个统一的研究框架,完整的研究体系还不健全。研究重点是从工程的角度定义音乐事件,并结合当下互联网各个环节的技术特性,从工程的角度定义音乐事件,构造一个小范围的IoMusT生态系统,从软硬件方面测试IoMusT性能,进而验证IoMusT设计架构效。     

An Overview of Control Charts for High‐quality Processes

Major difficulties in the study of high‐quality processes with traditional process monitoring techniques are a high false alarm rate and a negative lower control limit. The purpose of time‐between‐events control charts is to overcome existing problems in the high‐quality process monitoring setup. Time‐between‐events charts detect an out‐of‐control situation without great loss of sensitivity as compared with existing charts. High‐quality control charts gained much attention over the last decade because of the technological revolution. This article is dedicated to providing an overview of recent research and presenting it in a unifying framework. To summarize results and draw a precise conclusion from the statistical point of view, cross‐tabulations are also given in this article. Copyright © 2016 John Wiley & Sons, Ltd.     

MODIS imagery of turbid plumes in San Diego coastal waters during rainstorm events

Data of normalized water-leaving radiance, nLw, obtained from the Moderate Resolution Imaging Spectroradiometer (MODIS) on the Aqua satellite at spatial resolution of 250 m (band 1 centered at 645 nm) and 500 m (band 4 at 555 nm) are used to study turbid plumes in coastal waters of southern California during rainstorm events in winter of 2004-2005. Our study area includes San Diego coastal waters, which extend approximately 25 km offshore between Point Loma and 10 km south of the US-Mexican border. These waters are influenced by terrigenous input of particulate and dissolved materials from San Diego and Tijuana watersheds and non-point sources along the shore. Optimum threshold values of satellite-derived normalized water-leaving radiances at both wavebands were established for distinguishing the plume from ambient ocean waters. These threshold values were determined by searching for a maximum correlation between the estimates of satellite-derived plume area calculated using a broad range of nLw values and the environmental variables characterizing rainfall, river discharge, wind, and tides. A correlation analysis involving the amount of precipitated water accumulated during a storm event over the San Diego and Tijuana watersheds was selected as the basis for final determinations of the optimum threshold nLw_thr and subsequent calculations of the plume area. By applying this method to a sequence of MODIS imagery, we demonstrate the spatial extent and evolution of the plume during rainstorm events under various conditions of precipitation, river discharge, wind forcing, and coastal currents.     

神经网络极速学习方法研究进展

神经网络已经在模式识别、自动控制及数据挖掘等领域取得了广泛的应用,但学习方法的速度不能满足实际需求。传统的误差反向传播方法(BP)主要是基于梯度下降的思想,需要多次迭代;网络的所有参数都需要在训练过程中迭代确定,因此算法的计算量和搜索空间很大。ELM(Extreme Learning Machine,ELM)是一次学习思想使得学习速度提高很多,避免了多次迭代和局部最小值,具有良好的泛化性能、鲁棒性与可控性。但对于不同的数据集和不同的应用领域,无论ELM是用于数据分类或是回归,ELM算法本身还是存在问题,所以本文对已有方法深入对比分析,并指出极速学习方法未来的发展方向。     

粒子群优化算法的改进

针对粒子群优化算法搜索精度不高、对高维函数优化性能不佳的问题,提出一种改进的粒子群优化算法。以递增方式对粒子进行释放增强可利用的种群信息,通过释放粒子引导极值变化加强算法的运算效率。实验结果表明,与其他算法相比,改进算法具有更强的寻优能力和搜索精度,且适于高维复杂函数的优化。     

基于粒极值的最简规则提取算法

应用粗糙集的分辨关系,分别从表的行、列2个方面求出每个属性值的分辨、组合能力,以此确定出粒极值,将含粒极值的粒定义为极值粒。应用粒计算理论,以极值粒集为主要运算对象进行粒逻辑运算,使最终的组合粒最简,即所提取的规则最简。实验验证了算法的有效性。     

基于稀有事件仿真的商业银行IT操作风险评估研究

系统仿真是风险评价的一种重要手段,针对商业银行IT操作风险预警问题,提出了一种基于稀有事件仿真的IT操作风险评估方法。采用商业银行IT操作风险的概率作为衡量IT操作风险高低的标准,构造基于稀有事件的商业银行IT操作风险识别模型,利用交叉熵方法构建了一种稀有事件仿真的有效算法,并由此估计出发生损失的概率。实证分析结果表明,模型对商业银行IT操作风险具有很强的识别能力,从而提供了一个风险预警的新视角。     

SELM: Semi-supervised ELM with application in sparse calibrated location estimation

Indoor location estimation based on Wi-Fi has attracted more and more attention from both research and industry fields. It brings two significant challenges. One is requiring a vast amount of labeled calibration data. The other is real-time training and testing for location estimation task. Traditional machine learning methods cannot get high performance in both aspects. This paper proposed a novel semi-supervised learning method SELM (semi-supervised extreme learning machine) and applied it to sparse calibrated location estimation. There are two advantages of the proposed SELM. First, it employs graph Laplacian regularization to import large number of unlabeled samples which can dramatically reduce labeled calibration samples. Second, it inherits the good property of ELM on extreme training and testing speed. Comparative experiments show that with same number of labeled samples, our method outperforms original ELM and back propagation (BP) network, especially in the case that the calibration data is very sparse.     

基于极速神经网络的协作过滤方法研究

协作过滤是一种有效的个性化推荐技术,针对该技术随着用户和资源的增多,数据的高维稀疏特性严重导致推荐质量的下降和计算速度减慢的问题,研究并实现了一种基于极速神经网络的协作过滤方法。采用主成分分析解决数据高维稀疏性问题,采用极速神经网络技术解决计算速度慢的问题。实验结果表明,该方法具有良好的泛化性能和学习速度,能很好的满足个性化资源推荐的需求。     

基于LS-BSVR的旅游地理经济预测应用

基于改进的LS-BSVR(least squares B support vector regression)算法对旅游地理经济进行分析预测。提出了时政指数、景区景点分布指数概念,设计了时政指数、景区景点分布指数,并应用于旅游地理经济进行分析预测。设计了用于分析预测的数据模式。数值实验和预测平台结果表明,改进的LS-BSVR算法和设计的旅游地理经济数据指标以及数据模式是有效的。