一种基于投票策略的聚类融合算法

在分类算法和回归模型中,融合方法正得到越来越广泛的应用,但在非监督机器学习领域,由于缺乏数据集的先验知识,则不能直接用于聚类算法.提出并实现了一种基于投票策略的聚类融合算法,该算法利用k-means算法每次随机选取聚类中心而得到不同样本划分的特性,将多次运行得到的聚类结果通过投票的方式合并,从而得到最终的结果.通过一系列真实数据和合成数据集的实验证明,这种方法比单一的聚类算法能更有效地提高聚类的准确率.在此基础上,为了降低高维数据运算的复杂性,将随机划分属性子空间的方法应用到上述聚类融合算法中,实验证明,该方法同时也能够在一个属性子空间上获得好的聚类结果.     

基于MiniGUI的图像即时处理终端实现的关键技术

以现代嵌入式电子产品中图像处理终端为研究对象,本文对图像即时编辑功能进行了分析和设计。基于嵌入式删开发平台,使用MiniGUI嵌入式图形界面开发系统,最终实现一个可完成图像浏览、更新、标注、以及利用二分聚类算法实现搜索功能的图形界面应用程序,实现了图像的即时处理,并提高了图像检索速度。     

A fixed clustering protocol based on random relay strategy for EHWSN

In Energy Harvesting Wireless Sensor Networks (EHWSN), the communication protocol will directly affect the final performance of the network, so it is necessary to study the communication protocol based on EHWSN. In this paper, for the low-cost fixed clustering problem, a fixed clustering protocol RRCEH is based on random relaying. Our proposed RRCEH abandons the inefficient inter-cluster communication method of the traditional fixed clustering protocol. To coordinate the data upload of the cluster head, RRCEH allocates different random relay vectors to each ring area of the network, and combines all the random relay vectors into a random relay matrix of RRCEH. In each communication round, the cluster head node randomly selects its relay target node to send data according to the probability distribution in the random relay vector in the area. For two different cluster head configuration scenarios, by optimizing the random relay matrix, RRCEH can effectively reduce the network's configuration requirements for cluster head energy harvesting capability, thus reducing the deployment cost of EHWSN.     

Evaluation of Grid-Connected Photovoltaic Plants Based on Clustering Methods

Photovoltaic (PV) systems are electric power systems designed to supply usable solar power by means of photovoltaics, which is the conversion of light into electricity using semiconducting materials. PV systems have gained much attention and are a very attractive energy resource nowadays. The substantial advantage of PV systems is the usage of the most abundant and free energy from the sun. PV systems play an important role in reducing feeder losses, improving voltage profiles and providing ancillary services to local loads. However, large PV grid-connected systems may have a destructive impact on the stability of the electric grid. This is due to the fluctuations of the output AC power generated from the PV systems according to the variations in the solar energy levels. Thus, the electrical distribution system with high penetration of PV systems is subject to performance degradation and instabilities. For that, this project attempts to enhance the integration process of PV systems into electrical grids by analyzing the impact of installing grid-connected PV plants. To accomplish this, an indicative representation of solar irradiation datasets is used for planning and power flow studies of the electric network prior to PV systems installation. Those datasets contain lengthy historical observations of solar energy data, that requires extensive analysis and simulations. To overcome that the lengthy historical datasets are reduced and clustered while preserving the original data characteristics. The resultant clusters can be utilized in the planning stage and simulation studies. Accordingly, studies related to PV systems integration into the electric grid are conducted in an efficient manner, avoiding computing resources and processing times with easier and practical implementation.     

Disaster Monitoring of Satellite Image Processing Using Progressive Image Classification

The analysis of remote sensing image areas is needed for climate detection and management, especially for monitoring flood disasters in critical environments and applications. Satellites are mostly used to detect disasters on Earth, and they have advantages in capturing Earth images. Using the control technique, Earth images can be used to obtain detailed terrain information. Since the acquisition of satellite and aerial imagery, this system has been able to detect floods, and with increasing convenience, flood detection has become more desirable in the last few years. In this paper, a Big Data Set-based Progressive Image Classification Algorithm (PICA) system is introduced to implement an image processing technique, detect disasters, and determine results with the help of the PICA, which allows disaster analysis to be extracted more effectively. The PICA is essential to overcoming strong shadows, for proper access to disaster characteristics to false positives by operators, and to false predictions that affect the impact of the disaster. The PICA creates tailoring and adjustments obtained from satellite images before training and post-disaster aerial image data patches. Two types of proposed PICA systems detect disasters faster and more accurately (95.6%).     

Systematic review on privacy categorisation

In the modern digital world users need to make privacy and security choices that have far-reaching consequences. Researchers are increasingly studying people’s decisions when facing with privacy and security trade-offs, the pressing and time consuming disincentives that influence those decisions, and methods to mitigate them. This work aims to present a systematic review of the literature on privacy categorisation, which has been defined in terms of profile, profiling, segmentation, clustering and personae. Privacy categorisation involves the possibility to classify users according to specific prerequisites, such as their ability to manage privacy issues, or in terms of which type of and how many personal information they decide or do not decide to disclose. Privacy categorisation has been defined and used for different purposes. The systematic review focuses on three main research questions that investigate the study contexts, i.e. the motivations and research questions, that propose privacy categorisations; the methodologies and results of privacy categorisations; the evolution of privacy categorisations over time. Ultimately it tries to provide an answer whether privacy categorisation as a research attempt is still meaningful and may have a future.     

融合自动权重学习的深度子空间聚类

子空间聚类算法是一种面向高维数据的聚类方法,具有独特的数据自表示方式和较高的聚类精度。传统子空间聚类算法聚焦于对输入数据构建最优相似图再进行分割,导致聚类效果高度依赖于相似图学习。自适应近邻聚类（CAN）算法改进了相似图学习过程,根据数据间的距离自适应地分配最优邻居以构建相似图和聚类结构。然而,现有CAN算法在进行高维数据非线性聚类时,难以很好地捕获局部数据结构,从而导致聚类准确性及算法泛化能力有限。提出一种融合自动权重学习与结构化信息的深度子空间聚类算法。通过自编码器将数据映射到非线性潜在空间并降维,自适应地赋予潜在特征不同的权重从而处理噪声特征,最小化自编码器的重构误差以保留数据的局部结构信息。通过CAN方法学习相似图,在潜在表示下迭代地增强各特征间的相关性,从而保留数据的全局结构信息。实验结果表明,在ORL、COIL-20、UMIST数据集上该算法的准确率分别达到0.780 1、0.874 3、0.742 1,聚类性能优于LRR、LRSC、SSC、KSSC等算法。     

低功耗异构计算架构的高光谱遥感图像分类研究

高光谱图像分类算法通常需要逐点对图像中的像素点进行迭代处理,计算复杂度及并行程度存在较大差异。随着高光谱遥感图像空间、光谱和辐射分辨率的不断提升,这些算法无法满足实时处理海量遥感图像数据的需求。通过分析NPU存储计算一体化模式与遥感图像分类算法的实现步骤,设计低功耗CPU+NPU异构资源计算架构的低秩稀疏子空间聚类（LRSSC）算法,将数据密集型计算转移至NPU,并利用NPU数据驱动并行计算和内置AI加速,对基于机器学习算法的海量遥感数据进行实时分类。受到big.LITTLE计算范式的启发,CPU+NPU异构资源计算架构由8 bit和低精度位宽NPU共同组成以提高整体吞吐量,同时减少图网络推理过程中的能量损耗。实验结果表明,与CPU计算架构和CPU+GPU异构计算架构的LRSSC算法相比,CPU+NPU异构计算架构的LRSSC算法在Pavia University遥感数据集下的计算速度提升了3~14倍。     

Clustering of temporal gene expression data by regularized spline regression and an energy based similarity measure

Clustering analysis of temporal gene expression data is widely used to study dynamic biological systems, such as identifying sets of genes that are regulated by the same mechanism. However, most temporal gene expression data often contain noise, missing data points, and non-uniformly sampled time points, which imposes challenges for traditional clustering methods of extracting meaningful information. In this paper, we introduce an improved clustering approach based on the regularized spline regression and an energy based similarity measure. The proposed approach models each gene expression profile as a B-spline expansion, for which the spline coefficients are estimated by regularized least squares scheme on the observed data. To compensate the inadequate information from noisy and short gene expression data, we use its correlated genes as the test set to choose the optimal number of basis and the regularization parameter. We show that this treatment can help to avoid over-fitting. After fitting the continuous representations of gene expression profiles, we use an energy based similarity measure for clustering. The energy based measure can include the temporal information and relative changes of the time series using the first and second derivatives of the time series. We demonstrate that our method is robust to noise and can produce meaningful clustering results.