Process identification with multiple neural network models

This paper describes the identification of chemical processes with multiple neural network models. This concept is called a 'multimodel' approach. The multimodel approach presents a flexible framework which allows the integration of other model paradigms. Three different methods to construct neural network multimodels are presented. First, a priori knowledge is used to decompose the input domain into operating regimes. Each regime is modelled by a different neural network. In the second multimodel approach, unsupervised learning in form of clustering and SOMs is used to split the input domain. The third approach uses a gating neural network to divide the input space. In contrast to the first two approaches, the non-linear gated network approach allows the multim odel to simultaneously learn a suitable decomposition and the mapping in each regime. All three approaches are evaluated for a fed-batch fermentation.     

A novel clustering algorithm based on data transformation approaches

Clustering provides a knowledge acquisition method for intelligent systems. This paper proposes a novel data-clustering algorithm, by combining a new initialization technique, K-means algorithm and a new gradual data transformation approach to provide more accurate clustering results than the K-means algorithm and its variants by increasing the clusters’ coherence. The proposed data transformation approach solves the problem of generating empty clusters, which frequently occurs for other clustering algorithms. An efficient method based on the principal component transformation and a modified silhouette algorithm is also proposed in this paper to determine the number of clusters. Several different data sets are used to evaluate the efficacy of the proposed method to deal with the empty cluster generation problem and its accuracy and computational performance in comparison with other K-means based initialization techniques and clustering methods. The developed estimation method for determining the number of clusters is also evaluated and compared with other estimation algorithms. Significances of the proposed method include addressing the limitations of the K-means based clustering and improving the accuracy of clustering as an important method in the field of data mining and expert systems. Application of the proposed method for the knowledge acquisition in time series data such as wind, solar, electric load and stock market provides a pre-processing tool to select the most appropriate data to feed in neural networks or other estimators in use for forecasting such time series. In addition, utilization of the knowledge discovered by the proposed K-means clustering to develop rule based expert systems is one of the main impacts of the proposed method.     

基于文化算法的聚类分析

分析了K-均值聚类算法所存在的不足,提出了基于文化算法的新聚类算法,并给出该算法的两个实现版本：CA-version1利用规范知识调整变量变化步长,形势知识调整其变化方向;CA-version2利用规范知识调整变量变化步长及变化方向。文化算法所具有的双层结构特性,使其在问题求解过程中能够利用经验知识来指导搜索过程,从而具有较好的全局寻优性能。仿真实验亦表明,两个版本的文化算法均能有效地克服传统的K-均值算法的缺点,而且全局收敛性能优于基于遗传算法的K-均值聚类算法,同时还可以看出第二个版本的文化算法更适于求解聚类问题。     

Clustering based on median and closest string via rank distance with applications on DNA

This paper aims to present several clustering methods based on rank distance. Rank distance has applications in many different fields such as computational linguistics, biology and computer science. The K-means algorithm represents each cluster by a single mean vector. The mean vector is computed with respect to a distance measure. Two K-means algorithms based on rank distance are described in this paper. Hierarchical clustering builds models based on distance connectivity. This paper describes two hierarchical clustering techniques that use rank distance. Experiments using mitochondrial DNA sequences extracted from several mammals are performed to compare the results of the clustering methods. Results demonstrate the clustering performance and the utility of the proposed algorithms.     

Neural clustering networks based on global optimisation of prototypes in metric spaces

The utilisation of clustering algorithms based on the optimisation of prototypes in neural networks is demonstrated for unsupervised learning. Stimulated by common clustering methods of this type (learning vector quantisation [LVQ, GLVQ] and K-means) a globally operating algorithm was developed to cope with known shortcomings of existing tools. This algorithm and K-means (for the common methods) were applied to the problem of clustering EEG patterns being pre-processed. It can be shown that the algorithm based on global random optimisation may find an optimal solution repeatedly, whereas K-means provides different sub-optimal solutions with respect to the quality measure defined as objective function. The results are presented. The performance of the algorithms is discussed.     

A Sliding Mode Multiobserver Based on an Uncoupled Multimodel: An Application on a Transesterification Reaction

This paper concerns the design of robust sliding mode multiobserver for nonlinear systems. A discrete uncoupled multimodel structure is retained for the modeling of nonlinear systems. Unlike the classically used multimodel structures, the retained uncoupled multimodel is known by its flexibility of modeling, thus, the structures of the partial models are adapted to the complexity of the local models in each operating zone. Sufficient conditions are provided, in terms of linear matrix inequalities (LMIs), to ensure the asymptotic stability of the proposed sliding mode multiobserver. A convergence analysis is achieved to obtain the convergence radius. A numerical example and a real time application on a transesterification reactor are carried out to illustrate, once again, the performance of the proposed sliding mode multiobserver in terms of precision and rapidity of convergence.     

A Survey and Systematic Categorization of Parallel K-Means and Fuzzy-C-Means Algorithms

Parallel processing has turned into one of the emerging fields of machine learning due to providing consistent work by performing several tasks simultaneously, enhancing reliability (the presence of more than one device ensures the workflow even if some devices disrupted), saving processing time and introducing low cost and high-performance computation units. This research study presents a survey of parallel K-means and Fuzzy-c-means clustering algorithms based on their implementations in parallel environments such as Hadoop, MapReduce, Graphical Processing Units, and multi-core systems. Additionally, the enhancement in parallel clustering algorithms is investigated as hybrid approaches in which K-means and Fuzzy-c-means clustering algorithms are integrated with metaheuristic and other traditional algorithms.     

Optimal Systematic Determination of Models' Base for Multimodel Representation: Real Time Application

The multimodel approach is a powerful and practical tool to deal with analysis, modeling, observation, emulation and control of complex systems. In the modeling framework, we propose in this paper a new method for optimal systematic determination of models base for multimodel representation. This method is based on the classification of data set picked out of the considered system. The obtained cluster centers are exploited to provide the weighting functions and to deduce the corresponding dispersions and their models base. A simulation example and an experimental validation on a semi-batch reactor are presented to evaluate the effectiveness of the proposed method.     

Mining Projected Clusters in High-Dimensional Spaces

Clustering high-dimensional data has been a major challenge due to the inherent sparsity of the points. Most existing clustering algorithms become substantially inefficient if the required similarity measure is computed between data points in the full-dimensional space. To address this problem, a number of projected clustering algorithms have been proposed. However, most of them encounter difficulties when clusters hide in subspaces with very low dimensionality. These challenges motivate our effort to propose a robust partitional distance-based projected clustering algorithm. The algorithm consists of three phases. The first phase performs attribute relevance analysis by detecting dense and sparse regions and their location in each attribute. Starting from the results of the first phase, the goal of the second phase is to eliminate outliers, while the third phase aims to discover clusters in different subspaces. The clustering process is based on the K-means algorithm, with the computation of distance restricted to subsets of attributes where object values are dense. Our algorithm is capable of detecting projected clusters of low dimensionality embedded in a high-dimensional space and avoids the computation of the distance in the full-dimensional space. The suitability of our proposal has been demonstrated through an empirical study using synthetic and real datasets.     

一种改进的多视图聚类集成算法

近年来,针对大数据的数据挖掘技术和机器学习算法研究变得日趋重要。在聚类领域,随着多视图数据的大量出现,多视图聚类已经成为了一类重要的聚类方法。然而,大多数现有的多视图聚类算法受算法参数设置、数据样本等影响,具有聚类结果不稳定、参数需要反复调节等缺点。基于多视图K-means算法和聚类集成技术,提出了一种改进的多视图聚类集成算法,其提高了聚类的准确性、鲁棒性和稳定性。其次,由于单机环境下的多视图聚类算法难以对海量的数据进行处理,结合分布式处理技术,实现了一种分布式的多视图并行聚类算法。实验证明,并行算法在处理大数据时的时间效率有很大提升,适合于大数据环境下的多视图聚类分析。     

基于自适应步长的萤火虫划分聚类算法

在许多领域中,聚类是重要分析技术之一,如数据挖掘、模式识别和图像分析。针对K-means算法过度依赖初始聚类中心的选择而陷入局部最优的问题,提出了基于自适应步长的萤火虫划分聚类算法,简称ASFA。利用萤火虫算法的随机性和全局搜索性,来找到指定数量的初始簇中心,进一步利用K-means得到精确的簇划分。在萤火虫聚类优化算法中,采用自适应步长代替原有的固定步长,从而避免算法陷入局部最优,且能获得精度更高的解。为了提高算法性能,将改进的新算法用于不同规模大小的标准数据集中,实验结果表明,ASFA与K-means、GAK、PSOK对比显示更好的聚类性能和更好的稳定性和鲁棒性,与其他文献中算法相比,ASFA在寻优精度方面能取得更好的效果。     

A simple and fast algorithm for K-medoids clustering

This paper proposes a new algorithm for K-medoids clustering which runs like the K-means algorithm and tests several methods for selecting initial medoids. The proposed algorithm calculates the distance matrix once and uses it for finding new medoids at every iterative step. To evaluate the proposed algorithm, we use some real and artificial data sets and compare with the results of other algorithms in terms of the adjusted Rand index. Experimental results show that the proposed algorithm takes a significantly reduced time in computation with comparable performance against the partitioning around medoids.     

基于聚类与机器学习的零售商品销量预测

本文提出一种基于K-means聚类与机器学习回归算法的预测模型以解决零售行业多个商品的销售预测问题,首先通过聚类分析识别出具有相似销售模式的商品从而实现数据集的划分,然后分别在每个子数据集上训练了支持向量回归、随机森林以及XGBoost模型,通过构建数据池的方式增加了用于训练模型的数据量以及预测变量的选择范围.在一家零售企业的真实销售数据集上对提出的模型进行了验证,实验结果表明基于K-means和支持向量回归的预测模型表现最优,且所提出的模型预测效果明显优于基准模型以及不使用聚类的机器学习模型.     

Uncertain data stream algorithm based on clustering RBF neural network

In this paper, the existing algorithms for modeling uncertain data streams based on radial basis function neural networks have problems of low accuracy, weak stability and slow speed. A new clustering method for uncertain data streams is proposed. Radial basis function neural network of the algorithm. The algorithm firstly models the uncertain data stream, then combines the fuzzy theory and the neural network principle to obtain the radial basis function neural network, and then obtains the radial basis function neural network through the clustering algorithm of the regular tetrahedral uncertain vector. The central weight and width weights ultimately result in hidden layer output and output layer output results. The experimental results show that the proposed algorithm is an effective algorithm for modeling uncertain data streams using clustering radial basis function neural networks. It has higher precision, stability and speed than similar algorithms.     

数据驱动的蒸发器在线建模方法

针对蒸发器离线建模方法对变量运行工况范围要求较大的问题，利用K-means算法对辨识模型的观测数据进行聚类筛选处理，提出一种基于数据的蒸发器在线建模方法。首先利用DB准则和PSO算法提出K-means算法中最优分类数K*和最优初始聚类中心的确定方法，提高算法的收敛速度，并使用改进的K-means算法获得各簇聚类中心来代替辨识模型的观测数据，减少模型辨识的数据量。然后利用已有的蒸发器模型结构以及模型辨识方法，对模型进行辨识。实验结果表明：利用聚类筛选前、后的观测数据所辨识的模型精度基本相当，分别在±3%和±3.5%以内。最后利用在线观测数据到各聚类中心欧氏距离的分析判断，提出蒸发器的在线建模方法。该方法可以先采用小工况范围的少量离线数据辨识模型，再利用在线数据修正模型参数，扩大模型的适用范围。     

基于差分隐私保护的Stacking集成聚类算法研究

针对差分隐私保护下单一聚类算法准确性和安全性不足的问题,提出了一种基于差分隐私保护的Stacking集成聚类算法。使用Stacking集成多种异质聚类算法,将K-means聚类、Birch层次聚类、谱聚类和混合高斯聚类作为初级聚类算法,结合轮廓系数对初级聚类算法产生的聚类结果加权并入原始数据,将K-means算法作为次级聚类算法对扩展后的数据集进行聚类分析。其中,针对原始数据和初级聚类算法的聚类结果分别提出自适应的ε函数确定隐私预算,为不同敏感度的数据分配不同程度的Laplace噪声。理论分析和实验结果均表明,与单一聚类算法相比,该算法满足ε-差分隐私保护的同时有效提高了聚类准确性,实现了隐私保护与数据可用性的高度平衡。     

基于样本空间分布密度的初始聚类中心优化K-均值算法*

针对传统K-均值聚类算法对初始聚类中心敏感、现有初始聚类中心优化算法缺乏客观性,提出一种基于样本空间分布密度的初始聚类中心优化K-均值算法。该算法利用数据集样本的空间分布信息定义数据对象的密度,并根据整个数据集的空间信息定义了数据对象的邻域;在此基础上选择位于数据集样本密集区且相距较远的数据对象作为初始聚类中心,实现K-均值聚类。UCI机器学习数据库数据集以及随机生成的带有噪声点的人工模拟数据集的实验测试证明,本算法不仅具有很好的聚类效果,而且运行时间短,对噪声数据有很强的抗干扰性能。基于样本空间分布密度的初始聚类中心优化K-均值算法优于传统K-均值聚类算法和已有的相关K-均值初始中心优化算法。     

一种基于改进混合蛙跳的聚类算法

聚类分析是一种无监督的模式识别方式,它是数据挖掘中的重要技术之一。给出了一种基于改进混合蛙跳算法的聚类分析方法,该方法结合了K—均值算法和改进混合蛙跳算法各自的优点,引入了K—均值操作,再用改进混合蛙跳算法进行优化,很大程度上提高了该算法的局部搜索能力和收敛速度。通过仿真对基于改进混合蛙跳的聚类方法与其他已有的聚类方法进行了比较,验证了所提出算法的优越性。     

基于Ant-Tree算法的短文本聚类研究

短文本由于词频过低,使用常规的聚类算法如K-means效果不理想,难得到可接受的准确度。而最近结合使用生物启发及聚类内部有效性测量改进的方法,能够有效改善短文本的聚类效果。针对短文本聚类,提出了改进Ant-Tree的算法。该算法引入了轮廓系数作为内部效度测量,对K-means算法获得的初始聚类划分计算轮廓系数值,根据各聚簇样本值大小排序,将排序结果应用于Ant-Tree算法的初始化步骤中,使Ant-Tree算法性能得到提高。实验结果表明,该算法准确度超过了其它的算法。     

A genetic-based neuro-fuzzy approach for modeling and control ofdynamical systems

Linguistic modeling of complex irregular systems constitutes the heart of many control and decision making systems, and fuzzy logic represents one of the most effective algorithms to build such linguistic models. In this paper, a linguistic (qualitative) modeling approach is proposed. The approach combines the merits of the fuzzy logic theory, neural networks, and genetic algorithms (GAs). The proposed model is presented in a fuzzy-neural network (FNN) form which can handle both quantitative (numerical) and qualitative (linguistic) knowledge. The learning algorithm of a FNN is composed of three phases. The first phase is used to find the initial membership functions of the fuzzy model. In the second phase, a new algorithm is developed and used to extract the linguistic-fuzzy rules. In the third phase, a multiresolutional dynamic genetic algorithm (MRD-GA) is proposed and used for optimized tuning of membership functions of the proposed model. Two well-known benchmarks are used to evaluate the performance of the proposed modeling approach, and compare it with other modeling approaches.