一种用于神经网络样本划分的自聚类算法

建立神经网络模型时 ,能否合理地划分训练样本和检验样本直接关系到建模的效率 .在很多实际应用中 ,检验样本是随机抽取的 .本文提出了一种基于欧氏距离的自聚类算法 ,根据样本的空间分布情况对其自动分类 ,然后确定检验样本 .算例研究表明 ,应用此算法能够改善检验效果 ,从而提高建模效率 .

基于大数据技术的微博舆情快速自聚类方法研究

[目的/意义]针对海量的微博舆情信息及微博在网络舆论场中的重要作用,如何快速、准确地获取微博舆情,是提升网络舆情实时监测及分析能力的关键课题之一。目前已能以较高的准确度获取微博舆情,但仍存在舆情获取耗时长问题,为较有效地解决该问题,提出一种基于大数据技术的微博舆情快速自聚类方法。[方法/过程]该方法首先利用大数据技术抓取和处理海量的微博舆情信息,而后根据构建的微博文本相似度速算模型和文本自主聚类模型快速自主聚类微博舆情。文本相似度速算模型通过两文本间同名的特征词数与其特征词数量较小值的比值来度量文本相似度;而文本自主聚类模型以一个主题为聚类起始,自主自适应扩展聚类主题,并将文本相似度大于设定阈值的文本直接聚为一类。[结果/结论]实验结果表明:提出的微博舆情快速自聚类方法能快速、准确地获取微博舆情,且具有较低的舆情误报率和漏报率,可为网络舆情实时监测及分析能力的提升提供一定的方法支持。

基于网络自聚类的PBFT算法改进

联盟链是区块链技术在实际行业应用的主要形式,其共识机制多采用实用拜占庭容错算法(PBFT),在节点数量大时共识成功率与共识效率不高,存在扩展性问题。为此,提出一种基于网络自聚类拜占庭容错共识算法NAC-PBFT。利用行业应用中网络结构、系统节点等确知信息,在联盟链审核节点时指定种子节点,再以种子节点为中心自聚类为若干分组,组内通过优化实用拜占庭容错算法选举出代理人,由各组代理人共同完成全局共识。其中,组内选举时,通过定义可信度指标衡量节点作为筛选候选代理人的标准,确保每次选出的代理人具有良好的状态。通过对系统分析与性能测试,NAC-PBFT算法能有效降低消息量,在共识时间、系统吞吐量指标上有更好的表现,具备较好的扩展性。

基于数学形态学的QRS波自聚类方法

以数学形态学为基础,利用加权Hausdorff距离作为心电波形的相似性度量,提出了一种新的QRS波自聚类方法。本文详细介绍了二值图像的Hausdorff距离计算方法、加权Hausdorff距离和基于此的QRS波自聚类方法,并利用MIT-BIH心律失常数据库进行了算法检验。

K-means改进算法在电力用户聚类辨识中的应用

科学、准确的用户用电特征分析对掌握负荷发展变化规律,提高电力需求预测的准确性,保障系统规划和经济运行具有重要意义。文中在对K-means算法深入研究的基础上,结合电力负荷数据海量、多维等特点,通过归一化处理,异常数据剔除,改进的二分K-means算法进行自聚类,对各优化算法进行分析,克服了传统K-means算法对异常数据敏感和初始聚类中心的随机性问题。实验结果表明,优化的自聚类算法能够提高分类的准确性,提高收敛效率,实现用户数据特征自动辨识分类。

探讨复杂系统中的群体结构

为探讨恰当划分复杂系统群体结构的新方法,提出了改进边介数法。在对网络实例空手道俱乐部的研究中,将该方法与传统社会学聚类方法和GN算法进行了对照比较。其实证结果表明了该方法对于正确分析系统的群体结构的有效性。

Passive location estimation using scatterer information for non-line-of-sight environments

In order to improve the performance of the traditional hybrid time-of-arrival（TOA）/angle-of-arrival（AOA）location algorithm in non-line-of-sight（NLOS）environments,a new hybrid TOA/AOA location estimation algorithm by utilizing scatterer information is proposed.The linearized region of the mobile station（MS）is obtained according to the base station（BS）coordinates and the TOA measurements.The candidate points（CPs）of the MS are generated from this region.Then,using the measured TOA and AOA measurements,the radius of each scatterer is computed.Compared with the prior scatterer information,true CPs are obtained among all the CPs.The adaptive fuzzy clustering（AFC）technology is adopted to estimate the position of the MS with true CPs.Finally,simulations are conducted to evaluate the performance of the algorithm.The results demonstrate that the proposed location algorithm can significantly mitigate the NLOS effect and efficiently estimate the MS position.

Analysis on Chemical Compositions in Chinese Wolfberry from Different Producing Areas

[Objective] This study aimed to conduct correspondence cluster analysis of the trace elements in Chinese wolfberry from Qinghai and Ningxia regions, and to investigate the relationship among the quality of the wolfberry samples, the composition of trace elements and the sample sources. [Method] The determined contents of trace elements and ratios of zinc to copper （Zn/Cu） of wolfberry from 11 different producing areas of Qinghai and Ningxia regions were adopted to construct the raw measurement data matrix, to analyze the distribution characteristics of the trace ele- ments in wolfberry from Qinghai and Ningxia by using the corresponding cluster analysis method. [Result] The quality of wolfberry samples in 7hongning County, Zhongwei City, Pingluo County, Shizuishan City, Heicheng Town of Ningxia Hui Au-tonomous Region and Hehuang Valley, Golmud City of Qinghai Province is mainly related to the contents of Zn and Mn; Zn/Cu greatly affects the quality of Chinese wolfberry in Dulan County of Qinghai Province; Fe has great effect on the quality of Chinese wolfberry in Yinchuan City of Ningxia Hui Autonomous Region; Cu greatly affects the quality of Chinese wolfberry in Nuomuhong Village of Qinghai Province and a wolfberry research institute in Ningxia. [Conclusion] The relationship between the quality of wolfberry from different producing areas and the trace elements was investigated, which provides theoretical and practical basis for the cultivation, har- vesting, processing, and further development and utilization of Chinese wolfberry resources from different producing areas.

Recognition of Spontaneous Combustion in Coal Mines Based on Genetic Clustering

Spontaneous combustion is one of the greatest disasters in coal mines. Early recognition is important because it may be a potential inducement for other coalmine accidents. However, early recognition is difficult because of the complexity of different coal mines. Fuzzy clustering has been proposed to incorporate the uncertainty of spontaneous combustion in coal mines and it can give a clear degree of classification of combustion. Because FCM clustering tends to become trapped in local minima, a new approach of fuzzy c-means clustering based on a genetic algorithm is there- fore proposed. Genetic algorithm is capable of locating optimal or near optimal solutions to difficult problems. It can be applied in many fields without first obtaining detailed knowledge about correlation. It is helpful in improving the effec- tiveness of fuzzy clustering in detecting spontaneous combustion. The effectiveness of the method is demonstrated by means of an experiment.

Software for automated classification of probe-based confocal laser endomicroscopy videos of colorectal polyps

AIM:To support probe-based confocal laser endomi-croscopy (pCLE) diagnosis by designing software for the automated classification of colonic polyps. METHODS:Intravenous fluorescein pCLE imaging of colorectal lesions was performed on patients under-going screening and surveillance colonoscopies, followed by polypectomies. All resected specimens were reviewed by a reference gastrointestinal pathologist blinded to pCLE information. Histopathology was used as the criterion standard for the differentiation between neoplastic and non-neoplastic lesions. The pCLE video sequences, recorded for each polyp, were analyzed off-line by 2 expert endoscopists who were blinded to the endoscopic characteristics and histopathology. These pCLE videos, along with their histopathology diagnosis, were used to train the automated classification software which is a content-based image retrieval technique followed by k-nearest neighbor classification. The performance of the off-line diagnosis of pCLE videos established by the 2 expert endoscopists was compared with that of automated pCLE software classification. All evaluations were performed using leave-one-patient- out cross-validation to avoid bias. RESULTS:Colorectal lesions (135) were imaged in 71 patients. Based on histopathology, 93 of these 135 lesions were neoplastic and 42 were non-neoplastic. The study found no statistical significance for the difference between the performance of automated pCLE software classification (accuracy 89.6%, sensitivity 92.5%, specificity 83.3%, using leave-one-patient-out cross-validation) and the performance of the off-line diagnosis of pCLE videos established by the 2 expert endoscopists (accuracy 89.6%, sensitivity 91.4%, specificity 85.7%). There was very low power (< 6%) to detect the observed differences. The 95% confidence intervals for equivalence testing were:-0.073 to 0.073 for accuracy, -0.068 to 0.089 for sensitivity and -0.18 to 0.13 for specificity. The classification software proposed in this study is not a "black box" but an informative tool based on the query by example model that produces, as intermediate results, visually similar annotated videos that are directly interpretable by the endoscopist. CONCLUSION:The proposed software for automated classification of pCLE videos of colonic polyps achieves high performance, comparable to that of off-line diagnosis of pCLE videos established by expert endoscopists.

Adaptive clustering hierarchy routing for delay tolerant network

Adaptive clustering hierarchy routing(ACHR) establishes a clusters-based hierarchical hybrid routing algorithm with two-hop local visibility for delay tolerant network(DTN).The major contribution of ACHR is the combination of single copy scheme and multi-copy scheme and the combination of hop-by-hop and multi-hop mechanism ACHR,which has the advantages in simplicity,availability and well-expansibility.The result shows that it can take advantage of the random communication opportunities and local network connectivity,and achieves 1.6 times delivery ratio and 60% overhead compared with its counterpart.

Identifying Anomaly Aircraft Trajectories in Terminal Areas Based on Deep Autoencoder and Its Application in Trajectory Clustering

Anomalous trajectory detection and traffic flow classification for complicated airspace are of vital importance to safety and efficiency analysis.Some researchers employed density-based unsupervised machine learning method to exploit these trajectories related to air traffic control(ATC)actions.However,the quality of position data and the tiny density difference between traffic flows in the terminal area make it particularly challenging.To alleviate these two challenges,this paper proposes a novel framework which combines robust deep auto-encoder(RDAE)model and density peak(DP)clustering algorithm.Specifically,the RDAE model is utilized to reconstruct denoising trajectory and identify anomaly trajectories in the terminal area by two different regularizations.Then,the nonlinear components captured by the encoder of RDAE are input in the DP algorithm to classify the global traffic flows.An experiment on a terminal airspace at Guangzhou Baiyun Airport(ZGGG)with anomaly label shows that the proposed combination can automatically capture non-conventional spatiotemporal traffic patterns in the aircraft movement.The superiority of RDAE and combination are also demonstrated by visualizing and quantitatively evaluating the experimental results.

KFL： a clustering algorithm for image database

It is a fairly challenging issue to make image repositories easy to be searched and browsed. This depends on a technique--image clustering. Kernel-based clustering algorithm has been one of the most promising clustering methods in the last few years, beeanse it can handle data with high dimensional complex structure. In this paper, a kernel fuzzy learning （KFL） algorithm is proposed, which takes advantages of the distance kernel trick and the gradient-based fuzzy clustering method to execute the image clustering automatically. Experimental results show that KFL is a more efficient method for image clustering in comparison with recent renorted alternative methods.

Behavior Clustering for Anomaly Detection

We presented a novel framework for automatic behavior clustering and unsupervised anomaly detection in a large video set. The framework consisted of the following key components： 1 ） Drawing from natural language processing, we introduced a compact and effective behavior representation method as a stochastic sequence of spatiotemporal events, where we analyzed the global structural information of behaviors using their local action statistics. 2） The natural grouping of behavior patterns was discovered through a novel clustering algorithm. 3 ） A run-time accumulative anomaly measure was introduced to detect abnormal behavior, whereas normal behavior patterns were recognized when sufficient visual evidence had become available based on an online Likelihood Ratio Test （LRT） method. This ensured robust and reliable anomaly detection and normal behavior recognition at the shortest possible time. Experimental results demonstrated the effectiveness and robustness of our approach using noisy and sparse data sets collected from a real surveillance scenario.

Interactive Protein Data Clustering

In this paper, the authors present three different algorithms for data clustering. These are Self-Organizing Map （SOM）, Neural Gas （NG） and Fuzzy C-Means （FCM） algorithms. SOM and NG algorithms are based on competitive leaming. An important property of these algorithms is that they preserve the topological structure of data. This means that data that is close in input distribution is mapped to nearby locations in the network. The FCM algorithm is an algorithm based on soft clustering which means that the different clusters are not necessarily distinct, but may overlap. This clustering method may be very useful in many biological problems, for instance in genetics, where a gene may belong to different clusters. The different algorithms are compared in terms of their visualization of the clustering of proteomic data.

Turnout fault diagnosis based on DBSCAN/PSO-SOM

In order to diagnose the common faults of railway switch control circuit,a fault diagnosis method based on density-based spatial clustering of applications with noise(DBSCAN)and self-organizing feature map(SOM)is proposed.Firstly,the three-phase current curve of the switch machine recorded by the micro-computer monitoring system is dealt with segmentally and then the feature parameters of the three-phase current are calculated according to the action principle of the switch machine.Due to the high dimension of initial features,the DBSCAN algorithm is used to separate the sensitive features of fault diagnosis and construct the diagnostic sensitive feature set.Then,the particle swarm optimization(PSO)algorithm is used to adjust the weight of SOM network to modify the rules to avoid“dead neurons”.Finally,the PSO-SOM network fault classifier is designed to complete the classification and diagnosis of the samples to be tested.The experimental results show that this method can judge the fault mode of switch control circuit with less training samples,and the accuracy of fault diagnosis is higher than that of traditional SOM network.

A bottom-up method for module-based product platform development through mapping,clustering and matching analysis

Designing product platform could be an effective and efficient solution for manufacturing firms. Product platforms enable firms to provide increased product variety for the marketplace with as little variety between products as possible. Developed consumer products and modules within a firm can further be investigated to find out the possibility of product platform creation. A bottom-up method is proposed for module-based product platform through mapping, clustering and matching analysis. The framework and the parametric model of the method are presented, which consist of three steps:(1) mapping parameters from existing product families to functional modules,(2) clustering the modules within existing module families based on their parameters so as to generate module clusters, and selecting the satisfactory module clusters based on commonality, and(3) matching the parameters of the module clusters to the functional modules in order to capture platform elements. In addition, the parameter matching criterion and mismatching treatment are put forward to ensure the effectiveness of the platform process, while standardization and serialization of the platform element are presented. A design case of the belt conveyor is studied to demonstrate the feasibility of the proposed method.

Research on Novel Natural Image Reconstruction and Representation Algorithm based on Clustering and Modified Neural Network

In this paper, we conduct research on the novel natural image reconstruction and representation algorithm based on clustenng and modified neural network. Image resolution enhancement is one of the earliest researches of single image interpolation. Although the traditional interpolation and method for single image amplification is effect, but did not provide more useful information. Our method combines the neural network and the clustering approach. The experiment shows that our method performs well and satisfactory.

Research on natural language recognition algorithm based on sample entropy

Sample entropy can reflect the change of level of new information in signal sequence as well as the size of the new information. Based on the sample entropy as the features of speech classification, the paper firstly extract the sample entropy of mixed signal, mean and variance to calculate each signal sample entropy, finally uses the K mean clustering to recognize. The simulation results show that： the recognition rate can be increased to 89.2% based on sample entropy.

World Expo 2010 Pavilions Clustering Analysis Based on Self-Organizing Map

This paper reports the classification of 90 sample pavilions in Shanghai World Expo. An artificial intelligence based nonlinear clustering method known as Self-Organizing Map(SOM) has been used to classify expo pavilions. SOM is an efficient tool for visualization of multidimensional data. To conduct the classification, four characteristics namely Hurst exponent for queue length, Hurst exponent for waiting time, mean queue length and mean waiting time have been applied. The classification results show that Shanghai World Expo pavilions can be optimally classified into four classes. This result will shed light on further studies that how to manage the queue of World Expo pavilions in the future.