基于ART2的网络入侵检测算法

基于ART2的网络入侵检测算法是在自适应共振理论的基础上改进而来的。该算法对接收到的网络数据以及系统状态数据进行分析判断,实现入侵方式的自动分类,并且能够对新产生的入侵方式进行分类与记忆,实现了入侵检测系统的自适应性。该算法应用到入侵检测系统中能够解决入侵检测系统中可能出现的预分类不完全的问题,这对于检测新出现的入侵类型无疑具有很大的使用价值。     

快速分类神经网络在字符识别中的应用

本文以海明神经网络与自适应谐振理论(ART)模型学习算法为基础,从理论上分析了海明网络学习算法的缺陷,利用ART网络的思想,提出了一种快速分类的神经元网络的算法,命名为Improved Hamming算法(简称Im-H算法)。此算法主要优点在于阈值更新及引入了经验迭代次数。将此算法用于字符模式识别,大量的计算机实验结果表明了Im-H网络学习算法的有效性、快速性。     

基于ART-人工免疫网络的数据浓缩方法研究

基于智能互补融合观点,提出了一种新的数据浓缩方法.该方法整合了自适应共振网络(ART)和人工免疫网络各自的优点,人工免疫网络利用ART提供的有关系统的先验知识构造网络,对数据样本进行浓缩,获得原始数据集的内部映像,为进一步处理提供良好的数据样本.本文详细讨论了算法步骤,还提出新的数据浓缩度量指标--数据蒸发率.对R2空间分类和Fisher花瓣问题的实验结果表明,该方法有效克服了人工免疫网络计算量大等缺点,结果可靠,且数据蒸发率高.     

一种新的ART网络遥感影像分类方法

本文提出了一种基于模糊算子的ART2A-C遥感影像分类算法。算法结合原有几种高性能的ART网络对传统ART2A-C网络做了改进。论文分别利用现有网络和改进算法对遥感影像作了聚类,实验结果表明新算法的分类性能明显优于原算法。     

一种新的ART网络遥感影像分类方法

本文提出了一种基于模糊算子的ART2A-C遥感影像分类算法.算法结合原有几种高性能的ART网络对传统ART2A-C网络做了改进.论文分别利用现有网络和改进算法对遥感影像作了聚类,实验结果表明新算法的分类性能明显优于原算法.     

一种基于最小风险检测ART网分类器的设计与人脸识别

本文通过对传统的自适应共振网络(ART)的研究,分析了其优缺点及应用领域,在此基础上提出了一种适合训练样本数较少、样本特征值维数较高的图像目标分类的最小风险检测 ART 网络,并重点讨论了该网络在结构上的改进和运行原理,以及基于该网络的分类器的设计及其算法实现过程。通过对 ORL 人脸库和 Yale 人脸库的图像样本仿真结果表明,应用该分类方法进行人脸图像分类识别,同时获得了较高的分类速度和分类效果。     

慢速权值更新的ART2神经网络研究

ART2是基于自适应谐振理论的一种自组织神经网络,通过竞争学习和自稳机制原理实现分类,可以在非平稳的、有干扰的环境中进行无监督的自学习,其学习过程能迅速识别已学习过的样本,并能迅速适应未学习过的新对象。提出了一种基于慢速权值更新的ART2神经网络算法,该算法在对输入模式进行识别分类时,会减慢学习速率,降低模式漂移的速度。新的网络学习规则在分类实验中取得了较好的效果,并在一定程度上解决了模式漂移问题。     

一种抗漂移的改进ART2网络GSC-ART2研究

ART2是一种基于自适应谐振理论的无监督神经网络,由于其快速响应、实时学习等特点,被广泛地应用在实时聚类问题中。传统的ART2存在幅值信息丢失、容易产生模式漂移的问题,本文针对此不足提出了一种基于广义相似度和置信度的GSCART2网络。其通过引入广义相似度检测和竞争机制,解决了幅值信息丢失的问题。置信度结合广义相似度的权值调整方式抑制了模式漂移并使网络的连接权值更加准确。通过实验表明,GSC-ART2网络在处理幅值相关、样本渐变分类问题上的识别性能均优于传统ART2网络,从而证明了此GSC-ART2网络的有效性,也为解决模式识别中普遍存在的模式漂移问题找到了一种优良的解决方法。     

一种融合胶囊网络的分类方法

目前的ADMET分类方法在对具有多特征性和特征关联性的化合物数据进行ADMET分类时存在不足。而且,对ADMET分类结果不具备可解释性。针对上述问题,提出一种融合胶囊网络的分类模型（CapsMC）。CapsMC模型首先提出一种feature-to-image图像转换算法。使用该算法将特征之间的关联关系和依赖关系作为考量纳入到分类依据中,实现特征的多层次提取。其次,探索胶囊网络的高级应用,提出一种认知推理机制。使用该机制对特征进行认知推理,实现ADMET的可解释性分类。模型在五种ADMET数据集上的实验结果表明,CapsMC模型可以高效实现ADMET的可解释性分类。     

ART—2A的同相位不可分问题及其解决方法

自适应共振理论（ART）是一种典型的、无监督的、能够对复杂输入模式实现自组织识别的神经网络,本文经分析发现标准ART－2A算法中存在“同相位不可分问题”,由此提出新的F1层非线性变换函数和F2层竞争学习算法,把ART－2A算法的适用范围扩展到整个实数域,然后提出了相应的线性变换处理方法把“同相位不可分问题”转化为“整个坐标平面内的相位划分问题”,从而很好地解决上述问题,增强了ART－2A算法的适用性。验证算例的结果表明,新型ART－2A算法能够对更为广泛的数据模式进行自适应识别,分类结果准确。     

改进ART Ⅱ算法的仿真研究

ART Ⅱ网络以模式的相似性量度值为基础,能够对动态的输入模式样本进行自适应的聚类和识别,然而标准的ART Ⅱ网络在输入数据处理过程中,忽略了样本数据中的负数信息和幅值信息,造成信号畸变和"同相位不可分"问题,在权值调整过程中,聚类中心发生移动,容易造成"模式漂移"现象。针对上述问题结合相关文献提出了引入非线性函数对输入数据进行变换的方法解决"同相位不可分"问题,用待测数据与同一模式类中有限数据的欧氏距离与限定值进行比较实现聚类判定,抑制"模式漂移"现象。用Matlab仿真表明,改进算法性能优于标准算法。     

A RT 2神经网络的一种改进

传统A RT 2神经网络在聚类过程中模式的匹配度量仅仅与模式的相位信息相关,这种匹配度量忽略了模式的幅度信息的作用,在对相位信息相同而幅度信息不同的两个簇进行聚类时,效果很差;同时,它还存在输入域限制的问题。针对这些不足之处,提出了一种改进的A RT 2神经网络,在输入模式进入网络学习过程中,保存其幅值信息,放宽对负实数的非线性转换,并考虑输入模式到各个簇的中心点的最短距离,同时增加一个阈值对离群点进行判定,消除了离群点对聚类结果的影响。实验验证,改进的A RT 2网络在对相同相位的两个簇聚类时,性能明显优于传统的A RT 2网络。     

Degree of familiarity ART2 in knowledge-based landmine detection

The self-organizing network ART2 is extended to provide a fuzzy output value, which indicates the degree of familiarity of a new analog input pattern to previously stored patterns in the long-term memory of the network. The outputs of the multilayer perceptron and this modified ART2, provide an analog value to a fuzzy rule-based fusion technique which also uses a processed polarization resolved image as its third input. In real-time situations these two classifier outputs indicate the likelihood of a surface landmine target when presented with a number of multispectral and textural bands. Due to the modifications in ART2 this updated alternative architecture has improved real-time landmine detection capabilities although the registration of all bands is more critical to the accuracy of results in this case. The real-time fuzzy rule-based system in preliminary tests has detected two of the three landmines and the landmine surrogate with two false alarms. Advanced tests on 30 images using the fuzzy rule-based system further confirmed the distinct advantages of fusion and improved detection rates.     

A topological and temporal correlator network for spatiotemporalpattern learning, recognition, and recall

In this paper, we describe the design of an artificial neural network for spatiotemporal pattern recognition and recall. This network has a five-layered architecture and operates in two modes: pattern learning and recognition mode, and pattern recall mode. In pattern learning and recognition mode, the network extracts a set of topologically and temporally correlated features from each spatiotemporal input pattern based on a variation of Kohonen's self-organizing maps. These features are then used to classify the input into categories based on the fuzzy ART network. In the pattern recall mode, the network can reconstruct any of the learned categories when the appropriate category node is excited or probed. The network performance was evaluated via computer simulations of time-varying, two-dimensional and three-dimensional data. The results show that the network is capable of both recognition and recall of spatiotemporal data in an online and self-organized fashion. The network can also classify repeated events in the spatiotemporal input and is robust to noise in the input such as distortions in the spatial and temporal content.     

Self-organizing QRS-wave recognition in ECG using neural networks.

The author has developed a self-organizing QRS-wave recognition system for electrocardiograms (ECGs) using neural networks. An ART2 (adaptive resonance theory) network was employed in this self-organizing neural-network system. The system consists of a preprocessor, an ART2, network, and a recognizer. The preprocessor detects R points in the ECG and divides the ECG into cardiac cycles. A QRS-wave is the part of the ECG that is between a Q point and an S point. The input to the ART2 network is one cardiac cycle from which the ART2 network indicates the approximate locations of both the Q and S points. The recognizer establishes search regions for the Q and S points. Then, it locates the Q and S points in each search region. The system uses this method to recognize a QRS-wave. Then, the ART2 network learns the new QRS-wave pattern from the incoming ECG. The ART2 network self-organizes in response to the input ECG. The average recognition error of the present system is less than 1 ms in the recognition of the Q and S points.     

基于ART2改进算法的故障聚类研究

ART2（自适应谐振理论2）算法是神经网络中一种可以对模拟输入信号或二值信号进行无监督聚类的算法,所以ART2算法能够降低数据挖掘中原始数据的预处理的复杂度,提高挖掘效率。针对ART2算法中出现的聚类中心偏移的缺点,采用ART2算法与K-均值算法相结合的方法来抑制ART2中聚类中心偏移的现象。通过仿真对该方法进行了验证。     

Polytope ARTMAP: pattern classification without vigilance based on general geometry categories

This paper proposes polytope ARTMAP (PTAM), an adaptive resonance theory (ART) network for classification tasks which does not use the vigilance parameter. This feature is due to the geometry of categories in PTAM, which are irregular polytopes whose borders approximate the borders among the output predictions. During training, the categories expand only towards the input pattern without category overlap. The category expansion in PTAM is naturally limited by the other categories, and not by the category size, so the vigilance is not necessary. PTAM works in a fully automatic way for pattern classification tasks, without any parameter tuning, so it is easier to employ for nonexpert users than other classifiers. PTAM achieves lower error than the leading ART networks on a complete collection of benchmark data sets, except for noisy data, without any parameter optimization.     

New approaches to nesting rectangular patterns

In this study, two approaches are explored for the solution of the rectangular stock cutting problem: neuro-optimization, which integrates artificial neural networks and optimization methods; and genetic neuro-nesting, which combines artificial neural networks and genetic algorithms. In the first approach, an artificial neural network architecture is used to generate rectangular pattern configurations, to be used by the optimization model, with an acceptable scrap. Rectangular patterns of different sizes are selected as input to the network to generate the location and rotation of each pattern after they are combined. A mathematical programming model is used to determine the nesting of different sizes of rectangular patterns to meet the demand for rectangular blanks for a given planning horizon. The test data used in this study is generated randomly from a specific normal distribution. The average scrap percentage obtained is within acceptable limits. In the second approach, a genetic algorithm is used to generate sequences of the input patterns to be allocated on a finite width with infinite-length material. Each gene represents the sequence in which the patterns are to be allocated using the allocation algorithm developed. The scrap percentage of each allocation is used as an evaluation criterion for each gene for determining the best allocation while considering successive generations. The allocation algorithm uses the sliding method integrated with an artificial neural network based on the adaptive resonance theory (ART1) paradigm to allocate the patterns according to the sequence generated by the genetic algorithm. It slides an incoming pattern next to the allocated ones and keeps all scrap areas produced, which can be utilized in allocating a new pattern through the ART1 network. If there is a possible match with an incoming pattern and one of the scrap areas, the neural network selects the best match area and assigns the pattern. Both approaches gave satisfactory results. The second approach generated nests having packing densities in the range 95–97%. Improvement in packing densities was possible at the expense of excessive computational time. Parallel implementation of this unconventional approach could well bring a quick and satisfactory solution to this classical problem.     

一种改进的ART2网络学习算法

分析了现有ART2网络存在的问题,提出了一种改进的ART2算法。该算法首先利用样本数据自身来初始化权值,然后按照同一类中的数据点到其聚类中心的距离之和越小(即类内偏差越小),聚类效果越好的原则来设计特征表示场和类别表示场之间的权值修正公式,最后通过比较输入样本和聚类中心的模来有效地利用模式的幅度信息。分析证明了该算法不仅能有效解决模式漂移问题、充分利用幅度信息,而且能提高聚类速度。