An Intelligent Medical Expert System Using Temporal Fuzzy Rules and Neural Classifier

As per World Health Organization report which was released in the year of 2019,Diabetes claimed the lives of approximately 1.5 million individuals globally in 2019 and around 450 million people are affected by diabetes all over the world.Hence it is inferred that diabetes is rampant across the world with the majority of the world population being affected by it.Among the diabetics,it can be observed that a large number of people had failed to identify their disease in the initial stage itself and hence the disease level moved from Type-1 to Type-2.To avoid this situation,we propose a new fuzzy logic based neural classifier for early detection of diabetes.A set of new neuro-fuzzy rules is introduced with time constraints that are applied for thefirst level classification.These levels are further refined by using the Fuzzy Cognitive Maps(FCM)with time intervals for making thefinal decision over the classification process.The main objective of this proposed model is to detect the diabetes level based on the time.Also,the set of neuro-fuzzy rules are used for selecting the most contributing values over the decision-making process in diabetes prediction.The proposed model proved its efficiency in performance after experiments conducted not only from the repository but also by using the standard diabetic detection models that are available in the market.

Multi-source Fuzzy Information Fusion Method Based on Bayesian Optimal Classifier

为了做常规贝叶斯的最佳的分类器，拥有处理模糊信息并且认识到推理过程的自动化的能力，一个新贝叶斯的最佳的分类器被建议，模糊信息嵌入。它不能仅仅有效地处理模糊信息，而且保留贝叶斯的最佳的分类器的学习性质。另外根据模糊集合理论的进化，含糊的集合也是嵌入的进它产生含糊的贝叶斯的最佳的分类器。它能同时从积极、反向的方向模仿模糊信息的双重的特征。进一步，贝叶斯的最佳的分类器也是的集合对从积极、反向、不确定的方面就模糊信息的三方面的特征而言求婚了。最后，一个知识库的人工的神经网络(KBANN ) 被介绍认识到贝叶斯的最佳的分类器的自动推理。它不仅减少贝叶斯的最佳的分类器的计算费用而且改进它学习质量的分类。

An Evolving Fuzzy Classifier for Induction Motor Health Condition Monitoring

Induction motor (IM) is commonly used in various industrial applications. Reliable online IM health condition monitoring systems are critically needed in industries to improve operational accuracy and safety of the IMs and the machinery. A new evolving algorithm is proposed to provide more decision-making transparency, as well as better classification and processing efficiency. The effectiveness of the developed intelligent classifier is examined by simulation and experimental tests.

Keystroke Dynamics Based Authentication Using Possibilistic Renyi Entropy Features and Composite Fuzzy Classifier

This paper presents the formulation of the possibilistic Renyi entropy function from the Renyi entropy function using the framework of Hanman-Anirban entropy function. The new entropy function is used to derive the information set features from keystroke dynamics for the authentication of users. A new composite fuzzy classifier is also proposed based on Mamta-Hanman entropy function and applied on the Information Set based features. A comparison of the results of the proposed approach with those of Support Vector Machine and Random Forest classifier shows that the new classifier outperforms the other two.

Design of Polynomial Fuzzy Neural Network Classifiers Based on Density Fuzzy C-Means and L2-Norm Regularization

In this paper, polynomial fuzzy neural network classifiers (PFNNCs) is proposed by means of density fuzzy c-means and L2-norm regularization. The overall design of PFNNCs was realized by means of fuzzy rules that come in form of three parts, namely premise part, consequence part and aggregation part. The premise part was developed by density fuzzy c-means that helps determine the apex parameters of membership functions, while the consequence part was realized by means of two types of polynomials including linear and quadratic. L2-norm regularization that can alleviate the overfitting problem was exploited to estimate the parameters of polynomials, which constructed the aggregation part. Experimental results of several data sets demonstrate that the proposed classifiers show higher classification accuracy in comparison with some other classifiers reported in the literature.

改进模糊推理分类器进行木材树种近红外光谱开集分类识别研究

开集分类识别是近10多年来模式识别领域研究的热点,它能够识别训练集中已知类别的测试样本,同时还能够有效“拒识”未知类别的测试样本;这些未知类别样本不包含在训练集中。现有的开集分类识别算法主要是基于Support Vector Machine(SVM)和深度学习网络框架进行改进,并且主要应用在自然景物图像领域中;在光谱分析领域中还鲜有报道。将传统的闭集框架下的模糊推理分类器进行模型改进,提出了开集框架下的改进模糊推理分类器,并将其应用到木材树种近红外光谱分类识别中。首先,使用Flame-NIR近红外微型光谱仪采集木材样本横切面的近红外光谱曲线,采用Metric Learning算法进行光谱向量维度约简降维至4维(4D)。其次,改进闭集框架下的模糊推理分类器,根据模糊规则置信度和各维度隶属度概率的乘积构建Generalized Basic Probability Assignment(GBPA),再根据GBPA进行分类处理。在20个树种的具有不同的Openness指标下的近红外光谱数据集的分类识别对比实验表明,改进的开集模糊推理分类器(fuzzy reasoning classifier in an open set,FRCOS)优于现有的基于机器学习和深度学习的开集分类识别主流算法,具有较好的评价指标F-Score,Kappa系数及总体识别率。

GIS-based landslide susceptibility modeling:A comparison between fuzzy multi-criteria and machine learning algorithms

Hazards and disasters have always negative impacts on the way of life.Landslide is an overwhelming natural as well as man-made disaster that causes loss of natural resources and human properties throughout theworld.The present study aimed to assess and compare the prediction efficiency of different models in landslide susceptibility in the Kysuca river basin,Slovakia.In this regard,the fuzzy decision-making trial and evaluation laboratory combining with the analytic network process(FDEMATEL-ANP),Naïve Bayes(NB)classifier,and random forest(RF)classifier were considered.Initially,a landslide inventory map was produced with 2000 landslide and nonlandslide points by randomly dividedwith a ratio of 70%:30%for training and testing,respectively.The geospatial database for assessing the landslide susceptibility was generated with the help of 16 landslide conditioning factors by allowing for topographical,hydrological,lithological,and land cover factors.The ReliefF methodwas considered for determining the significance of selected conditioning factors and inclusion in the model building.Consequently,the landslide susceptibility maps(LSMs)were generated using the FDEMATEL-ANP,Naïve Bayes(NB)classifier,and random forest(RF)classifier models.Finally,the area under curve(AUC)and different arithmetic evaluation were used for validating and comparing the results and models.The results revealed that random forest(RF)classifier is a promising and optimum model for landslide susceptibility in the study area with a very high value of area under curve(AUC=0.954),lower value of mean absolute error(MAE=0.1238)and root mean square error(RMSE=0.2555),and higher value of Kappa index(K=0.8435)and overall accuracy(OAC=92.2%).

A Multiple Model Approach to Modeling Based on Fuzzy Support Vector Machines

A new multiple models(MM) approach was proposed to model complex industrial process by using Fuzzy Support Vector Machines(F -SVMs). By applying the proposed approach to a pH neutralization titration experiment, F -SVMs MM not only provides satisfactory approximation and generalization property, but also achieves superior performance to USOCPN multiple modeling method and single modeling method based on standard SVMs.

Using FCM to Select Samples in Semi-Supervised Classification

For a semi-supervised classification system, with the increase of the training samples number, the system needs to be continually updated. As the size of samples set is increasing, many unreliable samples will also be increased. In this paper, we use fuzzy c-means （FCM） clustering to take out some samples that are useless, and extract the intersection between the original training set and the cluster after using FCM clustering. The intersection between every class and cluster is reliable samples which we are looking for. The experiment result demonstrates that the superiority of the proposed algorithm is remarkable.

FUZZY SET STUDY OF WATER MASS MIXING IN THE SOURCE REGION OF THE TSUSHIMA WARM CURRENT

PFS-Fuzzy classification ( Lu, 1989) was used on observational data obtained during a cruise (July-August】 1987)to classify the water masses in the source area of the Tsushima Warm Current. Their mixing features were studied by using numerical index analysis of fuzzy sets. The calculated results showed there are nine water masses belonging to three basic types.The analyses suggest that, though, in summer, the Surface Water of the Tsushima Warm Current located in a strongly mixed area is a mixture of the East China Sea Mixed Water, the Kuroshio Surface Water and the Kyushu Western Coastal Water, it originates mainly from the Kuroshio Surface Water and its deep water comes from the Kuroshio Subsurface Water. This study reveals that 1) regions such as the intensely mixed region, the frontal zone and the transition zone, Water, it originates deep water comes from water, usually have a higher fuzzy degree ; 2) water masses with higher stability and little modification have a lower fuzzy degree ; and 3)

Diagnosis of Neem Leaf Diseases Using Fuzzy-HOBINM and ANFIS Algorithms

This paper proposes an approach to detecting diseases in neem leaf that uses a Fuzzy-Higher Order Biologically Inspired Neuron Model(F-HOBINM)and adaptive neuro classifier(ANFIS).India exports USD 0.28-million worth of neem leaf to the UK,USA,UAE,and Europe in the form of dried leaves and powder,both of which help reduce diabetesrelated issues,cardiovascular problems,and eye disorders.Diagnosing neem leaf disease is difficult through visual interpretation,owing to similarity in their color and texture patterns.The most common diseases include bacterial blight,Colletotrichum and Alternaria leaf spot,blight,damping-off,powdery mildew,Pseudocercospora leaf spot,leaf web blight,and seedling wilt.However,traditional color and texture algorithms fail to identify leaf diseases due to irregular lumps and surfaces,and rough ridges,as the classification time involved takes as long as a week.The proposed F-HOBINM algorithm recognizes the leaf intensity through the leaky capacitor,and uses subjective intensity and physical stimulus to interpret the diagnosis.Further,the processed leaf images from the HOBINM algorithm are applied to the ANFIS classifier to identify neem leaf diseases.The experimental results show 92.18%accuracy from a database of 1,462 neem leaves.

Design of Hybrid Fuzzy Neural Network for Function Approximation

In this paper, a hybrid Fuzzy Neural Network (FNN) system for function approximation is presented. The proposed FNN can handle numeric and fuzzy inputs simultaneously. The numeric inputs are fuzzified by input nodes upon presentation to the network while the Fuzzy rule based knowledge is translated directly into network architecture. The connections between input to hidden nodes represent rule antecedents and hidden to output nodes represent rule consequents. All the connections are represented by Gaussian fuzzy sets. The method of activation spread in the network is based on a fuzzy mutual subsethood measure. Rule (hidden) node activations are computed as a fuzzy inner product. For a given numeric o fuzzy input, numeric outputs are computed using volume based defuzzification. A supervised learning procedure based on gradient descent is employed to train the network. The model has been tested on two different approximation problems: sine-cosine function approximation and Narazaki-Ralescu function and shows its natural capability of inference, function approximation, and classification.

内存有效的快速双层深度规则分类器

深度规则分类器在处理大规模或高复杂度的分类任务时,模糊规则库的数量会过于庞大,导致其内部结构变得复杂,可解释性较差,测试时间也随着模糊规则数量的增加而线性递增,且在移动端设备内存有限的条件下无法保证内存的有效性.针对这些问题,提出一个新的内存有效的快速双层深度规则分类器,该分类器在深度规则分类器的基础上增设数据汇总和高描述性原型提取两个模块.仅当原型数量达到内存上限时执行一次数据汇总模块来删除部分原型,解决有限条件下内存不足无法训练出有效分类器的问题.高描述性原型提取模块将原型划分为底层和顶层的双层结构,底层由所有原型组成,用于展示分类器内部结构的全貌,顶层由少量高描述性原型组成,用于分类决策阶段.这样能有效地防止深度规则分类器在处理大规模数据时,因其复杂的内部结构导致较差的可解释性,同时提高分类决策阶段的效率.在基准数据集上的仿真实验验证了该方法的可行性和有效性.内存有效的快速双层深度规则分类器在处理大规模数据或数据流问题时,在达到较高分类性能的同时,保证模型的透明性、解释性和内存的有效性.

基于模糊关联的不平衡数据分类算法研究

由于信息时代数据量的爆炸式增加,高精度分类海量数据中的稀有种类数据是数据挖掘领域研究的热点。为提升不平衡数据集分类精度,深入研究基于Python的不平衡分类数据模糊关联混合算法,基于不平衡数据爬虫得到不平衡分类数据,采用AdaBoost.M1W集成学习算法训练不平衡数据集,从分类精度、分类器效率和分类器规模3方面进行对比。仿真实验结果表明,所提算法在自然不平衡数据集和人工不平衡数据集都具有较高精度,分类性能较优。

基于改进层次全局模糊熵和MCFS的滚动轴承损伤识别

针对传统的多尺度特征提取方法无法捕捉振动信号高频故障信息的问题,提出了一种基于改进层次全局模糊熵(IHGFE)全局全频段特征提取、多聚类特征选择(MCFS)特征降维和支持向量机分类的滚动轴承故障诊断方法。首先,提出了能够捕捉振动信号低频到高频的全局特征的IHGFE非线性动力学方法,并将其用于滚动轴承的故障特征提取;然后,利用MCFS对初始特征向量进行了维数约简和优化,构建了低维且对故障敏感的故障特征向量;最后,建立了基于支持向量机的多故障分类器,实现了滚动轴承损伤的智能化识别,并通过两个滚动轴承实验进行了对比分析。研究结果表明:IHGFE的分类准确率和识别稳定性均优于对比方法,证明了其在特征提取中能够在一定程度上解决现有方法无法同时考虑信号的高频特征和全局特征的问题,可为进一步扩展模糊熵方法在滚动轴承损伤识别中的应用提供参考。

面向不平衡数据的深度TSK模糊分类器

为了进一步提升Takagi-Sugeno-Kang(TSK)模糊分类器在不平衡数据集上的泛化能力和保持其较好的语义可解释性,受集成学习的启发,提出面向不平衡数据的深度TSK模糊分类器(A Deep TSK Fuzzy Classifier for Imbalanced Data,ID-TSK-FC).ID-TSK-FC主要由一个不平衡全局线性回归子分类器(Imbalanced Global Linear Regression Sub-Classifier,IGLRc)和多个不平衡TSK模糊子分类器(Imbalanced TSK Fuzzy Sub-Classifier,I-TSK-FC)组成.根据人类“从全局粗糙到局部精细”的认知行为和栈式叠加泛化原理,ID-TSK-FC首先在所有原始训练样本上训练一个IGLRc,获得全局粗糙的分类结果.然后根据IGLRc的输出,识别原始训练样本中的非线性分布训练样本.在非线性分布训练样本上,以栈式深度结构生成多个局部I-TSK-FC,获得局部精细的结果.最后,对于栈式堆叠IGLRc和所有I-TSK-FC的输出,使用基于最小距离投票原理,得到ID-TSK-FC的最终输出.实验表明,ID-TSK-FC不仅具有基于特征重要性的可解释性,而且具有至少相当的泛化性能和语义可解释性.

Optimised hybrid classification approach for rice leaf disease prediction with proposed texture features

This paper aims to frame a new rice disease prediction model that included three major phases.Initially,median filtering(MF)is deployed during pre-processing and then‘proposed Fuzzy Means Clustering(FCM)based segmentation’is done.Following that,‘Discrete Wavelet Transform(DWT),Scale-Invariant Feature Transform(SIFT)and low-level features(colour and shape),Proposed local Binary Pattern(LBP)based features’are extracted that are classified via‘MultiLayer Perceptron(MLP)and Long Short Term Memory(LSTM)’and predicted outcomes are obtained.For exact prediction,this work intends to optimise the weights of LSTM using Inertia Weighted Salp Swarm Optimisation(IW-SSO)model.Eventually,the development of IW-SSO method is established on varied metrics.

基于模糊关联空间的激光雷达三维扫描数据过滤研究

激光雷达三维扫描数据中存在大量无效点和有效点,但是两者之间的空间关联性阈值处在不断变动中,以固定关联规则的过滤方式很难准确区分,造成实际过滤应用效果较差。为此,提出基于模糊关联空间的激光雷达三维扫描无效数据过滤方法。采集激光雷达三维扫描样本数据构建无效数据识别规则库,将无效数据识别规则库中的无效数据集合转换为模糊集合,并利用聚类算法填补无效数据造成的数据空格,以避免出现数据过滤误差。填补后运用支持向量机分类器计算激光雷达三维扫描数据之间的模糊关联度,并对其进行分类。结合分类结果利用网闸过滤组织对激光雷达三维扫描无效数据过滤。实验结果表明:这种方法过滤精度高,具有较好的过滤效果,时间复杂度低、过滤耗时短,实际应用效果较好。

一类模糊重构度分类器

作为经典的分类器,模糊K近邻分类器(FKNN),有着较为广泛的应用,且衍生出许多改进分类器。不同于模糊K近邻分类器使用简单的投票策略构建模糊集,提出一种新的分类器,即基于重构度模糊分类器。得益于对样本重构方法的改良,新的分类器避免了传统FKNN需要根据不同的数据集合调整相应的K参数的不足,而是以数据集为驱动,进而无参数调整。同时,由于重构度从结构上对噪声所产生的影响有一定抑制作用,所以该分类器对于图片噪声的鲁棒性较强。实验结果也表明基于重构度的模糊分类器在加噪声的AR人脸库上都取得了超越其他类FKNN的表现。