this paper researched a neural networks based knowledge discovery and datamining (KDDM) methodology based on granular computing, neural computing, fuzzy computing, Linguistic computing, and patternrecognition. A gra...
详细信息
the Inhibition-Compensation learning Scheme (ICLS) has been proposed as a way of enhancing the performance of the Moving Window Classifier In this paper, the effect of ICLS on three n-tuple based classification techni...
详细信息
ISBN:
(纸本)076951695X
the Inhibition-Compensation learning Scheme (ICLS) has been proposed as a way of enhancing the performance of the Moving Window Classifier In this paper, the effect of ICLS on three n-tuple based classification techniques has been investigated. Pre-segmented handwritten characters from the NIST database have been used as the patterndata. Results show that approximately 2-6% gain in classification accuracy can be achieved in the OCR task domain with no adverse effect on the classification throughput.
the proceedings contain 49 papers. the special focus in this conference is on Application of Discovery to Natural Science, Knowledge Discovery from Unstructured and Semi-structured data. the topics include: Mathematic...
ISBN:
(纸本)3540001883
the proceedings contain 49 papers. the special focus in this conference is on Application of Discovery to Natural Science, Knowledge Discovery from Unstructured and Semi-structured data. the topics include: Mathematics based on learning;datamining with graphical models;on the eigenspectrum of the gram matrix and its relationship to the operator eigenspectrum;in search of the horowitz factor;learning structure from sequences, with applications in a digital library;discovering frequent structured patterns from string databases;discovery in hydrating plaster using machinelearning methods;revising qualitative models of gene regulation;structure extraction using summaries;model complexity and algorithm selection in classification;experiments with projection learning;improved dataset characterisation for meta-learning;racing committees for large datasets;from ensemble methods to comprehensible models;learningthe causal structure of overlapping variable sets;extraction of logical rules from data by means of piecewise-linear neural networks;structuring neural networks through bidirectional clustering of weights;toward drawing an atlas of hypothesis classes;datascape survey using the cascade model;learning hierarchical skills from observation;image analysis for detecting faulty spots from microarray images;inferring gene regulatory networks from time-ordered gene expression data using differential equations;modeling state transition of typhoon image sequences by spatio-temporal clustering;structure-sweetness relationships of aspartame derivatives by GUHA;a hybrid approach for Chinese named entity recognition;extraction of word senses from human factors in knowledge discovery;event pattern discovery from the stock market bulletin and email categorization using fast machinelearning algorithms.
In this paper, we synthesize the main findings of three repeat purchase modelling case studies using real-life direct marketing data. Historically, direct marketing - more recently, targeted web marketing - has been o...
详细信息
ISBN:
(纸本)9729805067
In this paper, we synthesize the main findings of three repeat purchase modelling case studies using real-life direct marketing data. Historically, direct marketing - more recently, targeted web marketing - has been one of the most popular domains for the exploration of the feasibility and the viable use of novel business intelligence techniques. Many a datamining technique has been field tested in the direct marketing domain. this can be explained by the (relatively) low-cost availability of recency, frequency, monetary (RFM) and several other customer relationship data, the (relatively) well-developed understanding of the task and the domain, the clearly identifiable costs and benefits, and because the results can often be readily applied to obtain a high return on investment. the purchase incidence modelling cases reported on in this paper were in the first place undertaken to trial run state-of-the-art supervised Bayesian learning multilayer perceptron (MLP) and least squares support vector machine (LS-SVM) classifiers. For each of the cases, we also aimed at exploring the explanatory power (relevance) of the available RFM and other customer relationship related variable operationalizations for predicting purchase incidence in the context of direct marketing.
this article presents a survey of models of rough neurocomputing that have their roots in rough set theory. Historically, rough neurocomputing has three main threads: training set production, calculus of granules, and...
详细信息
Functional Electrical Stimulation (FES) is an effective and developing method to restore functions for paraplegic patients. In this research, we focused on the switching problem of FES, which is one of the obstacles t...
详细信息
Functional Electrical Stimulation (FES) is an effective and developing method to restore functions for paraplegic patients. In this research, we focused on the switching problem of FES, which is one of the obstacles that prevent FES from further practical uses. An adaptive switching for FES control for the lower limbs' activities of hemiplegic patients was developed, based on the consideration that, lower limbs' activities need the synchronization of limbs of both sides. Electromyogram (EMG) signals detected from normal side of hemiplegic patients were used to recognize the activities that the patients intend to do. the recognition results were utilized as the switching signals. However, motion patterns represented and analyzed by EMG are distinctive of individual variations and characteristic alternation, which inevitably result in classification errors in EMG analyzing. To overcome these problems, a feed-forward artificial neural network (ANN) was embedded in an on-line process to form an analyzing system that can adapt to individual characteristics and trace the nonstationary factor. Furthermore, in order to enable the analyzing system to recognize right timings from the EMG-described dynamical processes of activities, such as standing-up and walking, a practical training-set construction method that utilizes additional reference data was proposed. the proposed switching system was applied to a FES system that supports standing-up and walking for a hemiplegics subject, to verify the effectiveness. (C) 2002 Elsevier Science B.V. All rights reserved.
In patternrecognition, the goal of classification can be achieved from two different types of learning strategy-discriminative teaming and informative learning. Discriminative learning focuses on extracting the discr...
详细信息
ISBN:
(纸本)9810475241
In patternrecognition, the goal of classification can be achieved from two different types of learning strategy-discriminative teaming and informative learning. Discriminative learning focuses on extracting the discriminative information between classes. Informative learning emphasizes the learning of the class information such as class densities. We review major discriminative learning methods, namely, principal component analysis (PCA), linear discriminant analysis (LDA), minimum classification error (MCE) training algorithm and support vector machine (SVM) and one informative learning method-Gaussian mixture models (GMM). We also discuss the combination of the two types of learning and give the corresponding experiments results.
A method for pattern classification on large-scale training data is presented in this paper, which is based upon the genetic algorithm (GA) and support vector machine (SVM). the initial training data are optimized wit...
详细信息
ISBN:
(纸本)0780374886
A method for pattern classification on large-scale training data is presented in this paper, which is based upon the genetic algorithm (GA) and support vector machine (SVM). the initial training data are optimized with GA in order to find a sample subset including the important samples that can preserve or improve the discrimination ability of SVM. Training on the subset is equal to that on the initial sample sets. the training time is greatly shortened. Following the result, we take advantage of the excellent classification performance of SVM to accomplish the pattern classification.
A new method, called adaptive projective algorithm, which is able to extract support vectors from given training examples, is put forward as a support vector algorithm. the method greatly reduces the training samples ...
详细信息
ISBN:
(纸本)0780374886
A new method, called adaptive projective algorithm, which is able to extract support vectors from given training examples, is put forward as a support vector algorithm. the method greatly reduces the training samples and so improves the speed of the support vector machine, while the ability of the support vector machine in pattern classification is unaffected: Our experimental results show remarkable improvement of speed to support our idea.
Functional Electrical Stimulation (FES) is an effective and developing method to restore functions for paraplegic patients. In this research, we focused on the switching problem of FES, which is one of the obstacles t...
详细信息
ISBN:
(纸本)1586030787
Functional Electrical Stimulation (FES) is an effective and developing method to restore functions for paraplegic patients. In this research, we focused on the switching problem of FES, which is one of the obstacles that prevent FES from further practical uses. An adaptive switching for FES control for the lower limbs' activities of hemiplegic patients was developed, based on the consideration that, lower limbs' activities need the synchronization of limbs of both sides. Electromyogram (EMG) signals detected from normal side of hemiplegic patients were used to recognize the activities that the patients intend to do. the recognition results were utilized as the switching signals. However, motion patterns represented and analyzed by EMG are distinctive of individual variations and characteristic alternation, which inevitably result in classification errors in EMG analyzing. To overcome these problems, a feed-forward artificial neural network (ANN) was embedded in an on-line process to form an analyzing system that can adapt to individual characteristics and trace the nonstationary factor. Furthermore, in order to enable the analyzing system to recognize right timings from the EMG-described dynamical processes of activities, such as standing-up and walking, a practical training-set construction method that utilizes additional reference data was proposed. the proposed switching system was applied to a FES system that supports standing-up and walking for a hemiplegics subject, to verify the effectiveness. (C) 2002 Elsevier Science B.V. All rights reserved.
暂无评论