Withthe rapid development of online shopping, the ability to intelligently collect and analyze information about E-shoppers has become a key source of competitive advantage for firms. this paper presents an optimal a...
详细信息
ISBN:
(纸本)9780769533049
Withthe rapid development of online shopping, the ability to intelligently collect and analyze information about E-shoppers has become a key source of competitive advantage for firms. this paper presents an optimal algorithm of modeling dynamic architecture for artificial neural networks (ANN) and a novel machine-learning algorithm for extracting rules from databases via using genetic algorithm. Classical architecture Of AAW is arbitrary. Before training ANN, the number of hidden layers and hidden nodes has already been fixed In the dynamic architecture, the number of hidden layers and the number of hidden nodes are sequentially and dynamically generated until a level of performance accuracy is reached In addition, in this paper, a new genetic algorithm is proposed, which does not need the computational complexity. the genetic algorithm is used to find the optimal values of input attributes (chromosome), X-m, which maximizes output function OK of output node k. the optimal chromosome is decoded and used to obtain a rule belonging to class k. the better result is achieved by applying the two new algorithms to a given database for customers buying computer.
In this paper, followed the assumption that the gene expression data of tumor may be sampled from the data with a probability distribution on a sub-manifold of ambient space, a supervised version of locally linear emb...
详细信息
ISBN:
(纸本)9783540859833
In this paper, followed the assumption that the gene expression data of tumor may be sampled from the data with a probability distribution on a sub-manifold of ambient space, a supervised version of locally linear embedding (LLE), named locally linear discriminant embedding (LLDE), is proposed for tumor classification. In the proposed algorithm, we construct a vector translation and distance rescaling model to enhance the recognition ability of the original LLE from two aspects. To validate the efficiency, the proposed method is applied to classify two different DNA microarray datasets. the prediction results show that our method is efficient and feasible.
this paper presents the use of coefficients derived from linear predictive coding (LPC) based on the dynamic time warping (DTW). the derived coefficients are called the DTW frame-fixing coefficients (DTW-FF), they are...
详细信息
ISBN:
(纸本)9780889867307
this paper presents the use of coefficients derived from linear predictive coding (LPC) based on the dynamic time warping (DTW). the derived coefficients are called the DTW frame-fixing coefficients (DTW-FF), they are used as input to the back-propagation neural network for speech patternrecognition. this paper also presents the study of pitch as a contributing input feature added to the DTW-FF coefficients. the results showed a good performance and improvement as high as 100% when using pitch along with DTW-FF feature. It is known that back-propagation NN is capable of handling large learning problems and is a very promising method due to its ability to train data and classify them. Current method of back- propagation is using steepest gradient descent whereby this method is exposed to bad local-minima. In this study, the network is designed to handle the parallel processing of multiplesamples/words, therefore it caused the network to compute a large amount of connection weights and error updates at a time, therefore longer time is taken for network convergence to its global minima. Since the Conjugate Gradient method has been proven of being able to accelerate the network convergence, it is applied into the back-propagation mechanism to replace the steepest gradient descent algorithm. the outcome showed that the convergence rate was improved when conjugate gradient method is used in the back-propagation algorithm.
this paper proposes a new texture image retrieval scheme based on contourlet transform and support vector machines (SVMs). In the scheme, the energies and the generalized Gaussian distribution (GGD) parameters are use...
详细信息
ISBN:
(纸本)9783540881919
this paper proposes a new texture image retrieval scheme based on contourlet transform and support vector machines (SVMs). In the scheme, the energies and the generalized Gaussian distribution (GGD) parameters are used to present the contourlet subband features. Using the representations, a two-run SVM retrieval algorithm which employs an one-class SVM followed by a two-class SVM is proposed to carry out the perceptual similarity measurement. For the query image, the one-class SVM is used to obtain the effective initial training set with positive and negative samples. Using these initial samples, the two-class SVM is applied to refine on the image classification subject to the user's relevance feedback. Compared with existing texture image retrieval methods, the proposes retrieval scheme is demonstrated respectively to be effective on the VisTex database of 640 texture images and the Brodatz database of 1760 texture images. Experimental results have shown that the proposed retrieval scheme can attain 99.38% and 98.07% of the average rates respectively for the two databases.
Predictive Toxicology (PT) is one of the newest targets of the Knowledge Discovery in databases (KDD) domain. Its goal is to describe the relationships between the chemical structure of chemical compounds and biologic...
详细信息
Predictive Toxicology (PT) is one of the newest targets of the Knowledge Discovery in databases (KDD) domain. Its goal is to describe the relationships between the chemical structure of chemical compounds and biological and toxicological processes. In real PT problems there is a very important topic to be considered: the huge number of the chemical descriptors. Irrelevant, redundant, noisy and unreliable data have a negative impact, therefore one of the main goals in KDD is to detect these undesirable proprieties and to eliminate or correct them. this assumes data cleaning, noise reduction and feature selection because the performance of the applied machinelearning algorithms is strongly related withthe quality of the data used. In this paper, we present some of the issues that can be taken into account for preparing data before the actual knowledge discovery is performed.
DNA microarray allows the measurement of transcript abundances for thousands of genes in parallel. though, it is an important procedure to select informative genes related to tumor from those gene expression profiles ...
详细信息
ISBN:
(纸本)9783540859833
DNA microarray allows the measurement of transcript abundances for thousands of genes in parallel. though, it is an important procedure to select informative genes related to tumor from those gene expression profiles (GEP) because of its characteristics such as high dimensionality, small sample set and many noises. In this paper we proposed a novel method for feature extraction that is named as Orthogonal Discriminant Projection (ODP). this method is a linear approximation base on manifold learning approach. the ODP method characterizes the local and non-local information of manifold distributed data and explores an optimum subspace which can maximize the difference between non-local scatter and the local scatter. Moreover, it introduces the class information to enhance the recognition ability. A trick has been employed to handle the Small Sample Site (SSS). Experimental results on Non-small Cell Lung Cancer (NSCLC) and glioma dataset validates its efficiency compared to other widely used dimensionality reduction methods such as Principle Component Analysis (PCA), Linear Discriminant Analysis (LDA).
Artificial Immune System (AIS) is all emerging technique for the classification task and proved to be a reliable technique. In previous studies, many classifiers including AIS classifiers require the data to be in num...
详细信息
ISBN:
(纸本)9783540850717
Artificial Immune System (AIS) is all emerging technique for the classification task and proved to be a reliable technique. In previous studies, many classifiers including AIS classifiers require the data to be in numerical or categorical data. types prior to processing. the transformation of data into any other specific types from their original form call degrade the originality of the data and consume more space and pre processing time. this paper introduces AIS model using immune network for classifying heterogeneous data in its original types. the model is able to process the data withthe types as represented in the database and it solves some bias problems highlighted in the AIS review papers. To ensure the consistent conditions and fair comparison, the selected existing algorithms use the same set of data as used in the proposed model. Experimental results show that this network-based model produces a better accuracy rate than the existing population-based immune algorithm and than the standard classifiers on most of the data from University of California, Irvive (UCI) machinelearning Repository (MLR) and University of California, Riverside (UCR) Time Series data (TSR).
Committee machines approach has shown to be useful in different applications. Protein primary structure data contain valuable information to extract. In this paper we mine these data and predict protein contact map ba...
详细信息
Committee machines approach has shown to be useful in different applications. Protein primary structure data contain valuable information to extract. In this paper we mine these data and predict protein contact map based on committee machines. Contact map is the simplified, two dimensional representation of protein spatial structure. Contact map prediction is of great interest due to its application in fold recognition and predicting protein tertiary structure. the results show that the performance of the committee is considerably better than a single model.
this book - in conjunction withthe two volumes CCIS 0015 and LNCS 5226 - constitutes the refereed proceedings of the 4thinternationalconference on Intelligent Computing, ICIC 2008, held in Shanghai, China, in Septe...
详细信息
ISBN:
(数字)9783540859840
ISBN:
(纸本)9783540859833
this book - in conjunction withthe two volumes CCIS 0015 and LNCS 5226 - constitutes the refereed proceedings of the 4thinternationalconference on Intelligent Computing, ICIC 2008, held in Shanghai, China, in September 2008. the 136 revised full papers were carefully reviewed and selected from 2336 submissions. the papers address all current issues in the field of intelligent computing technology, including neural networks, evolutionary computing and genetic algorithms, fuzzy systems and soft computing, particle swarm optimization and niche technology, supervised and semi-supervised learning, unsupervised and reinforcement learning, fine feature extraction methods, combinatorial and numerical optimization, neural computing and optimization, case based reasoning and autonomy-oriented computing, as well as artificial life and artificial immune systems. the volume is rounded off by three special sessions on computational analysis and datamining in biological systems, on datamining and fusion in bioinformatics, and on intelligent patternrecognition and understanding in image processing.
this paper focuses on neural networks with complex-valued (CV) neurons as well as on selected aspects of neural networks learning, pruning and rule extraction. CV neurons can be used as versatile substitutes in real-v...
详细信息
this paper focuses on neural networks with complex-valued (CV) neurons as well as on selected aspects of neural networks learning, pruning and rule extraction. CV neurons can be used as versatile substitutes in real-valued perceptron networks. learning of CV layers is discussed in context of traditional multilayer feedforward architecture. Such learning is derivative-free and it usually requires networks of reduced size. Selected examples and applications of CV-networks in bioinformatics and patternrecognition are discussed. the paper also covers specialized learning techniques for logic rule extraction. Such techniques include learning with pruning, and can be used in expert systems, and other applications that rely on models developed to fit measured data.
暂无评论