datamining is an expanding research frontier that provides numerous efficient and scalable methods to extract patterns of interest in datasets. In this paper , Computer Aided Diagnosis ( CAD ) is applied to brain MRI...
详细信息
datamining is an expanding research frontier that provides numerous efficient and scalable methods to extract patterns of interest in datasets. In this paper , Computer Aided Diagnosis ( CAD ) is applied to brain MRI image processing. Four features based on texture as proposed by Harlick are extracted and stored in a transactional database. the system is then trained withthe proposed efficient associative classifier. the existing CBA algorithm was extended to select only essential rules which help diagnosis of abnormal MRI of the brain. Our work is optimized in the sense it combines feature selection and discretization thereby reducing the mining complexity. the results showed higher sensitivity ( upto 98% ) and accuracy ( upto 97% ) allowing us to claim that association rules can effectively aid in the diagnosing task.
the proceedings contain 67 papers. the topics discussed include: on concentration of discrete distributions with applications to supervised learning of classifiers;multi-source data modelling: integrating related data...
详细信息
ISBN:
(纸本)9783540734987
the proceedings contain 67 papers. the topics discussed include: on concentration of discrete distributions with applications to supervised learning of classifiers;multi-source data modelling: integrating related data to improve model performance;an empirical comparison of ideal and empirical ROC-based reject rules;outlier detection with kernel density functions;generic probability density function reconstruction for randomization in privacy-preserving datamining;an incremental fuzzy decision tree classification method for miningdata streams;on the combination of locally optimal pairwise classifiers;an agent-based approach to the multiple-objective selection of reference vectors;on applying dimension reduction for multi-labeled problems;nonlinear feature selection by relevance feature vector machine;a bounded index for cluster validity;and varying density spatial clustering based on a hierarchical tree.
the CENIT research project GAD (Demand Active Management), partly supported by the Spanish government, has as its objective the demand side management (DSM) of domestic end-users, smoothing the peaks of energy demand,...
详细信息
the CENIT research project GAD (Demand Active Management), partly supported by the Spanish government, has as its objective the demand side management (DSM) of domestic end-users, smoothing the peaks of energy demand, therefore enhancing the conditions of transport and distribution networks and the quality of energy delivery and service. One of the objectives in the development of the project is the classification of users according to their patterns of daily energy load profiles. Based on daily hour measures from a sample of residential users, a self-organizing map (SOM) has been trained to classify users according to a specific number of resulting patterns of daily load profiles, attached to a number of indices that define the users' energy consumption. the so-trained SOM classifier allows the characterization of future users based on their load profiles, thus estimating their energy consumption habits and potentially manageable energy.
Nowadays, the power production and transmission are substantial elements for the society, and will probably play a more important role in the future, but are also associated with various negative effects, which impose...
详细信息
Nowadays, the power production and transmission are substantial elements for the society, and will probably play a more important role in the future, but are also associated with various negative effects, which impose additional costs, the so called external costs. As long as these are not included in the electricity prices, market failures most often arise. In order to define expansion projects, right price signals are needed. thus, the internalization of external costs in power production, which typically leads to higher marginal production costs for conventional power plants, can help in the identification of sustainable future transmission and production plans. the aggregated benefit of the internalization of external costs is usually higher than the social welfare deficit, depending on the energy mix, and can be further used in order to reinforce the existing electricity network, financing ldquogreenrdquo investments. the idea in this paper is the coordination of generation, transmission and policy planning, and their interaction, towards a common electricity market. the proposed methodology is based on a cost benefit analysis in order to support the final decision taken. Several scenarios, considering demand changes, power plants installations and investments in transmission are implemented by means of a European ldquocopperplaterdquo model.
By applying the method of support vector machine, we carry on a classification study of existing diagram data of the Chinese herbal medicine fingerprint. We compared the effect of support vector machine algorithm on t...
详细信息
ISBN:
(纸本)9781424422388
By applying the method of support vector machine, we carry on a classification study of existing diagram data of the Chinese herbal medicine fingerprint. We compared the effect of support vector machine algorithm on two and several types of data identification withthat of the existing computer classification methods. It is shown that the support vector machine method can be used to identify Chinese herbal medicine fingerprint diagram. Some suggestions are put forward to identifying the Chinese herbal medicine fingerprint diagram.
Dimensionality reduction with prior information is considered. the semi-supervised Laplacian eigenmap algorithm is proposed. It is shown that the performance of dimensionality reduction algorithms can be improved by t...
详细信息
ISBN:
(纸本)9781424422388
Dimensionality reduction with prior information is considered. the semi-supervised Laplacian eigenmap algorithm is proposed. It is shown that the performance of dimensionality reduction algorithms can be improved by taking into account the label, information of the data. the data analysis and experiments show the validity of our algorithm.
the problem of learning a kernel is considered based on minimizing the generalization bounds. According to the bounds, a bi-regularization criterion is developed for learning a kernel from the data. the relations betw...
详细信息
ISBN:
(纸本)9781424422388
the problem of learning a kernel is considered based on minimizing the generalization bounds. According to the bounds, a bi-regularization criterion is developed for learning a kernel from the data. the relations between the criterion and some established criteria, such as kernel-target alignment and the regularization criterion, is discussed. Using the relations, we connect the kernel-target alignment and the generalization of kernel-based algorithms. Moreover, we consider the kernel-learning problem withthe bi-regularization criterion when the kernel is in the convex hull of basic kernels which are continuously parameterized by a compact set. We show that there always exists an optimal kernel which is the convex combination of at most n+1 basic kernels, where n is the sample size. And a saddle theorem is developed to characterize the optimal kernel.
Manifold learning has currently become a hot issue in the field of machinelearning, patternrecognition and datamining. Locally linear embedding (LLE) is one of several promising manifold learning methods. But ordin...
详细信息
ISBN:
(纸本)9780769533056
Manifold learning has currently become a hot issue in the field of machinelearning, patternrecognition and datamining. Locally linear embedding (LLE) is one of several promising manifold learning methods. But ordinary LLE can not distinguish effectively the low-dimensional embeddings of noise data. By introducing the reconstruction similarity into LLE, this paper proposes a generalized locally linear embedding algorithm based on local reconstruction similarity. Experimental results show on Columbia object image datathat the new generalized version is superior to LLE in revealing the visualization of high-dimensional image dataset containing noise images.
Feature selection is an important task in machinelearning, patternrecognition and datamining. this paper proposed a new feature selection method for classification, named SD, which is based on scatter matrix used i...
详细信息
ISBN:
(纸本)9781424420957
Feature selection is an important task in machinelearning, patternrecognition and datamining. this paper proposed a new feature selection method for classification, named SD, which is based on scatter matrix used in linear discriminant analysis. the main feature of SD is its simplicity and independency of learning algorithms. High-dimensional data samples are first projected into a lower dimensional subspace of the original feature space by means of a linear transformation matrix, which can be attained according to the scatter degree of each feature, and then the scatter degree is used to measure the importance of each feature. A comparison of SD and some popular feature selection methods (information gain and X-2-test) is conducted, and the results of experiment carried out on 19 data sets show the advantages of SD.
the present study deals withthe analysis and mapping of Swiss franc interest rates. Interest rates depend on time and maturity, defining term structure of the interest rate curves (IRC). In the present study IRC are ...
详细信息
the present study deals withthe analysis and mapping of Swiss franc interest rates. Interest rates depend on time and maturity, defining term structure of the interest rate curves (IRC). In the present study IRC are considered in a two-dimensional feature space - time and maturity. Exploratory data analysis includes a variety of tools widely used in econophysics and geostatistics. Geostatistical models and machinelearning algorithms (multilayer perceptron and Support Vector machines) were applied to produce interest rate maps. IR maps can be used for the visualisation and pattern perception purposes, to develop and to explore economical hypotheses, to produce dynamic asset-liability simulations and for financial risk assessments. the feasibility of an application of interest rates mapping approach for the IRC forecasting is considered as well. (C) 2008 Elsevier B.V. All rights reserved.
暂无评论