In many applications such as dealing with database, continuous environment and humanoid robots, the machine often deals with large amount of data every day of work. Dealing with large amount of data requires fast as w...
详细信息
ISBN:
(纸本)9781479956869
In many applications such as dealing with database, continuous environment and humanoid robots, the machine often deals with large amount of data every day of work. Dealing with large amount of data requires fast as well as accurate learning algorithms to do the classification. A new supervised non parametric Partial Histogram Bayes learning algorithm (PHBayes) is proposed and presented in this paper. The proposed algorithm was tested on image database and compared with other standard algorithms like Naive Bayes, Gaussian Mixture Model based Classifier, 1st Nearest Neighbor and Nearest Class Mean for classification purpose. The experimental results showed that the proposed algorithm is faster as well as more accurate compare with other algorithms, which makes it worthy to be considered in classification applications.
The objective of the current study was the development of a reliable modeling platform to calculate in real time the personal exposure and the associated health risk for filling station employees evaluating current en...
详细信息
The objective of the current study was the development of a reliable modeling platform to calculate in real time the personal exposure and the associated health risk for filling station employees evaluating current environmental parameters (traffic, meteorological and amount of fuel traded) determined by the appropriate sensor network. A set of Artificial Neural Networks (ANNs) was developed to predict benzene exposure pattern for the filling station employees. Furthermore, a Physiology Based Pharmaco-Kinetic (PBPK) risk assessment model was developed in order to calculate the lifetime probability distribution of leukemia to the employees, fed by data obtained by the ANN model. bayesian algorithm was involved in crucial points of both model sub compartments. The application was evaluated in two filling stations (one urban and one rural). Among several algorithms available for the development of the ANN exposure model, bayesian regularization provided the best results and seemed to be a promising technique for prediction of the exposure pattern of that occupational population group. On assessing the estimated leukemia risk under the scope of providing a distribution curve based on the exposure levels and the different susceptibility of the population, the bayesian algorithm was a prerequisite of the Monte Carlo approach, which is integrated in the PBPK-based risk model. In conclusion, the modeling system described herein is capable of exploiting the information collected by the environmental sensors in order to estimate in real time the personal exposure and the resulting health risk for employees of gasoline filling stations.
bayesian and maximum likelihood algorithms are considered. A comparative analysis of their effectiveness is performed. The results are specified for an image in the shape of an ellipse with linearly varying intensity.
bayesian and maximum likelihood algorithms are considered. A comparative analysis of their effectiveness is performed. The results are specified for an image in the shape of an ellipse with linearly varying intensity.
Long-term wireless neural recording systems which are subject to stringent power consumption, are highly desired to reduce the rate of data transmission and computation complexity. In this paper, we propose using a co...
详细信息
ISBN:
(纸本)9781467363525;9781467363518
Long-term wireless neural recording systems which are subject to stringent power consumption, are highly desired to reduce the rate of data transmission and computation complexity. In this paper, we propose using a combination of on-chip neural action potentials ('spikes') detection system and compressive sensing (CS) techniques to reduce the power required for data transmission and a random Bernoulli matrix to reduce the computation complexity consequently further reduce the power consumption. By analyzing the data detected by nano platinum black modified microelectrode array implanted in the hippocampus of the Sprague-Dawley (SD) rat, we prove that spikes are compressible in the wavelet domain. We use the bayesian CS algorithm to reconstruct them. Our results show that the mean compression ratio is 26: 1 achieved for 16-dB SNDR recovery using this mechanism.
Weight of criteria can only be changed through discussion or analysis of experts under traditional supplier selection method, which brings up two problems: (1) human effect problem and (2) incapability in timely decis...
详细信息
ISBN:
(纸本)9781479909865
Weight of criteria can only be changed through discussion or analysis of experts under traditional supplier selection method, which brings up two problems: (1) human effect problem and (2) incapability in timely decision making on updates of weights. In this paper, a new supplier selection method based on integration of Analytic Hierarchy Process, the bayesian Classifier algorithm and dynamic probabilities (AHP-BCA) is proposed. The method makes predictions with the probabilities of occurrences of criteria values based on historical records to avoid any human effects in decision making. It is also equipped with an instant self-update function to instantly update the probability values with new data, and be ready for next calculation. A simulation experiment is conducted to compare the performance of the proposed approach with a remarkable traditional approach in literature with historical data. Results show that the proposed approach can outperform
A new Cloud-based Trust Awareness and Interaction Model (CTAIM) is proposed for trust evaluation based content filtering in social interactive data. The research is based on diversified information from social interac...
详细信息
ISBN:
(纸本)9781479928293
A new Cloud-based Trust Awareness and Interaction Model (CTAIM) is proposed for trust evaluation based content filtering in social interactive data. The research is based on diversified information from social interactive data to analyse people's intention and evaluate people's trust. The proposed model is composed of bayesian content filtering algorithm and bayesian inference algorithm in Dirichlet distribution, and it's capable to provide 3rd party trustworthiness evaluation according to node behavior and interaction history with high-efficiency, security, and neutrality. Additionally, MapReduce-based computing and HBase storage framework are implemented for parallel computing among mass interactive data.
According to the problem of the constant increase of internet data and the current Internet information retrieval system has been unable to meet the needs of data and information retrieval, this paper puts forward the...
详细信息
ISBN:
(纸本)9781479927166
According to the problem of the constant increase of internet data and the current Internet information retrieval system has been unable to meet the needs of data and information retrieval, this paper puts forward the data mining based Internet information retrieval system. The paper first uses data mining to excavate association rules, and evaluates the entire behavior simulation of users resources need according to a certain factor of user's retrieval resources, and then uses bayesian network algorithm to correlate the mining data. Experimental simulation results show that the application of bayesian network algorithm into Internet information retrieval has implemented intelligence and personalization of information retrieval to a certain extent, which has certain research value.
With an upsurge in biomedical literature,using data-mining method to search new knowledge from literature has drawing more attention of *** this study,taking the mining of non-coding gene literature from the network d...
详细信息
With an upsurge in biomedical literature,using data-mining method to search new knowledge from literature has drawing more attention of *** this study,taking the mining of non-coding gene literature from the network database of PubMed as an example,we first preprocessed the abstract data,next applied the term occurrence frequency(TF) and inverse document frequency(IDF)(TF-IDF) method to select features,and then established a biomedical literature data-mining model based on bayesian ***,we assessed the model through area under the receiver operating characteristic curve(AUC),accuracy,specificity,sensitivity,precision rate and recall *** 1 000 features are selected,AUC,specificity,sensitivity,accuracy rate,precision rate and recall rate are 0.868 3,84.63%,89.02%,86.83%,89.02% and 98.14%,*** results indicate that our method can identify the targeted literature related to a particular topic effectively.
Background: We assessed the specificity of wide QRS complex tachycardia (WCT) differentiating algorithms in patients with preexistent left bundle branch block (LBBB) and heart failure. Methods: Three hundred fourteen ...
详细信息
Background: We assessed the specificity of wide QRS complex tachycardia (WCT) differentiating algorithms in patients with preexistent left bundle branch block (LBBB) and heart failure. Methods: Three hundred fourteen patients with resynchronization devices were retrospectively screened. electrocardiograms with supraventricular LBBB rhythm were used as a surrogate for supraventricular tachycardia QRS morphology. The Pava lead II criterion, ventricular activation velocity ratio (Vi/Vt) ratio in V-2, Vereckei aVR, Brugada, Griffith, and bayesian algorithms were investigated. Results: The WCT algorithms had a lower specificity (33%-69%) in patients with LBBB than in general WCT populations. The Pava lead II criterion and Brugada algorithm had higher specificity than other algorithms (P<.05). Several of the single criteria (absence of an RS complex in V-1 through V-6, initial R wave in aVR, Vi/Vt <1 in V-2) had specificities of 92% to 99%. Conclusions: In patients with heart failure and LBBB, an electrocardiographic diagnosis of ventricular tachycardia should be based on selected, specific criteria rather than on WCT algorithms. (C) 2012 Elsevier Inc. All rights reserved.
暂无评论