We first examine the techniques,development,and application future of the current recommender systems in the film *** recommendation techniques in current applications and the K-nearest neighbor(***) algorithm,in pa...
详细信息
We first examine the techniques,development,and application future of the current recommender systems in the film *** recommendation techniques in current applications and the K-nearest neighbor(***) algorithm,in particular,is then introduced in *** is followed by an introduction to the expectationmaximization(***) algorithm based on the Bayesian classifier,which has been applied to the classification and similarity calculations of ***,the movieeviews data in the NLTK(Natural Language Toolkit)library is used to facilitate *** evaluate the classification accuracy of the KNN algorithm and the EM algorithm based on the Bayesian *** experimental results demonstrate that,the classification accuracy of the EM algorithm for films is higher than that of the KNN algorithm and it is feasible and useful to apply the EM algorithm to films classification.
Magnetic Resonance Image segmentation is a fundamental task in a wide variety of computed-based medical applications that support therapy, diagnostic and medical applications. In this work, spatial information is incl...
详细信息
ISBN:
(纸本)9783642166860
Magnetic Resonance Image segmentation is a fundamental task in a wide variety of computed-based medical applications that support therapy, diagnostic and medical applications. In this work, spatial information is included for estimating paramaters of a finite mixture model, with gaussian distribution assumption, using a modified version of the well-know expectation maximization algorithm proposed in [3]. Our approach is based on aggregating a transition step between E-step and M-step, that includes the information of spatial dependences between neighboring pixels. Our proposal is compared with other approaches proposed in the image segmentation literature using the size and shape test, obtaining accurate and robust results in the presence of noise.
DNA microarrays provide a simple tool to identify and quantify the gene expression for tens of thousands of genes simultaneously. Image processing is an important step in microarrays experiments. This paper presents a...
详细信息
ISBN:
(纸本)9781424447138
DNA microarrays provide a simple tool to identify and quantify the gene expression for tens of thousands of genes simultaneously. Image processing is an important step in microarrays experiments. This paper presents a novel technique for removing gene's noises based on the offset vector field and segmenting genes using the expectation maximization algorithm. Simulations show that the new technique for microarray images filtering and segmentation has better performance than most of the common ways. The results of experiments are computationally attractive, have excellent performance and can preserve spots' data while efficiently suppress noises.
The assumptions on the hidden Markov model will limit the output of a model to be a piecewise stationary random sequence that may not be a good fit for real processes. In this paper we propose a piecewise polynomial h...
详细信息
ISBN:
(纸本)9781509043149
The assumptions on the hidden Markov model will limit the output of a model to be a piecewise stationary random sequence that may not be a good fit for real processes. In this paper we propose a piecewise polynomial high-order hidden Markov model so that the output of a model can be more versatile. We derived formulas for the calculation of the probability that a given sequence is produced by a model. We also derived the posterior probability of the states and the state transitions and used them in an expectation maximization algorithm to update the model parameters. Experiments on speech recognition of Mandarin digits were conducted to investigate the effectiveness of the proposed model. Experimental results showed that the proposed model can reduce the recognition error rate compared to a baseline hidden Markov model.
A key issue in mission planning for aerial reconnaissance is to use the sensor ressources in an appropriate way. The sensor mission planning requires knowledge, e.g., about the optimal sensor type (IR/EO) or the neces...
详细信息
ISBN:
(纸本)9781510600928
A key issue in mission planning for aerial reconnaissance is to use the sensor ressources in an appropriate way. The sensor mission planning requires knowledge, e.g., about the optimal sensor type (IR/EO) or the necessary flying altitude for a specific task. There are various types of task that can be part of the mission, e.g., to detect a vehicle or investigate a bridge. The goal of this work is to examine knowledge-based approaches like ontologies and the use of them to automatically derive all needed parameters for an optimal sensor mission planning based on the task. The task-oriented mission planning is processed on the tactical level. Based on the task the aerial image analyst defines the specific evaluation conditions. Various parameters are part of the task-oriented mission planning. For example there are the scene, the flight mode, the sensor, the system and the image analysis parameters. We introduce an idea to represent the sensor mission planning task at a relatively high-level knowledge-based approach. We want to create a representation of a useful sensor mission planning where all aspects and its components are considered, what these parts do, how they relate to each other and define the rules and constraints within. The main framework is based on the definitions of target categories and purpose codes (STANAG 3596) and NIIRS level.
Classical wireless communication technologies are threatened with so many challenges for meeting the desires of ubiquity and mobility for the cellular systems. Hostile wireless channels features and restricted frequen...
详细信息
ISBN:
(纸本)9781509002108
Classical wireless communication technologies are threatened with so many challenges for meeting the desires of ubiquity and mobility for the cellular systems. Hostile wireless channels features and restricted frequency bandwidths are obstacles in future generation systems. In order to deal with these limitations, different advanced signal processing approaches, such as expectation-maximization (EM) algorithm, SAGE algorithm, Baum-Welch algorithm, Kalman filters and their extensions etc. were proposed. In this paper, estimation of unknown channel parameter and detection of data at receiver end has been performed. MIMO Rayleigh and Rician channels are taken for wireless communication. To find the initial point for EM algorithm FCM clustering algorithm is used. In this work, algorithm is implemented using MATLAB R2012a. The performance matrices of the algorithm are bit error rate (BER) and mean square error (MSE) at different values of signal to noise ratio.
In this paper, we consider Generalized Gaussian (GG) distribution to model the additive noise source in underwater acoustic (UWA) communication. Since communication in oceanic medium is dominated by both prevailing an...
详细信息
ISBN:
(纸本)9781509017461
In this paper, we consider Generalized Gaussian (GG) distribution to model the additive noise source in underwater acoustic (UWA) communication. Since communication in oceanic medium is dominated by both prevailing and spontaneous noise sources, we model the resultant noise distribution as mixture of GG distribution. Owing to the complexity in optimal detector design with GG noise model, we apply expectationmaximization (EM) algorithm to decompose the resultant channel distribution in terms of weighted sum of Gaussian density functions. By having multiple antennas at the receiver, we also exploit spatial diversity to improve error performance at the receiver. In this context, we compute decision boundary for detecting the binary phase shift keying (BPSK) modulated signal. In addition, we also discuss variation in decision boundary under various signal to noise ratio (SNR) levels observed at receiver front end. Finally, we compare the detector performance under new decision boundary with traditional detectors and validate the approach by showing improvement in symbol error rate performance.
This paper considers the problem of networks reconstruction from heterogeneous data using a Gaussian Graphical Mixture Model (GGMM). It is well known that parameter estimation in this context is challenging due to lar...
详细信息
This paper considers the problem of networks reconstruction from heterogeneous data using a Gaussian Graphical Mixture Model (GGMM). It is well known that parameter estimation in this context is challenging due to large numbers of variables coupled with the degenerate nature of the likelihood. We propose as a solution a penalized maximum likelihood technique by imposing an l1 penalty on the precision matrix. Our approach shrinks the parameters thereby resulting in better identifiability and variable selection. Cet article considère le probléme de la reconstruction de réseaux à partir de données hétérogènes en utilisant le modèle graphique gaussien mélangé (GGMM en Anglais). Il est connu que l'estimation paramétrique, dans ce contexte, n'est pas aisé à cause du grand nombre de variable et de la nature dégénérée de la vraisemblance. Nous proposons comme une solution une méthode de pénalisation du maximum de vraisemblance en imposant une pénalité de type l1 sur la précision de la matrice. Notre méthode réduit les paramétres et ainsi aboutit à une meilleure identification et à un meilleur choix des variables.
We consider a system defined as a collection of two types of components. The number of failures of each component is described as a stochastic process, with one of the processes depending on the other. None of the pro...
详细信息
We consider a system defined as a collection of two types of components. The number of failures of each component is described as a stochastic process, with one of the processes depending on the other. None of the processes is observed directly. The only available information is the number of type 1 components at risk in the system. Because of this missing data situation, different algorithms relying on an expectationmaximization (EM) strategy are proposed to obtain the MLE of the intensity parameters for both processes so we can assess the reliability of type 1 and type 2 components. To overcome the computational limits of EM, a Monte Carlo EM (MCEM) algorithm using a Metropolis procedure is presented. Stochastic EM (SEM) algorithms including a Bayesian approach are also described. The methods are applied to simulated data to demonstrate their efficiency.
Terrain and landscape complexities can limit the accurate discrimination of land use categories with similar spectral signatures, as well as the accurate detection of land use change in temporal analyses of landscape ...
详细信息
Terrain and landscape complexities can limit the accurate discrimination of land use categories with similar spectral signatures, as well as the accurate detection of land use change in temporal analyses of landscape dynamics. Studies based on misclassified land use data can generate biased parameter estimates and standard errors, inaccurate predictions, and incorrect policy recommendations. To address these challenges and improve the accuracy of land use analyses, we implement a post-classification strategy to detect misclassified land use observations using a latent multinomial logit model. This strategy is tested using both Monte Carlo simulations and a time series dataset based on supervised classification of remotely sensed data corresponding to land use decisions observed in a Mexican coffee growing region during the period 1984-2006. The results indicate that the strategy is useful for identifying land use observations with a high probability of being wrongly classified, even between categories with low discriminative spectral signatures. Reclassification of the land use data, based on the model results, increases the magnitudes of the marginal effects of the analyzed land use drivers in the theoretically expected directions, and in some cases improves the statistical significance of the parameter estimates. (C) 2015 Elsevier Inc. All rights reserved.
暂无评论