Effective connectivity analysis of fMRI data based on network motifs

Exploring effective connectivity between neuronal assemblies at different temporal and spatial scales is an important issue in human brain research from the perspective of pervasive computing. At the same time, network motifs play roles in network classification and analysis of structural network properties. This paper develops a method of analyzing the effective connectivity of functional magnetic resonance imaging (fMRI) data by using network motifs. Firstly, the directed interactions between fMRI time-series are analyzed based on Granger causality analysis (GCA), by which the complex network is built up to reveal the causal relationships among different brain regions. Then the effective connectivity in complex network is described with a variety of network motifs, and the statistical properties of fMRI data are characterized according to the network motifs topological parameters. Finally, the experimental results demonstrate that the proposed method is feasible in testing and measuring the effective connectivity of fMRI data.     

A Bayesian multilevel model for fMRI data analysis

Bayesian approaches have been proposed by several functional magnetic resonance imaging (fMRI) researchers in order to overcome the fundamental limitations of the popular statistical parametric mapping method. However, the difficulties associated with subjective prior elicitation have prevented the widespread adoption of the Bayesian methodology by the neuroimaging community. In this paper, we present a Bayesian multilevel model for the analysis of brain fMRI data. The main idea is to consider that all the estimated group effects (fMRI activation patterns) are exchangeable. This means that all the collected voxel time series are considered manifestations of a few common underlying phenomena. In contradistinction to other Bayesian approaches, we think of the estimated activations as multivariate random draws from the same distribution without imposing specific prior spatial and/or temporal information for the interaction between voxels. Instead, a two-stage empirical Bayes prior approach is used to relate voxel regression equations through correlations between the regression coefficient vectors. The adaptive shrinkage properties of the Bayesian multilevel methodology are exploited to deal with spatial variations, and noise outliers. The characteristics of the proposed model are evaluated by considering its application to two real data sets.     

A topological approach to dynamic graph connectivity

A dynamic connectivity problem consists of an initial graph, and a sequence of operations consisting of graph modifications and graph connectivity tests. The size n of the problem is the sum of the maximum number of vertices and edges of the derived graph, plus the number of operations to be executed. Each graph modification is a deletion of either an edge or an isolated vertex. Each graph connectivity test is to determine if there exists a path in the current graph between two given vertices (the vertices can vary for distinct tests). The best previously known time for this dynamic connectivity problem was Ω(n²).Our main result is an O(ng+n log n) time algorithm for the dynamic connectivity problem in the case of the maximum genus of the derived graph being g.     

A new approach for determining the spectral data of multichannel harmonic signals in noise

In this paper we present a new recursive procedure for identifying both the frequencies and amplitude matrices corresponding to a multichannel harmonic signal in additive correlated noise. The procedure follows from a complete characterization of the eigenstructure of the adjoint operator of the Naimark dilation for the observation correlation sequence. This research was supported in part by the Research Fund of Indiana University and the Natural Science Foundation under Grant No. ECS 841935.     

A dynamic Bayesian network data fusion algorithm for estimating leaf area index using time-series data from in situ measurement to remote sensing observations

Leaf area index (LAI) products retrieved from remote sensing observations have been widely used in the fields of ecosphere, atmosphere etc. However, because satellite-observed images are captured instantaneously and sometimes screened by cloud, some current LAI products are inherently discontinuous in time and their accuracy may not meet the needs of users well. To solve these problems, we proposed a dynamic Bayesian network (DBN)-based data fusion algorithm that integrates dynamic crop growth information, a canopy reflectance (CR) model and remote sensing observations from the perspective of Bayesian probability. Using the proposed algorithm, LAI was estimated using data sets from both field measurements for winter wheat in Beijing, China, and MODIS reflectance data at two American flux tower sites. Results showed good agreement between the LAI estimated by the DBN-based data fusion method and the true ground LAI, with a correlation coefficient of (R) 0.95 and 0.96, respectively, and a corresponding root mean square error (RMSE) of 0.35 and 0.49, respectively. In addition, the LAI estimated by the DBN-based data fusion method formed a continuous time series and was consistent with the variety law of vegetation growth at both plot and flux tower site scales. It has been demonstrated that the proposed DBN-based data fusion algorithm has the potential to be used to accurately estimate LAI and to fill the temporal gap by integrating information from multiple sources.     

A neural network approach for data masking

In this letter we present a neural network based data masking solution, in which the database information remains internally consistent yet is not inadvertently exposed in an interpretable state. The system differs from the classic data masking in the sense that it can understand the semantics of the original data and mask it using a neural network which is a priori trained by some rules. Our adaptive data masking (ADM) concentrates on data masking techniques such as shuffling, substitution, masking and number variance in an intelligent fashion with the help of adaptive neural network. The very nature of being adaptive makes data masking easier and content agnostic, and thus finds place in various vertical domains and systems.     

Bayesian robust linear dynamic system approach for dynamic process monitoring

In this paper, a Bayesian robust linear dynamic system approach is proposed for process modeling. Traditional linear dynamic system (LDS) constructed with Kalman filter is designed by Gaussian assumption which can be easily violated in non-Gaussian modeling situations, especially those with outliers. To deal with this issue, the conventional Gaussian-based Kalman filter is modified with heavy tailed Student's t-distribution so as to deal with the non-Gaussian noise and modeling outliers. Then, a variational Bayesian expectation maximization (VBEM) algorithm is developed for learning parameters of the robust linear dynamic system. For process monitoring, traditional monitoring scheme are discussed and the residual space monitoring mechanism has been improved. To explore the feasibility and effectiveness, the proposed method is applied for fault detection, with detailed comparative studies with several other methods through the Tennessee Eastman benchmark.     

A semiparametric Bayesian approach for joint-quantile regression with clustered data

Based on a semiparametric Bayesian framework, a joint-quantile regression method is developed for analyzing clustered data, where random effects are included to accommodate the intra-cluster dependence. Instead of posing any parametric distributional assumptions on the random errors, the proposed method approximates the central density by linearly interpolating the conditional quantile functions of the response at multiple quantiles and estimates the tail densities by adopting extreme value theory. Through joint-quantile modeling, the proposed algorithm can yield the joint posterior distribution of quantile coefficients at multiple quantiles and meanwhile avoid the quantile crossing issue. The finite sample performance of the proposed method is assessed through a simulation study and the analysis of an apnea duration data.     

基于动态贝叶斯网络的社交圈归属匹配模型

为了帮助社交网中新成员寻找与之最为合适的社交圈,尝试采用动态贝叶斯网（DBN）理论解决社交网分析应用中成员（节点）与社交圈（集合）的匹配问题。将圈内成员个人的多项兴趣爱好程度作为描述社交圈基本属性的特征向量,对每一类圈子建立了带有辅助信息形式的DBN模型,求解最大输出概率即为最佳匹配对象。结果表明,在客观测试和主观评价两方面,该模型都收到了较为满意的结果。     

混合式贝叶斯网络结构增量学习算法研究

提出一种混合式贝叶斯网络结构增量学习算法.首先提出多项式时间的限制性学习技术,为每个变量建立候选父节点集合;然后,依据候选父节点集合,利用搜索技术对当前网络进行增量学习.该算法的复杂度显著低于目前最优的贝叶斯网络增量学习算法.理论与实验均表明,所处理的问题越复杂,该算法在计算复杂度方面的优势越明显.     

一种基于Bayesian信念网络的客户行为预测方法

提出一种基于Bayesian信念网络（BN）的客户行为预测方法．通过知识学习构建客户行为Bayesian网络（CBN）,根据CBN对预实例计算联合分布概率,准确预测了一对一营销优化中的客户行为．CBN学习算法包括连线和定向部分,复杂度为O（N^4）条件相关测试．在零售行业一对一营销实际应用表明,CBN学习算法较现有BN学习算法更快构建CBN,预测精度高于朴素Bayesina分类法．     

A new approach for generating efficient sample from market basket data

Classical data mining algorithms require expensive passes over the entire database to generate frequent items and hence to generate association rules. With the increase in the size of database, it is becoming very difficult to handle large amount of data for computation. One of the solutions to this problem is to generate sample from the database that acts as representative of the entire database for finding association rules in such a way that the distance of the sample from the complete database is minimal. Choosing correct sample that could represent data is not an easy task. Many algorithms have been proposed in the past. Some of them are computationally fast while others give better accuracy. In this paper, we present an algorithm for generating a sample from the database that can replace the entire database for generating association rules and is aimed at keeping a balance between accuracy and speed. The algorithm that is proposed takes into account the average number of small, medium and large 1-itemset in the database and average weight of the transactions to define threshold condition for the transactions. Set of transactions that satisfy the threshold condition is chosen as the representative for the entire database. The effectiveness of the proposed algorithm has been tested over several runs of database generated by IBM synthetic data generator. A vivid comparative performance evaluation of the proposed technique with the existing sampling techniques for comparing the accuracy and speed has also been carried out.     

A dynamic classifier ensemble selection approach for noise data

Dynamic classifier ensemble selection (DCES) plays a strategic role in the field of multiple classifier systems. The real data to be classified often include a large amount of noise, so it is important to study the noise-immunity ability of various DCES strategies. This paper introduces a group method of data handling (GMDH) to DCES, and proposes a novel dynamic classifier ensemble selection strategy GDES-AD. It considers both accuracy and diversity in the process of ensemble selection. We experimentally test GDES-AD and six other ensemble strategies over 30 UCI data sets in three cases: the data sets do not include artificial noise, include class noise, and include attribute noise. Statistical analysis results show that GDES-AD has stronger noise-immunity ability than other strategies. In addition, we find out that Random Subspace is more suitable for GDES-AD compared with Bagging. Further, the bias-variance decomposition experiments for the classification errors of various strategies show that the stronger noise-immunity ability of GDES-AD is mainly due to the fact that it can reduce the bias in classification error better.     

离散动态贝叶斯网络推理的编程计算算法

离散动态贝叶斯网络是对时间序列进行建模和推理的重要工具,具有广泛的建模应用价值,但是其推理算法还有待进一步完善。针对构离散动态贝叶斯网络的推理算法难以理解、编程计算难、推理速度慢的问题,给出了实现离散动态贝叶斯推理算法的数据结构,推导了进行计算机编程计算的推理算法和编程步骤,并通过实例进行了算理验证。     

A Bayesian approach to demand forecasting for new equipment programs

Demand forecasting is a fundamental component in a range of industrial problems (e.g., inventory management, equipment maintenance). Forecasts are crucial to accurately estimating spare or replacement part demand to determine inventory stock levels. Estimating demand becomes challenging when parts experience intermittent demand/failures versus demand at more regular intervals or high quantities. In this paper, we develop a demand forecasting approach that utilizes Bayes’ rule to improve the forecast accuracy of parts from new equipment programs where established demand patterns have not had sufficient time to develop. In these instances, the best information available tends to be “engineering estimates” based on like /similar parts or engineering projections. A case study is performed to validate the forecasting methodology. The validation compared the performance of the proposed Bayesian method and traditional forecasting methods for both forecast accuracy and overall inventory fill rate performance. The analysis showed that for specific situations the Bayesian-based forecasting approach more accurately predicts part demand, impacting part availability (fill rate) and inventory cost. This improved forecasting ability will enable managers to make better inventory investment decisions for new equipment programs.     

A new effective dynamic program for an investment optimization problem

After a series of publications of T.E. O’Neil et al. (e.g. in 2010), dynamic programming seems to be the most promising way to solve knapsack problems. Some techniques are known to make dynamic programming algorithms (DPA) faster. One of them is the graphical method that deals with piecewise linear Bellman functions. For some problems, it was previously shown that the graphical algorithm has a smaller running time in comparison with the classical DPA and also some other advantages. In this paper, an exact graphical algorithm (GrA) and a fully polynomial-time approximation scheme based on it are presented for an investment optimization problem having the best known running time. The algorithms are based on new Bellman functional equations and a new way of implementing the GrA.     

A Bayesian robust CP decomposition approach for missing traffic data imputation

The inevitable problem of missing data is ubiquitous in the real transportation system, which makes the data-driven intelligent transportation system suffer from incorrect response. We propose a Bayesian robust Candecomp/Parafac (CP) tensor decomposition (BRCP) approach to deal with missing data and outliers by integrating the general form of transportation system domain knowledge. Specifically, when the lower rank tensor captures the global information, the sparse tensor is added to capture the local information, which can robustly predict the distribution of missing items and under the fully Bayesian treatment, the effective variational reasoning can prevent the over fitting problem. Real and reliable traffic data sets are used to evaluate the performance of the model in two data missing scenarios, which the experimental results show that the proposed BRCP model achieves the best imputation accuracy and is better than the most advanced baseline (Bayesian Gaussian CP decomposition (BGCP), high accuracy low-rank tensor completion (HaLRTC) and SVD-combined tensor decomposition (STD)), even in the case of high missed detection rate, the model still has the best performance and robustness.

     

A neural network approach to a Bayesian statistical decisionproblem

Generalized mean-squared error (GMSE) objective functions are proposed that can be used in neural networks to yield a Bayes optimal solution to a statistical decision problem characterized by a generic loss function.