Search

Unsupervised domain adaptation aims to transfer knowledge from labeled source domain to unlabeled target domain. Recently, multisource domain adaptation (MDA) has begun to attract attention. Its performance should go beyond simply mixing all source domains together for knowledge transfer. In this article, we propose a novel prototype-based method for MDA. Specifically, for solving the problem that the target domain has no label, we use the prototype to transfer the semantic category information from source domains to target domain. First, a feature extraction network is applied to both source and target domains to obtain the extracted features from which the domain-invariant features and domain-specific features will be disentangled. Then, based on these two kinds of features, the named inherent class prototypes and domain prototypes are estimated, respectively. Then a prototype mapping to the extracted feature space is learned in the feature reconstruction process. Thus, the class prototypes for all source and target domains can be constructed in the extracted feature space based on the previous domain prototypes and inherent class prototypes. By forcing the extracted features are close to the corresponding class prototypes for all domains, the feature extraction network is progressively adjusted. In the end, the inherent class prototypes are used as a classifier in the target domain. Our contribution is that through the inherent class prototypes and domain prototypes, the semantic category information from source domains is transformed into the target domain by constructing the corresponding class prototypes. In our method, all source and target domains are aligned twice at the feature level for better domain-invariant features and more closer features to the class prototypes, respectively. Several experiments on public data sets also prove the effectiveness of our method. [ABSTRACT FROM AUTHOR]

The article focuses on the user-centered design (UCD) practice. User-Centered Design (UCD) is a multidisciplinary design approach based on the active involvement of users to improve the understanding of user and task requirements, and the iteration of design and evaluation. It is widely considered the key to product usefulness and usability, an effective approach to overcoming the limitations of traditional system-centered design. In the past, many of the UCD methods in the literature were found ineffective or impractical for a variety of reasons. However, the growing popularity of e-commerce has greatly bolstered the appeal of usability and UCD, as users can take their business elsewhere with just one mouse click. In the past, the lack of widespread UCD implementation was thought to be partly due to two common misconceptions: usability could not be measured, and development work did not have usability goals. A series of hierarchical regression analyses was conducted to examine organizational properties and characteristics of UCD processes as potential factors affecting UCD effectiveness.

Abstract Latent variable hidden Markov models (LVHMMs) are important statistical methods in exploring the possible heterogeneity of data and explaining the pattern of subjects moving from one group to another over time. Classic subject- and/or time-homogeneous assumptions on transition matrices in transition model as well as the emission distribution in the observed process may be inappropriate to interpret heterogeneity at the subject level. For this end, a general extension of LVHMM is proposed to address the heterogeneity of multivariate longitudinal data both at the subject level and the occasion level. The main modeling strategy is that the observed time sequences are first grouped into different clusters, and then within each cluster the observed sequences are formulated via latent variable hidden Markov model. The local heterogeneity at the occasion level is characterized by the distribution related to the latent states, while the global heterogeneity at the subject level is identified with the finite mixture model. Compared to the existing methods, an appeal underlying the proposal is its capacity of accommodating non-homogeneous patterns of state sequences and emission distributions across the subjects simultaneously. As a result, the proposal provides a comprehensive framework for exploring various kinds of relevance among the multivariate longitudinal data. Within the Bayesian paradigm, Markov Chains Monte Carlo (MCMC) method is used to implement posterior analysis. Gibbs sampler is used to draw observations from the related full conditionals and posterior inferences are carried out based on these simulated observations. Empirical results including simulation studies and a real example are used to illustrate the proposed methodology. Highlights • This paper proposes a general extension of latent variable hidden Markov model to address the heterogeneity of multivariate longitudinal data both at the subject level and the occasion level. • The local heterogeneity at the occasion level is characterized by the distribution related to the latent states, while the global heterogeneity at the subject level is identified with the finite mixture model. • A Bayesian procedure coupled with Gibbs sampler is developed to carry out posterior inference. • Empirical results including simulation studies and a real example are presented to illustrate the proposed methodology. [ABSTRACT FROM AUTHOR]

Sparsity in features presents a big technical challenge to existing clustering methods for categorical data. Hierarchical Bayesian Bernoulli mixture model (HBBMM) incorporates constrained empirical Bayes priors for model parameters, so the resulting Expectation Maximization (EM) algorithm of estimator searching is confined in a proper region. The EM algorithm enables to obtain the maximum a posterior (MAP) estimation, in which cluster labels are simultaneously assigned. Three criteria are proposed to identify defining features of individual clusters, leading to understanding of the underlying data structures. Information based model selection criterion is applied to determine the number of clusters. Estimation consistency and performance of model selection criteria are investigated. Two real-world sparse categorical datasets are analyzed with the proposed method. [ABSTRACT FROM AUTHOR]

The article discusses three successful digital technology start-up companies in China, namely coffee company Saturnbird Coffee, sugar-free beverage company Yuan Qi Sen Lin (YQSL), and cosmetics company Perfect Diary. It particularly focuses upon the firms' product innovations in relation to consumer desires, use of social media, and the integration of the Chinese economy into the worldwide supply chain and global value chain.

The communication accessibility of the lunar rover from the Earth to the Moon can provide a basis for the choice of ideal landing sites. In this paper, a mathematical model is established to study the communication accessibility affected by topography, by combining lunar digital elevation models (DEM) with known parameters related to lunar and earth' s orbit. The laser altimeter data obtained by 'Chang' E-1' can provide high accuracy DEM and can be as the data basis for this study. The research region is Sinus Iridum region (the preferred landing area of Chinese lunar exploration) and the lunar polar regions. The calculation period is from Oct. 1, 2013 to Oct. 30, 2013. The results show that (1) in the Sinus Iridum region, the communication condition between tracking stations (Beijing, Kunming, Shanghai and Urumchi) and the lunar rover is expedite, which is consistent with the region's flat terrain; (2) in the polar regions, the influence of topography on the communication condition is great because of the topography complexity. [ABSTRACT FROM AUTHOR]

Abstract: This paper concerns stability and stabilization issues for switched neutral systems and presents new classes of piecewise Lyapunov functionals and multiple Lyapunov functionals, based on which, two new switching rules are introduced to stabilize the neutral systems. One switching rule is designed from the solution of the so-called Lyapunov–Metzler linear matrix inequalities. The other is based on the determination of average dwell time computed from a new class of linear matrix inequalities (LMIs). And then, state-feedback control is derived for the switched neutral control system mainly based on the state switching rules. Finally, three examples are given to demonstrate the effectiveness of the proposed method. [Copyright &y& Elsevier]

Abstract: In this paper, by using the theories and methods of ecology and ordinary differential equation, an ecological model with impulsive control strategy and distributed time delay is established. By using the theories of impulsive equation, small amplitude perturbation skills and comparison technique, we get the condition which guarantees the global asymptotical stability of the lowest level prey and top predator eradication periodic solution. Further, influences of the impulsive perturbation and the parameter a on the inherent oscillation are studied numerically, which shows rich dynamics, such as period-doubling bifurcation, period-halving bifurcation, chaotic band, narrow or wide periodic window, chaotic crises, etc. Moreover, computation of the largest Lyapunov exponent demonstrates the chaotic dynamic behavior of the model. Meanwhile. we investigate the qualitative nature of strange attractor by using Fourier spectra. All these results may be useful for study of the dynamic complexity of ecosystems. [Copyright &y& Elsevier]

Purpose - The purpose of this paper is to present a symmetrical method to extract smooth signal from linear mixtures in the frequency domain; with experimentations, the method is thereafter evaluated. Design/methodology/approach - The Second-order Frequency Identification (SOFI) algorithm has been presented to retrieve baseband signals which have inactive bands and different bandwidths in the frequency domain. However, the SOFI method will bring about accumulative error, therefore an improved method is proposed in this paper by using symmetric extraction mode. In contrast to the SOFI algorithm, the ISOFI method can extract simultaneously the signal with the highest degree of smoothness and the signal with the lowest degree of smoothness. This means that the signals are not extracted one by one; instead, they are extracted in parallel. Experimental results in both noise-free and noisy scenarios verified that the proposed method has a significant improvement compared with the SOFI algorithm. Findings - An improved SOFI (ISOFI) method is proposed to reduce the accumulated error encountered in the SOFI algorithm. In the proposed method, the symmetric mode is utilized to extract in parallel the signals with different smooth degrees. Experimental results demonstrated that the ISOFI has a higher accuracy and lower accumulated error compared to the original algorithm. Originality/value - The paper demonstrates use of the symmetric extraction mode to overcome the disadvantage of accumulated errors existing in the SOFI algorithm. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, the issue of pth moment exponential synchronization of a class of stochastic neural networks with mixed delays is investigated. By establishing two new integro-differential inequalities, some new sufficient conditions ensuring pth moment exponential synchronization are obtained. Compared with previous method, our method does not resort to any Lyapunov function. The results obtained in this paper generalize some earlier works reported in the literature. Some strict constraints of time delays and kernel functions are removed. Two numerical examples are presented to illustrate the validity of the main results. [Copyright &y& Elsevier]

Abstract: In this paper, a tabu search based clustering approach called TS-Clustering is proposed to deal with the minimum sum-of-squares clustering problem. In the TS-Clustering algorithm, five improvement operations and three neighborhood modes are given. The improvement operation is used to enhance the clustering solution obtained in the process of iterations, and the neighborhood mode is used to create the neighborhood of tabu search. The superiority of the proposed method over some known clustering techniques is demonstrated for artificial and real life data sets. [Copyright &y& Elsevier]