Search

Multiple kernel k‐means clustering (MKKC) can efficiently incorporate multiple base kernels to generate an optimal kernel. Many existing MKKC methods all need two‐step operation: learning clustering indicator matrix and performing clustering on it. However, the optimal clustering results of two steps are not equivalent to those of original problem. To address this issue, in this paper we propose a novel method named one‐step multiple kernel k‐means clustering based on block diagonal representation (OS‐MKKC‐BD). By imposing a block diagonal constraint on the product of indicator matrix and its transpose, this method can encourage the indicator matrix to be block diagonal. Then the indicator matrix can produce explicit clustering indicator, so as to implement one‐step clustering, which avoids the disadvantage of two‐step operation. Furthermore, a simple kernel weighting strategy is used to obtain an optimal kernel, which boosts the quality of optimal kernel. In addition, a three‐step iterative algorithm is designed to solve the corresponding optimization problem, where the Riemann conjugate gradient iterative method is used to solve the optimization problem of the indicator matrix. Finally, by extensive experiments on eleven real data sets and comparison of clustering results with 10 MKC methods, it is concluded that OS‐MKKC‐BD is effective. [ABSTRACT FROM AUTHOR]

Among the methods of multiple kernel clustering (MKC), some adopt a neighborhood kernel as the optimal kernel, and some use local base kernels to generate an optimal kernel. However, these two methods are not synthetically combined together to leverage their advantages, which affects the quality of the optimal kernel. Furthermore, most existing MKC methods require a two-step strategy to cluster, i.e., first learn an indicator matrix, then executive clustering. This does not guarantee the optimality of the final results. To overcome the above drawbacks, a one-step clustering with adaptively local kernels and a neighborhood kernel (OSC-ALK-ONK) is proposed in this paper, where the two methods are combined together to produce an optimal kernel. In particular, the neighborhood kernel improves the expression capability of the optimal kernel and enlarges its search range, and local base kernels avoid the redundancy of base kernels and promote their variety. Accordingly, the quality of the optimal kernel is enhanced. Further, a soft block diagonal (BD) regularizer is utilized to encourage the indicator matrix to be BD. It is helpful to obtain explicit clustering results directly and achieve one-step clustering, then overcome the disadvantage of the two-step strategy. In addition, extensive experiments on eight data sets and comparisons with six clustering methods show that OSC-ALK-ONK is effective. [ABSTRACT FROM AUTHOR]

In this paper, the heat transfer characteristics of a 2D gas–solid spout fluidized bed with a hot gas jet are investigated using computational fluid dynamics-discrete element method. The initial temperature of the background gas and particles in the spouted bed was set to 300 K. The particle temperature distribution after injection of 500 K gas from the bottom, center of the bed, is presented. The simulation results indicate well heat transfer behavior in the bed. Then, statistical analysis is conducted to investigate the influence of inlet gas velocity and particle thermal conductivity on the heat transfer at particle scale in detail. The results indicate that the particle mean temperature and convective heat transfer coefficient (HTC) linearly increase with the increase in inlet gas velocity, while the conductive HTC and the uniformity of particle temperature distribution are dominated by the particle thermal conductivity. The conductive and convective heat transfer play different roles in the spout fluidized bed. These results should be useful for the further research in such flow pattern and the optimization of operating such spouted fluidized beds. [ABSTRACT FROM AUTHOR]

This study aimed to explore the effects of dominant and compliant personalities, on both flow experience and the external characteristics of flow experience. A total of 48 participants were recruited to play an online game and subsequently asked to recall the songs they had heard while they were playing the game. Eye blink rate was recorded. The results demonstrated that (1) the participant was immersed in the game more if he/she was relatively dominant or noncompliant; (2) the perceptions about the external environment declined remarkably while being in a flow state; and (3) eye blink rates decreased only when the flow happened at the beginning of the game, rather than throughout the whole process. The results suggested that gamers who tend to be dominant or noncompliant were more likely to experience flow. Eye blink rate and perceptions of the external environment could be objective indicators of flow experience. [ABSTRACT FROM AUTHOR]

Objective: The aim of the present study is to investigate Chinese handwriting on mobile touch devices, considering the effects of three characteristics of the human finger (type, length, and width) and three characteristics of Chinese characters (direction of the first stroke, number of strokes, and structure).Background: Due to the popularity of touch devices in recent years, finger input for Chinese characters has attracted more attention from both industry and academia. However, previous studies have no systematical consideration on the effects of human finger and Chinese characters on Chinese handwriting performance.Method: An experiment was reported in this article to illustrate the effects of the human finger and Chinese characters on the Chinese handwriting performance (i.e., input time, accuracy, number of protruding strokes, mental workload, satisfaction, and physical fatigue).Results: The experiment results indicated that all six factors have significant effects on Chinese handwriting performance, especially on the input time, accuracy, and number of protruding strokes.Conclusion: Finger type, finger length, finger width, direction of the first stroke, number of strokes, and character structures are significantly influencing Chinese handwriting performance. These factors should be taken into more consideration in future research and the practical design for Chinese handwriting systems. [ABSTRACT FROM AUTHOR]