A mixed methods approach to the assessor's targeting behavior during online peer assessment: effects of anonymity and underlying reasons

This study examined the effects of identity revelation and concealment on the number of times students' work was assessed in an online peer assessment context. It also examined the underlying reasons guiding the assessor's targeting behavior. Two fifth-grade classes participated. The one-group pretest–posttest experimental research design coupled with qualitative research method was adopted. The results from social network analysis (SNA) showed a high redundancy rate for both high- and low-targeted assessees in identity-revealed and identity-concealed conditions for both classes. Moreover, the density metrics indicated that all sociograms were similarly sparse. The Wilcoxon tests further confirmed that there were no significant differences in the assessees' ranking in the two conditions for both classes. Finally, analysis of students' self-reflective data revealed that the features/content of the work to be assessed, rather than who the question-author was, was the determining factor when it came to deciding the target for online peer assessment.     

A probabilistic music recommender considering user opinions and audio features

A recommender system has an obvious appeal in an environment where the amount of on-line information vastly outstrips any individual’s capability to survey. Music recommendation is considered a popular application area. In order to make personalized recommendations, many collaborative music recommender systems (CMRS) focus on capturing precise similarities among users or items based on user historical ratings. Despite the valuable information from audio features of music itself, however, few studies have investigated how to utilize information extracted directly from music for personalized recommendation in CMRS. In this paper, we describe a CMRS based on our proposed item-based probabilistic model, where items are classified into groups and predictions are made for users considering the Gaussian distribution of user ratings. In addition, this model has been extended for improved recommendation performance by utilizing audio features that help alleviate three well-known problems associated with data sparseness in collaborative recommender systems: user bias, non-association, and cold start problems in capturing accurate similarities among items. Experimental results based on two real-world data sets lead us to believe that content information is crucial in achieving better personalized recommendation beyond user ratings. We further show how primitive audio features can be combined into aggregate features for the proposed CRMS and analyze their influences on recommendation performance. Although this model was developed originally for music collaborative recommendation based on audio features, our experiment with the movie data set demonstrates that it can be applied to other domains.     

Less conservative stabilization conditions for Markovian jump systems with partly unknown transition probabilities

This paper is devoted to the robust stabilization problem for a class of MJLSs with mode transition probabilities taken to be partly known, unknown, and unknown but with known bounds. To meet the needs of less conservative stabilization conditions, this paper proposes a valuable approach that (1) can express the conditions in terms of matrix inequalities with homogeneous polynomial dependence on partly unknown transition probabilities and (2) can include all possible slack variables related to transition probability constraints in the relaxation process coupled with deriving a finite set of linear matrix inequalities (LMIs). Finally, two numerical examples are reported to illustrate the effectiveness of the derived stabilization conditions.     

Characterization of anomalous movements of spherical living cells on a silicon dioxide glassy substrate

The random walk of spherical living cells on a silicon dioxide glassy substrate was studied experimentally and numerically. This random walk trajectory exhibited erratic dancing, which seemingly obeyed anomalous diffusion (i.e., Lévy-like walk) rather than normal diffusion. Moreover, the angular distribution (−π to π) of the cells'' trajectory followed a “U-shaped pattern” in comparison to the uniform distribution seen in the movements of negatively charged polystyrene microspheres. These effects could be attributable to the homeostasis-driven structural resilient character of cells and physical interactions derived from temporarily retained nonspecific binding due to weak forces between the cells and substrates. Our results provide new insights into the stochastic behavior of mesoscopic biological particles with respect to structural properties and physical interactions.     

The relationship between university student learning outcomes and participation in social network services,social acceptance and attitude towards school life

Educators try to use social network services (SNSs) in their teaching because they are regarded as beneficial to student learning. However, little research to date has empirically investigated whether the use of an SNS increases university student learning outcomes. A total of 730 university students in the capital area of South Korea participated in the present study. Based on social learning theory, we developed a general structural model that included SNS participation (SP) as an exogenous variable and a range of endogenous variables: cognitive domain learning outcomes, affective domain learning outcomes, social domain learning outcomes, social acceptance (SA) and attitude towards university life (AT). The study results confirmed that the model adequately explained causal relationships between student learning outcomes and SA, AT and SP. SA was identified as a key variable because of its direct effect on learning outcomes. However, SP was considered more important because of its large indirect effects on all learning outcomes through both SA and AT.     

Decentralized low-complexity tracking of uncertain interconnected high-order nonlinear systems with unknown high powers

A global decentralized low-complexity tracker design methodology is proposed for uncertain interconnected high-order nonlinear systems with unknown high powers. It is assumed that interconnected nonlinearities are bounded by completely unknown nonlinearities, rather than, a linear combination of high-ordered state variables. Compared with the existing decentralized results for interconnected nonlinear systems with known high powers, the decentralized robust controller, which achieves the pre-designable transient and steady-state tracking performance for each subsystem, is designed by employing nonlinear error surfaces with time-varying performance functions, regardless of unknown nonlinear interactions and high powers related to virtual and actual control variables. The proposed decentralized continuous robust low-complexity tracker is realized without the use of any adaptive or function approximation techniques for estimating unknown parameters and nonlinearities. The stability and preassigned tracking performance of the resulting decentralized low-complexity control system are thoroughly analyzed in the Lyapunov sense. Finally, simulation results on coupled underactuated mechanical systems are provided to show the effectiveness of the proposed theoretical result.