In this paper, we develop a novel semi-supervised learning algorithm called active hybrid deep belief networks (AHD), to address the semi-supervised sentiment classification problem with deep learning. First, we construct the previous several hidden layers using restricted Boltzmann machines (RBM), which can reduce the dimension and abstract the information of the reviews quickly. Second, we co...

Icons, or pictorial and graphical objects, are commonly used in human-computer interaction (HCI) fields as the mediator in order to communicate information to users. Yet there has been little studies focusing on a majority of the world’s population – semi-literate communities – in terms of the fundamental knowhow for designing icons for such population. In this study, two sets of icons belongin...

Semi-supervised learning reduces the cost of labeling the training data of a supervised learning algorithm through using unlabeled data together with labeled data to improve the performance. Co-Training is a popular semi-supervised learning algorithm, that requires multiple redundant and independent sets of features (views). In many real-world application domains, this requirement can not be sa...

To investigate the seismic behavior of end-plate connection semi-rigid space steel frames, three 1/4-scale specimens were tested under cyclic load. Finite element analysis which took initial geometric imperfections into consideration was also conducted, and the results conform to experimental results. The failure modes, hysteretic behavior, deformation capacity and energy-dissipation capacity o...

We report a case of a 35-year-old housewife who had a repetitive ventricular fibrillation following cardiac surgery. This arrhythmia was induced by sinus or supraventricular ectopic beats conducted to the ventricles and interrupting the preceding T or T U wave during incomplete A-V dissociation. The cycle length (Q-Q') and the Q-Q'/Q-T ratio of the beats preceding fibrillation were significantl...