This paper introduces a novel rotation-based framework for arbitrary-oriented text detection in natural scene images. We present the Rotation Region Proposal Networks (RRPN), which are designed to generate inclined proposals with orientation angle information. The information is then adapted bounding box regression make more accurately fit into region terms of orientation. Region-of-Interest (RRoI) pooling layer proposed project feature map classifier. whole built upon region-proposal-based...

In China, dengue remains an important public health issue with expanded areas and increased incidence recently. Accurate timely forecasts of in China are still lacking. We aimed to use the state-of-the-art machine learning algorithms develop accurate predictive model dengue.Weekly cases, Baidu search queries climate factors (mean temperature, relative humidity rainfall) during 2011-2014 Guangdong were gathered. A index was constructed for developing models combination factors. The observed...

Crack is one of the most common road distresses which may pose safety hazards. Generally, crack detection performed by either certified inspectors or structural engineers. This task is, however, time-consuming, subjective and labor-intensive. In this paper, a novel algorithm based on deep learning adaptive image segmentation proposed. Firstly, convolutional neural network trained to determine whether an contains cracks not. The images containing are then smoothed using bilateral filtering,...

This article introduces collaborative learning models based on sociocultural theories and discusses their potential for developing more effective opportunities in information literacy teaching. As described the research of Vygotsky other theoreticians, are learner-centered provide insight into approaches to student learning. These take account social cultural aspects acquiring knowledge. Collaborative learning, as outlined literature review, is an means increasing achievement cognitive...

Autoregressive integrated moving average (ARIMA) model has been successfully applied as a popular linear for economic time series forecasting. In addition, during the recent years, artificial neural networks (ANNs) have used to capture complex relationships with variety of patterns they serve powerful and flexible computational tool. However, most these studies characterized by mixed results in terms effectiveness ANN s compared ARIMA model. this paper, we propose hybrid model, which is...

Despite the significant financial losses associated with data breaches, little is known about (in)effectiveness of tools that firms have to protect their value following a breach. Drawing on cognitive dissonance theory and research cue diagnosticity crisis management, this study examines relative efficacy firm reputation range post-breach response strategies. The results indicate an important asset in protecting value. However, only certain strategies are found mitigate negative impact...

Up to now, little effort has been made exploit large-area high-throughput patterning by block copolymer (BCP) lithography generate nanostructured substrates with periods well below 100 nm for surface-enhanced Raman scattering (SERS). We show that simple BCP-templated galvanic displacement reactions yield dense arrays of mushroom-shaped gold nanopillars a period 50 nm. The nanoporous BCP films used as templates were obtained swelling-induced reconstruction reverse micelle monolayers deposited...

This paper studies the problem of finding best rank-1 approximations for both symmetric and nonsymmetric tensors. For tensors, this is equivalent to optimizing homogeneous polynomials over unit spheres; multiquadratic forms multispheres. We propose semidefinite relaxations, based on sum squares representations, solve these polynomial optimization problems. Their special properties structures are studied. In applications, resulting programs often large scale. The recent Newton-CG augmented...

The poor predictive power of existing preclinical models has spurred efforts to develop human-relevant for accurate assessment drug safety. In this work, we developed a multi-organoids-on-a-chip system derived from human induced pluripotent stem cells (hiPSCs), which allows the cardiac safety an antidepressant drug, following liver metabolism in vitro. This liver-heart organoids-on-chip device contains compartmentalized chambers separated by porous membrane, permits co-culture 3D organoids...

To better understand users' technology utilisation behaviour, a construct named dependence is suggested; as well model proposed and empirically tested. Based on the literature marketing, information systems' utilisation, psychology, comprehensive set of constructs hypotheses are compiled with methodology for testing them in this paper. A questionnaire was designed data were collected from 255 users smartphones Korea, mainly consisting students academics. Structural equation modelling then...

Purpose Mobile health service (MHS) can provide users with convenient services and information to reduce their medical costs from aging or other issues. Previous studies confirm the underdevelopment of Chinese MHS. The purpose this paper is analyze factors that affect intention use Design/methodology/approach This develops a research model integrates personal differences theory planned behavior (TPB) protection motivation (PMT). empirically tested using data 494 valid questionnaires....

Blockchain technology is introduced into the dual-channel supply chain system of online direct marketing and offline traditional retailing to solve problem opaque product sources information asymmetry while also considering consumers' privacy concerns increase their willingness buy improve enterprises' profitability. Based on introduction blockchain technology, paper considers concerns, uses manufacturer-dominated Stackelberg game model equilibrium, compares analyzes optimal pricing...

The relationship of mentoring and network resources, the two components social capital, with career success was investigated in a sample 113 Chinese white-collar workers. findings suggested that prevalence workplace is substantially higher than Anglo-Saxon workplace, employees do not distinguish their ties or Guanxi into expressive instrumental. These results were line presumption an integral part culture view society cannot exist purely instrumental form. In contrast to hypotheses, however,...

Drawing on the example of a recent study (Wang, 2010), this paper discusses application sociocultural approach to information literacy research and curricular design. First, it describes foundation in theories, particular Vygotsky's theory. Then presents key theoretical principles arising from how enabled establishment collaborative partnerships between professionals academic teaching support staff community practice for integration.

Abstract Virtual enterprises (VEs) have become one of main types organisations in the manufacturing sector through which consortium companies organise their activities. To be competitive, a VE relies on complementary core competences among members resource sharing and agile capacity. Manufacturing grid (M-Grid) is platform production resources can shared. In this article, an M-Grid-based operation (MGVEOP) presented as it enables geographically distributed enterprises. The performance...

Abstract Organizing single cells along a developmental trajectory has emerged as powerful tool for understanding how gene regulation governs cell fate decisions. However, learning the structure of complex single-cell trajectories with two or more branches remains challenging computational problem. We present Monocle 2, which uses reversed graph embedding to reconstruct in fully unsupervised manner. 2 learns an explicit principal describe data, greatly improving robustness and accuracy its...

With the development of globalization, English continues to be widely used, and teaching is given considerable importance. In recent years, Chinese universities have introduced more native speakers as language teachers. However, ingrained native-speakerism in field (ELT) claims that English-speaking teachers (NESTs) are ideal teachers, while non-native (NNESTs) often marginalized. Recognizing issue ELT, this paper investigates university students' teachers' attitudes towards NESTs NNESTs...

Prior literature argues that information technology (IT) capability, an organization's ability to effectively use IT-based resources in combination with other organizational resources, can create unique and sustainable competitive advantages thus intangible assets for a company. However, current accounting rules do not allow the capitalization of IT-enabled assets. We hypothesize are value-relevant provide incremental explanatory power firm valuation beyond traditional information. IT...