Search

Cities have become the principal platform of economic growth and residents’ settlements. The importance of urban growth in residents’ quality of life is increasingly evident, because of the intimate interaction between urban characteristics and human activities. Cities shape urban residents’ welfare, bringing benefits and disadvantages as well. The complex way in which urban growth and residents’ activities interact makes it difficult to figure out which kinds of urban features are positively or instead negatively associated with residents’ quality of life. This puzzle makes it harder to clarify whether cities can provide enough net welfare to their residents. Thus, there is a major challenge to understand and evaluate how and to what extent urban growth shapes the welfare of urban residents. The link between cities and welfare has been hotly debated, yet we still have a limited grasp of how urban growth may impact residents’ welfare. Most studies concentrate on a certain dimension of urban welfare and a specific geographical scale. However, the interaction between urban growth and resident welfare may touch upon multiple dimensions of welfare coincidently, and it may take different forms across geographical scales. This thesis focuses on urban growth in China, against the background of a newly industrializing and developing country. Through the housing market dynamic, its central claim is to explore the tensions between the positive and negative effects of urban growth on resident welfare at multiple dimensions (economic, social, environmental, and policy) and scales (intra-city, inter-city, and countrywide) from a spatial inequality perspective.

Dialog systems have attracted attention as they are promising in many intelligent applications. Generating fluent and informative responses is of critical importance for dialog systems. Some recent studies introduce documents as extra knowledge to improve the performance of dialog generation. However, it is hard to understand the unstructured document and extract crucial information related to dialog history and current utterance. This leads to uninformative and inflexible responses in existing studies. To address this issue, we propose a generative model of a neural network with an attention mechanism for document-grounded multi-turn dialog. This model encodes the context of utterances that contains the given document, dialog history, and the last utterance into distributed representations via a triple-channel. Then, it introduces a hierarchical attention interaction between dialog contexts and previously generated utterances into the decoder for generating an appropriate response. We compare our model with various baselines on dataset CMU_DoG in terms of the evaluation criteria. The experimental results demonstrate the state-of-the-art performance of our model as compared to previous studies. Furthermore, the results of ablation experiments show the effectiveness of the hierarchical attention interaction and the triple channel for encoding. We also conduct human judgment to evaluate the informativeness of responses and the consistency of responses with dialog history.

Morphological parameters of coarse aggregates, including the angularity index (AI), flat and elongated ratio, flatness ratio, and sphericity, strongly affect the performance of asphalt concrete. A device must be developed to efficiently and accurately measure these parameters. Herein, a charge-coupled device camera and a linear profile scanner are combined to evaluate the parameters. The AI is calculated using three methods. The results show that the fitted-ellipse method provides superior results among the three AI measurement methods. Moreover, the use of a linear profile scanner improves the AI measurement accuracy. Therefore, the proposed combination has high measurement efficiency and is suitable for batch inspections.

This paper discusses the effects of industrial agglomeration on electrical energy consumption from a spatial perspective by highlighting multiple agglomeration types. It is argued that different types of industry agglomeration may exert differentiated effects on energy consumption and their possible spatial interactions. With city sector panel data of China's 285 prefecture-level cities for the period 2003–2013, we adopted the Spatial Durbin Model (SDM) with fixed effects to examine the spatial effects of industrial agglomeration types on energy consumption. The model found that all types of industrial agglomeration play significant and different roles in explaining overall local energy consumption. Second, geographical proximity, diversity, and related variety show the spatial spillover effects on surrounding areas. Third, the spatial energy effects of industrial agglomeration across regions are evidently heterogeneous. Diversity and one of its sub-forms-related variety maintains consumption-saving effects in regions with the initial or intermediate stage of an industry life cycle, while unrelated variety changes over different phases. The results suggest that policymakers scientifically discriminate regional features and guide the agglomerated types of industries in order to balance economic growth with energy conservation.

Wet markets, recently thrust into the limelight during the coronavirus pandemic, play a necessary role in daily life in Asia. Yet, the value of wet markets has not ever been sufficiently explored. While an increasing body of work has studied the implicit value of urban public amenities through the hedonic price model and traditional measurements of accessibility and density, the subjective perceptions of amenities have been neglected, specifically for consumption amenity. This is especially the case with wet markets: traditional approaches are unable to capture their comprehensive effects. By employing data on Beijing housing transactions in 2019 and online review scores of wet markets, we reduce this knowledge gap by exploring both the amenity and dis-amenity effects of wet markets and capturing the objective and subjective perspectives. Our results indicate a nonlinear relationship between wet market accessibility and urban housing prices. Considering the perceived quality of wet markets, this paper further indicates that housing prices near high-scoring wet markets appreciate while housing prices depreciate near low-scoring markets. Interestingly, the negative influence of low-scoring markets is statistically larger than the positive influence of high-scoring wet markets. Taking the housing price as a reflection of an owner's wealth and income level, we argue that high-income dwellers tend to pay more for perceived quality than for convenience. Our findings, therefore, offer new and refined insights for scholars and urban planning decision-makers.

A rich body of literature is stressing the crucial importance of migration and market evolution on the underutilization of housing in cities. Rural housing is unique given its less mature market structure. Drawing on an empirical case of Sunan, the work reported in this paper addresses underlying mechanisms of housing land underutilization in rural China. Employing a multi-level modeling approach, results suggest that the likelihood of underutilization relates to household and village features, such as family attributes, housing/parcel characteristics, type of villages, and geographical locations. Additionally, underutilization was also closely associated with regional contexts including local economic development levels and migration patterns. Theoretically, we conceptualized rural housing land underutilization as a land use transition with villagers' awareness of policy change possibilities in the context of both rapid urbanization and rural transformation with nested hybrid results. We argue that urban-rural differences do not induce underutilization. Rather, policy-induced overbuilding of new houses and insistence on retaining uninhabited older houses combined with the tendency for villagers’ to view investments as a mechanism to retain ties to their rural hometowns drove underutilizaiton.

The morphological parameters of coarse aggregates, including the angularity, flat and elongated (FE) and particle size, strongly influence the performance of asphalt concrete. At present, the measu...

The environmental externalities of industrial agglomeration have generated intense debate, yet few studies have considered their effects and inner influence mechanism based on varied agglomeration types. This study distinguished between industrial density and proximity (spatially) and industrial specialization and diversity, as well as related variety and unrelated variety (organizationally). Using the panel data from China's 285 cities from 2003 to 2013, we examined the different effects of multiple agglomeration types of industrial agglomeration on industrial SO2 emissions, and their inner influence mechanism from two aspects of industrial structure and technological progress. First, we found that various agglomeration types have different environmental externalities and that industrial density and proximity both have significant reduction effects on SO2 emissions; diversity and related variety effectively reduce SO2 discharges, but specialization and unrelated variety are linked to increased emissions. Second, technological progress and industrial structure are the critical channels of industrial agglomeration affecting pollution emissions while technological progress plays a greater role in the reducing emission effect of industrial agglomeration. Our findings reveal the importance of different agglomeration types and technological progress between industrial agglomeration and pollution emission.

Background: Diabetic nephropathy is a frequent microvascular complication of diabetes mellitus that causes end-stage renal disease most of the time. In China, Shenkang injection is one of widely used traditional Chinese medicine for treating chronic kidney disease, but its efficacy and safety have not yet been clarified. We will systematically review the current randomized controlled trial (RCT) evidence to summarize the efficacy and safety of Shenkang injection in treating diabetic nephropathy. Methods: We will search 7 literature databases including PubMed, EMBASE, Cochrane Library, Sinomed, Chinese National Knowledge Infrastructure, Wanfang, and VIP. Two trial registry platforms will also be searched. The time frame of the search will be from the inceptions of the databases to December 31, 2020. RCTs assessing Shenkang injection combined with basic treatments versus basic treatments alone for treating diabetic nephropathy will be included. The risk of bias within the individual RCTs will be evaluated using criteria proposed by the Cochrane Handbook 5.1.0. The primary outcomes to be investigated are glomerular filtration rate and serum creatinine; the secondary outcome will include 24-hour urine albumin excretion rate, blood urea nitrogen, fasting blood glucose, postprandial blood glucose, hemoglobin A1c, total cholesterol, triglyceride, response to treatment, and incidence of adverse events. The effect data of individual RCTs by performing random-effects model meta-analysis. Statistical heterogeneity will be measured by the Cochran Q test and I-squared statistics. Three subgroup analyses, set based on clinical experience, will be performed to explore the sources of heterogeneity. Sensitivity analyses excluding RCTs with high risk of bias and using fixed effect model will be done to test the robustness of the meta-analytic results. Publication bias across included RCTs will be evaluated by funnel plots and Egger test. Results: This study will provide systematic review on the efficacy and safety of Shenkang Injection as adjuvant therapy in patients with diabetic nephropathy by rigorous quality assessment and reasonable data synthesis. The results will be submitted to a peer-reviewed journal for publication. Conclusion: This systematic review will provide the best evidence currently on Shenkang Injection as adjuvant therapy in patients with diabetic nephropathy. INPLASY registration number: INPLASY2020110014.

In recent years, image style transfer has been greatly improved by using deep learning technology. However, when directly applied to clothing style transfer, the current methods cannot allow the users to self-control the local transfer position of an image, such as separating specific T-shirt or trousers from a figure, and cannot achieve the perfect preservation of clothing shape. Therefore, this paper proposes an interactive image localized style transfer method especially for clothes. We introduce additional image called outline image, which is extracted from content image by interactive algorithm. The interaction consists simply of dragging a rectangle around the desired clothing. Then, we introduce an outline loss function based on distance transform of the outline image, which can achieve the perfect preservation of clothing shape. In order to smooth and denoise the boundary region, total variation regularization is employed. The proposed method constrains that the new style is generated only in the desired clothing part rather than the whole image including background. Therefore, in our new generated images, the original clothing shape can be reserved perfectly. Experiment results show impressive generated clothing images and demonstrate that this is a good approach to design clothes.

We study theoretically the localization and coupling properties of defect modes in octagonal quasi-crystals made of conjugated polymers for the light modulation of organic light-emitting devices (OLED) and organic lasers. Due to the rotational symmetry of quasicrystals and the low refractive-index of organic semiconductors, it is observed that localized modes only occur in the complex defect-cavity structures. Unlike conventional photonic crystals with high-index, some unusual characteristics of defect modes coupling in quasicrystals under the low-index contrast are revealed and analyzed, such as the loss of defect modes, composite resonant cavity, and phase-opposition of modal tails. These results will allow us to better utilize the light control in organic photonic crystal microcavity lasers, especially in the localization region, and further to apply in OLED and VCSEL (Vertical Cavity Surface Emitting Laser) fields.

Along with the development of social media on the internet, dialogue systems are becoming more and more intelligent to meet users’ needs for communication, emotion, and social intercourse. Previous studies usually use sequence-to-sequence learning with recurrent neural networks for response generation. However, recurrent-based learning models heavily suffer from the problem of long-distance dependencies in sequences. Moreover, some models neglect crucial information in the dialogue contexts, which leads to uninformative and inflexible responses. To address these issues, we present a bichannel transformer with context encoding (BCTCE) for document-driven conversation. This conversational generator consists of a context encoder, an utterance encoder, and a decoder with attention mechanism. The encoders aim to learn the distributed representation of input texts. The multihop attention mechanism is used in BCTCE to capture the interaction between documents and dialogues. We evaluate the proposed BCTCE by both automatic evaluation and human judgment. The experimental results on the dataset CMU_DoG indicate that the proposed model yields significant improvements over the state-of-the-art baselines on most of the evaluation metrics, and the generated responses of BCTCE are more informative and more relevant to dialogues than baselines.

Magnetic hyperthermia using ferrite nanoparticles has emerged as a eutherapeutic approach for cancer treatment. In this study, we had constructed folate conjugated poly(ethyleneimine) (PEI) modified magnetic nanoparticles (FA-PEI-MNPs) that generate heat in an alternating magnetic field (AMF). Transmission electron microscope (TEM) showed that FA-PEI-MNPs were nearly spherical in shape. The mean diameter of these nanoparticles determined by DLS was 30.7 +/- 7.7 nm. The cytotoxicity assay showed that FA-PEI-MNPs exhibited no significant cytotoxicity against L929 cells. FA-PEI-MNPs have good hyperthermia effect on KB cells in vitro by magnetic field induced hyperthermia, especially at 45 degrees C and 47 degrees C. In in vivo hyperthermia treatment using FA-PEI-MNPs in a subcutaneous tumor xenograft model of KB cells tumor, the tumor temperature increased to 48 degrees C whereas the body temperature remained at around 37.5 degrees C. Thermotherapy with FA-PEI-MNPs led up to prolongation of survival over magnetic field controls, 66.7% (4/6) of subcutaneous tumors regressed completely. Histological examinations after treatment revealed large necrotic areas close to particle deposits. These results suggest that hyperthermia using FA-PEI-MNPs is an effective and specific therapy for tumor. This therapy may provide an alternative way to treat recurrent cancer refractory to other modalities.

Hot mix asphalt (HMA) mixture is a major building material in paving engineering. To decrease energy consumption in HMA, a prediction model of energy consumption was investigated systematically in this paper, employing kernel principal component analysis (KPCA) and a support vector machine (SVM). The purpose of the work is to optimize production and structure parameters of an HMA plant. A prediction model of energy consumption cannot only be used for understanding but can also be used to develop new asphalt mixing plants and to achieve optimization. The main relationships between energy consumption and production parameters are studied in some field tests. To build a multi-parameter model of energy consumption, eigenvectors of many factors are optimized employing KPCA. The three kernel principal components of higher contribution rates are chosen. A prediction model of energy consumption is built using KPCA and SVM. Energy consumption of aggregate drying is predicted using the constructed model. The prediction model is optimized by particle swarm optimization (PSO). Influence coefficients of energy consumption are obtained by SVM and piecewise least-squares regression (PLSR) and are found to be consistent with test results. There is about a 5% error in energy consumption between the model prediction and the test. The prediction error can meet engineering requirements in the hot mix asphalt mixing plant.