Cross-Attention Fusion Based Spatial-Temporal Multi-Graph Convolutional Network for Traffic Flow Prediction

Accurate traffic flow prediction is essential to building a smart transportation city. Existing research mainly uses a given single-graph structure as a model, only considers local and static spatial dependencies, and ignores the impact of dynamic spatio-temporal data diversity. To fully capture the characteristics of spatio-temporal data diversity, this paper proposes a cross-Attention Fusion Based Spatial-Temporal Multi-Graph Convolutional Network (CAFMGCN) model for traffic flow prediction. First, introduce GCN to model the historical traffic data’s three-time attributes (current, daily, and weekly) to extract time features. Second, consider the relationship between distance and traffic flow, constructing adjacency, connectivity, and regional similarity graphs to capture dynamic spatial topology information. To make full use of global information, a cross-attention mechanism is introduced to fuse temporal and spatial features separately to reduce prediction errors. Finally, the CAFMGCN model is evaluated, and the experimental results show that the prediction of this model is more accurate and effective than the baseline of other models.

Regional Atmospheric Light Optimisation Algorithm for Heterogeneous Image Dehazing

Under foggy and other severe weather conditions, image acquisition equipment is not effective. It often produces an image with low contrast and low scene brightness, which is difficult to use in other image-based applications. The dark channel prior dehazing algorithm will cause the brightness of the image to decrease and sometimes introduce halos in the sky area. To solve this problem, we proposed a region similarity optimisation algorithm based on a dark channel prior. First, a vector comprising RGB layer dark channel value was obtained as the original atmospheric ambient light, and then, the proposed regional similarity linear function was used to adjust the atmospheric ambient light matrix. Next, the transmittance of different colour channels was derived and the multichannel soft matting algorithm was employed to produce more effective transmittance. Finally, the atmospheric ambient light and transmittance were substituted into the atmospheric scattering model to calculate clean images. Experimental results show that the proposed algorithm outperformed the existing mainstream dehazing algorithms in terms of both visual judgement and quality analysis with nonhomogeneous haze datasets. The algorithm not only improves the image details but also improves the brightness and saturation of the dehazing result; therefore, the proposed algorithm is effective in the restoration of the hazy image.

Diversity and distribution of the superfamily Grylloidea (Orthoptera: Ensifera: Gryllidea) in the Nearctic region

Based on the geographic distribution database of the Orthoptera Species File, the diversity and distribution of the superfamily Grylloidea in the Nearctic region was studied using the statistics and Sorensen dissimilarity coefficient. A total of 164 species or subspecies belonging to 4 families, 9 subfamilies and 27 genera were recorded from this region; among which Gryllidae (93, 56.70%), followed by Trigonidiidae (44, 26.83%), Mogoplistidae (25, 15.24%), and Phalangopsidae (2, 1.22%). The diversity exhibits an asymmetric distribution pattern, with the southeastern coastal plain, the Interior Plateau and Piedmont of the United States was the most abundant. At the same time, the regional similarity of species distribution was analyzed, and the Nearctic was divided into four subregions: Boreal & Arctic zone of North America, Eastern temperate North America, Northeast temperate North America, and Southern North America & western temperate North America.

The Effect of Functional Fiber on Microbiota Composition in Different Intestinal Segments of Obese Mice

The gastrointestinal tract is a heterogeneous ecosystem with distinct, stratified environments, which leads to different microbial composition in different intestinal segments. The regional heterogeneity of intestinal microbiota complicates the relationship between diet and microbiota. Few studies have focused on the effects of different diets on microbiota in different intestinal segments. This study aimed to investigate the effects of functional fiber on the microbial composition in multiple intestinal segments from a high-fat diet compared with a normal chow diet. We found that the response of microbiota from different intestinal segments to diet was related to the intestinal physiologic function and the physicochemical properties of dietary nutrients. A high-fat diet drove changes in the microbial composition in the hindgut, possibly by affecting the digestive environment of the foregut, and increased the regional heterogeneity of the whole intestinal microbiota. The supplementation of functional fiber promoted the microbial transfer and colonization from the anterior to the posterior intestinal segments, and increased the regional similarity of intestinal microbiota accordingly, particularly within the hindgut. The gut fermentation of the functional fiber, which mainly occurred in the hindgut, resulted in a significant change in the microbial composition and metabolism in the cecum and colon, with richer carbohydrate metabolism-related bacteria, including Mucispirillum, Prevotella, Anaerostipes, Oscillospira, Ruminococcus, Bacteroides, Coprococcus, Ruminococcus (Lachnospiraceae), and Allobaculum, and higher production of acetate and butyrate. We concluded that multiple regulatory mechanisms of diets which affect microbiota composition exist, including microbial metabolism, microbial migration, and the regulation of the intestinal environment.

Research on Super-resolution Reconstruction Method of Lidar 3D Range Profile

At present, how to use low-cost and superior algorithms to obtain high-resolution 3D range image is the focus of lidar research. In this letter, the low-resolution Gm-APD lidar is combined with the high-resolution ICCD lidar to obtain the registered low-resolution range image and high-resolution intensity image. This letter proposes an improved image guidance algorithm. The algorithm uses a Markov random field model to define a global energy function. This function combines the distance fidelity term and the regularization term to obtain a high-resolution 3D range image by solving the optimization model. The experimental results show that compared with the traditional algorithms, the algorithm improves the resolution of the range images, the edge of the reconstructed image is sharper than the regional similarity guidance algorithm, and the image quality evaluation index has the better value.

Populational Level Driving Force of COVID-19 Epidemic and Fatality: A Global Analysis

BackgroundThe COVID-19 has become a pandemic worldwide. 216 countries and regions reported their occurrence of COVID-19 cases by 19th Sep 2020. Meanwhile, the COVID-19 had infected more than 30 million people and caused almost one million deaths until 19th Sep 2020.MethodWe scoped data related to reported epidemic durations, incidences, fatalities, and epidemic risk factors of the studied countries. Disease Development Speed (DDS), Population-Level Incidence (PLI), and Case-Fatality Rate (CFR) were calculated to assess the COVID-19 pandemic globally. The Spearman rank correlation was applied to further explore the relationship among DDS, PLI, CFR, and their influencing factors.Result216 countries and regions had reported their COVID-19 cases by 19th Sep 2020. Globally, the medians of epidemic duration, DDS, PLI, and CFR were 167 days (IQR: 158-175 days; Range: 110-221 days),4.6 (IQR: 3.1-6.0; Range: 1.8-10.1), 17.0 (IQR: 5.3-56.0; Range: 0.36-1758.3), and 2.2% (IQR: 1.3%-3.5%) respectively. DDS, PLI, and CFR were found enjoying positive correlations with the urban population proportion, UHC (Universal Health Coverage) service coverage, the population with basic handwashing facilities at home, the number of reported deaths, and the number of doctors and nurses.ConclusionNationwide development of the COVID-19, its incidence and fatality suggested regional similarity in the world. National population structure, human resources of medical staff, structure of national health expenditure, and the sanitation facility were revealed as vital risk factors for the COVID-19 in epidemiology.

Does Social Similarity Pay Off? Homophily and Venture Capitalists’ Deal Valuation, Downside Risk Protection, and Financial Returns in India

We ask how social similarity between start-up founders and venture capitalists (VCs) influences VCs’ pricing decisions and returns on investments. We conceptualize how regional and caste similarity, two salient aspects of social similarity in India, affect two distinct aspects of deal pricing: premoney valuation and investors’ downside risk protection in the Indian venture capital market. We theorize that VCs reflect the benefits and costs of social similarity by setting higher premoney valuation when investing in companies led by socially similar founders while also minimizing their downside risks in these investments. We expect that social similarity’s impact on pricing is amplified when VCs face greater subjective uncertainty, such as for early-stage deals or if the VCs lack expertise in the start-up company’s product market. Finally, we claim that VCs achieve superior returns on investments when their deal pricing accurately reflects the impact of social similarity. We tested our conceptual model using both parametric and nonparametric methods on a hand-collected data set of all deals that occurred during 2005–2012, and we supplemented our analyses with in-depth, qualitative interviews that contextualize our findings. The pattern of findings on regional similarity are consistent with our model, but the effects of caste in our data are theoretically anomalous. Post hoc analyses to resolve the anomaly suggest an “intrinsic quality” mechanism, whereby higher-caste VCs set higher valuations when matching with lower-caste founders that signal high quality. Overall, our findings offer evidence that VCs incorporate social attributes into deal pricing in nuanced yet boundedly rational ways.

locStra: Fast analysis of regional/global stratification in whole genome sequencing (WGS) studies

AbstractlocStra is an R-package for the analysis of regional and global population stratification in whole genome sequencing studies, where regional stratification refers to the substructure defined by the loci in a particular region on the genome. Population substructure can be assessed based on the genetic covariance matrix, the genomic relationship matrix, and the unweighted/weighted genetic Jaccard similarity matrix. Using a sliding window approach, the regional similarity matrices are compared to the global ones, based on user-defined window sizes and metrics, e.g. the correlation between regional and global eigenvectors. An algorithm for the specification of the window size is provided. As the implementation fully exploits sparse matrix algebra and is written in C++, the analysis is highly efficient. Even on single cores, for realistic study sizes (several thousand subjects, several million RVs per subject), the runtime for the genome-wide computation of all regional similarity matrices does typically not exceed one hour, enabling an unprecedented investigation of regional stratification across the entire genome. The package is applied to three WGS studies, illustrating the varying patterns of regional substructure across the genome and its beneficial effects on association testing.