Search

In response to the problem of missing traffic flow data on highways, to solve the problem of insufficient mining of traffic flow characteristics using existing spatiotemporal correlation repair methods, a missing data repair method based on spatiotemporal fusion adversarial network is proposed based on analyzing the spatiotemporal characteristics of traffic flow, this method utilizes the fusion of GRU and GCN to capture the fine-grained spatiotemporal relationships of traffic flow, optimizes the generator and discriminator of the adversarial network, and achieves accurate repair of missing data. Experiments based on real data have shown that under various missing modes and rates, the model can learn the topological relationships of the road network, capture the temporal regularity and spatiotemporal correlation in the data, and effectively impute missing data.

With the rapid development of economy, the increasing number of motor vehicles and the total road mileage, which leads to the increasingly prominent traffic safety problems. In order to explore the quantitative relationship between the built environment and the risk of urban road traffic safety, this paper reconstructs the built environment system based on the ‘5D' element model of the built environment combined with the factors influencing traffic safety risks, and describes the built environment from multiple aspects such as density, diversity &traffic design etc, and then build the gradient lift decision tree model to explore the importance and dependency of variables. The empirical analysis selects a district in Chongqing as the research unit, and the results show that: the RMSE the model was 0.0036, the MAPE was 1.9%, and the determination coefficient R2 was 0.84. GBDT algorithm results shows: the cumulative importance of population density, road facilities, intersection density, secondary road and branch road density, average intersection distance, land use mix, and economic density reaches 77.87%. Some variables show obvious nonlinearity and threshold effect.

The rapid development of highways greatly affects the flow of people, finance, goods, and information between cities, and monitoring the OD flow of travel has become a very important task for intelligent transportation systems (ITS). The temporal dynamics and complex spatial correlations of OD traffic distribution, as well as the sparsity and incompleteness of data caused by uneven traffic distribution, make OD traffic prediction complex and challenging. This paper proposes a multi-task prediction model for OD traffic between highway stations. The model adopts a hard parameter shared multi-task learning network structure, which is divided into sub-task learning inflow trend modules, sub-task learning outflow trend modules, and main task learning modules for OD traffic. At the same time, the attraction intensity matrix between stations is constructed using the population density data as the external feature of the sub-task module for outlet outflow flow, and stronger constraints between tasks are introduced to achieve better fitting results. Finally, an OD flow prediction case experiment was conducted between stations on highways in Sichuan Province. The experimental results showed that the proposed model not only had higher accuracy in predicting results than other baseline models, but also had better effectiveness and robustness. [ABSTRACT FROM AUTHOR]

In order to solve the problem of high time-consuming and resource-consuming quantitative calculation of large-scale industrial control network attack graphs, a key path analysis method for large-scale industrial control networks was proposed. Firstly, the idea of cut set was used to calculate the key nodes set of Bayesian attack graph by combining the atomic attack income in industrial control network, which solved the problem that the current cut set algorithm only considers the key nodes in graph structure. Secondly, a dynamic updating strategy of Bayesian attack graph which only updated the attack probability of key nodes was proposed to efficiently calculate the attack probability of the whole graph and analyze the key path of attack graph. The experimental results show that the proposed method can not only ensure the reliability of the calculation results of large-scale industrial control attack graphs, but also can significantly reduce the time consumption and have a significant improvement in the calculation efficiency.

Background: Chemotactic cytokines play a crucial role in the development of acute myeloid leukemia (AML). Thus, investigating the mechanisms of chemotactic cytokine-related genes (CCRGs) in AML is of paramount importance. Methods: Using the TCGA-AML, GSE114868 , and GSE12417 datasets, differential expression analysis identified differentially expressed CCRGs (DE-CCRGs). These genes were screened by overlapping differentially expressed genes (DEGs) between AML and control groups with CCRGs. Subsequently, functional enrichment analysis and the construction of a protein-protein interaction (PPI) network were conducted to explore the functions of the DE-CCRGs. Univariate Cox regression, least absolute shrinkage and selection operator (LASSO), and multivariate Cox regression analyses identified relevant prognostic genes and developed a prognostic model. Survival analysis of the prognostic gene was performed, followed by functional similarity analysis, immune analysis, enrichment analysis, and drug prediction analysis. Results: Differential expression analysis revealed 6,743 DEGs, of which 29 DE-CCRGs were selected for this study. Functional enrichment analysis indicated that DE-CCRGs were primarily involved in chemotactic cytokine-related functions and pathways. Six prognostic genes (CXCR3, CXCR2, CXCR6, CCL20, CCL4, and CCR2) were identified and incorporated into the risk model. The model's performance was validated using the GSE12417 dataset. Survival analysis showed significant differences in AML overall survival (OS) between prognostic gene high and low expression groups, indicating that prognostic gene might be significantly associated with patient survival. Additionally, nine different immune cells were identified between the two risk groups. Correlation analysis revealed that CCR2 had the most significant positive correlation with monocytes and the most significant negative correlation with resting mast cells. The tumor immune dysfunction and exclusion score was lower in the high-risk group. Conclusion: CXCR3, CXCR2, CXCR6, CCL20, CCL4, and CCR2 were identified as prognostic genes correlated to AML and the tumor immune microenvironment. These findings offerred novel insights into the prevention and treatment of AML. [ABSTRACT FROM AUTHOR]

PbS quantum dots (QDs) have received considerable attention due to their broad and tunable absorption spectra, and because of the in situ passivation effect from PbBr2, the PbS QDs synthesized by PbBr2 precursor aroused many scholars’ interest. However, its synthesis process is complex, and the reaction mechanism remains unclear. In this work, we simplified the synthesis process by introducing oleic acid (OA) and 1-octadecene (ODE) into the lead precursor solution. Monodisperse PbS QDs ranging in size from 3.6 nm to 5.4 nm were synthesized successfully by adjusting the contents of ODE and oleylamine. Moreover, the use of PbBr2 as a lead precursor effectively reduced the surface defects of the PbS QDs by replacing surface hydroxyl (OH) groups. This resulted in improvement of the photoluminescence (PL) decay lifetime and stability of the QDs. Finally, the synthesized QDs were successfully applied in the solar cells. These findings contribute to a deeper understanding of leveraging the advantages of PbS QDs, particularly in the photovoltaic industry. [ABSTRACT FROM AUTHOR]

With the development of NAND flash, the storage density has become larger and larger, but the reliability has become lower and lower. To address this problem, stronger error correction codes, low-density parity-check (LDPC) codes, are widely used in flash memory. However, the direct use of LDPC codes for error correction has a great impact on the read and write performance of SSDs. Given this, this paper proposes a new mentality that combines replica and LDPC decoding methods to improve the read performance of solid state drives (SSDs). Considering the characteristics of the quantization level of LDPC soft-decision decoding, this paper proposes the IDSD (LDPC soft-decision decoding and duplication method) and IDHSD (a method combining duplication and LDPC hybrid decoding)schemes. These schemes are mainly for hot data. When reading the data, the IDSD scheme needs to first judge whether it is soft-decision decoding. Then, according to the number of quantization levels recorded, it needs to select whether to use duplication or soft-decision decoding to correct the error data. The IDHSD scheme does not need to judge the soft-decision decoding but directly uses the duplication after the hard-decision decoding fails. Both schemes reduce the average response time to a certain extent and optimize the read performance of the SSD. The experimental results show that compared with the Native Scheme, the average response delay time of the IDSD scheme decreases by 5–30% and the IDHSD scheme slashes by 5–38%. Compared with the Traditional Refresh Scheme, the average response delay time of the IDSD scheme cuts down by 1–30% and the IDHSD scheme lowers by 1–29%. [ABSTRACT FROM AUTHOR]

Wiskott-Aldrich syndrome (WAS) is a rare X-linked recessive genetic disease characterized by clinical symptoms such as eczema, thrombocytopenia with small platelets, immune deficiency, prone to autoimmune diseases, and malignant tumors. This disease is caused by mutations of the WAS gene encoding WASprotein (WASP). The locus and type of mutations of the WAS gene and the expression quantity of WASP were strongly correlated with the clinical manifestations of patients. We found a novel mutation in the WAS gene (c. 931 + 5 G > C), which affected splicing to produce three abnormal mRNA, resulting in an abnormally truncated WASP. This mutation led to a reduction but not the elimination of the normal WASP population, resulting in causes X-linked thrombocytopenia (XLT) with mild clinical manifestations. Our findings revealed the pathogenic mechanism of this mutation. [ABSTRACT FROM AUTHOR]

This article employs qualitative research methods to explore the urban adaptation and adaptation processes of Chinese migrant children. Through twenty-one in-depth interviews with migrant children, the researchers discovered: The participant migrant children showed a fairly high level of adaptation to the city; their process of urban adaptation emerged as three categories: the U-Type, the J-type, and the "flat line" type. Also explored were long-term acquisition of explicit behaviors, implicit notions, adaptation to the socio-cultural environment, and success in education as difficult aspects of adaptation. After comparing these three processes, the research team developed a theory of migrant children urban adaptation. It hypothesizes that during the urban adaptation process, migrant children will encounter four stages of development: excitement and curiosity, shock and resistance, exploration and adaptation, and finally integration and assimilation.