Search

Your search keyword '"Wang, Keli"' showing total 252 results
Start Over Author "Wang, Keli"
252 results on '"Wang, Keli"'

1. Hierarchical accompanying and inhibiting patterns on the spatial arrangement of taxis' local hotspots

Author

Chen, Xiao-Jian, Dong, Quanhua, Xiao, Changjiang, Huang, Zhou, Wang, Keli, Zhang, Weiyu, and Liu, Yu

Subjects
Computer Science - Computers and Society
Abstract

The spatial arrangement of taxi hotspots indicates their inherent distribution relationships, reflecting spatial organization structure and has received attention in urban studies. Previous studies mainly explore large-scale hotspots by visual analysis or simple indexes, where hotspots usually cover the entire central business district, train stations, or dense residential areas, reaching a radius of hundreds or even thousands of meters. However, the spatial arrangement patterns of small-scale hotspots, reflecting the specific popular pick-up and drop-off locations, have not received much attention. This study quantitatively examines the spatial arrangement of fine-grained local hotspots in Wuhan and Beijing, China, using taxi trajectory data. Hotspots are adaptatively identified with sizes of 90m*90m in Wuhan and 105m*105m in Beijing according to identification method. Findings show popular hotspots are typically surrounded by less popular ones, though regions with many popular hotspots inhibit the presence of less popular ones. We term these configurations as hierarchical accompany and inhibiting patterns. Finally, inspired by both patterns, a KNN-based model is developed to describe these relationships, successfully reproducing the spatial distribution of less popular hotspots based on the most popular ones. These insights enhance understanding of local urban structures and support urban planning.

Published
2023

2. Time series anomaly detection with reconstruction-based state-space models

Author

Wang, Fan, Wang, Keli, and Yao, Boyu

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence
Abstract

Recent advances in digitization have led to the availability of multivariate time series data in various domains, enabling real-time monitoring of operations. Identifying abnormal data patterns and detecting potential failures in these scenarios are important yet rather challenging. In this work, we propose a novel unsupervised anomaly detection method for time series data. The proposed framework jointly learns the observation model and the dynamic model, and model uncertainty is estimated from normal samples. Specifically, a long short-term memory (LSTM)-based encoder-decoder is adopted to represent the mapping between the observation space and the latent space. Bidirectional transitions of states are simultaneously modeled by leveraging backward and forward temporal information. Regularization of the latent space places constraints on the states of normal samples, and Mahalanobis distance is used to evaluate the abnormality level. Empirical studies on synthetic and real-world datasets demonstrate the superior performance of the proposed method in anomaly detection tasks.

Published
2023
Full Text
View/download PDF

10. Cost-effectiveness Analysis of Antiepidemic Policies and Global Situation Assessment of COVID-19

Author

Xu, Liyan, Zhang, Hongmou, Deng, Yuqiao, Wang, Keli, Li, Fu, Lu, Qing, Yin, Jie, Di, Qian, Liu, Tao, Yin, Hang, Zhang, Zijiao, Du, Qingyang, Yu, Hongbin, Liu, Aihan, Jiang, Hezhishi, Guo, Jing, Yuan, Xiumei, Zhang, Yun, Liu, Liu, and Liu, Yu

Subjects
Physics - Physics and Society, Quantitative Biology - Populations and Evolution
Abstract

With a two-layer contact-dispersion model and data in China, we analyze the cost-effectiveness of three types of antiepidemic measures for COVID-19: regular epidemiological control, local social interaction control, and inter-city travel restriction. We find that: 1) intercity travel restriction has minimal or even negative effect compared to the other two at the national level; 2) the time of reaching turning point is independent of the current number of cases, and only related to the enforcement stringency of epidemiological control and social interaction control measures; 3) strong enforcement at the early stage is the only opportunity to maximize both antiepidemic effectiveness and cost-effectiveness; 4) mediocre stringency of social interaction measures is the worst choice. Subsequently, we cluster countries/regions into four groups based on their control measures and provide situation assessment and policy suggestions for each group.

Published
2020

22. Mesoscale Ordering 3D Mosaic Self-Assembly of Dopant-Free Hole Transport Material for Perovskite Solar Cells

Author

Liu, Xiaoyuan, primary, Wang, Keli, additional, Li, Yuheng, additional, You, Shuai, additional, Liu, Tuo, additional, Lv, Yifan, additional, Li, Yuan, additional, Wang, Yu, additional, He, Haoxiang, additional, Li, Yinxin, additional, Chen, Qian, additional, Xie, Pengfei, additional, Chen, Yonghua, additional, Li, Xiong, additional, and Xu, Zong-Xiang, additional

Published
2024
Full Text
View/download PDF

32. Theoretical Study of the Temperature- and Pressure-Dependent Rate Constants for Nine Reactions between COn(n= 0–4), Om(m= 1–3), C2O, and C3O2during the Radiolysis of Carbon Dioxide

Author

Tang, Yuhong, Yang, Yu, Cao, Qi, Wang, Keli, Ma, Dandan, Liu, Beibei, Zhu, Quan, and Ma, Jianyi

Abstract

We investigate the reaction pathways of nine important CO2-related reactions using the revDSD-PBEP86-D3(BJ)/jun-cc-pV(T+d)Z level and simultaneously employ an accurate composite method (jun-Cheap) based on coupled-cluster (CC) theory. Subsequently, the Rice–Ramsperger–Kassel–Marcus/master equation (RRKM/ME) is solved to calculate the temperature- and pressure-dependent rate constants. This work investigates reactions involving transition states that have been overlooked in previous literature, including the dissociation of singlet-state C3O2, the triple channel formation of C2O + CO to form C3O2, and the formation of O3+ CO. The results show that CO3is highly prone to dissociation at high temperatures. Finally, the kinetic data show that over a wide temperature range, our calculations are consistent with previous experimental measurements. The majority of the reaction rate constants studied exhibit significant pressure dependence, while the O3+ CO reaction is pressure-independent at low temperatures. These results are instrumental in the development of detailed kinetic models for the CO2radiolysis reaction network.

Published
2024
Full Text
View/download PDF

33. Boosting the Capacitor Performance of the Metal Organic Frameworks Derived V2O5Nanoflower by Cu Doping

Author

Wang, Keli, Wang, Fangxuan, Qian, Lei, Wang, Dandan, and Liu, Yongqi

Abstract

Due to its ultrahigh theoretical capacitance, vanadium pentoxide (V2O5) is considered to be a valid candidate for advanced supercapacitors. However, because of the low electron/electrolyte transfer rate, the capacitive performance still remains to be improved. In this report, Cu doping is adopted to improve the capacitive performance by a two-steps strategy consisting of microwave-assisted solvothermal and postannealing treatments. The electrochemical results indicate that the Cu doping was beneficial for improving the specific capacitance, extending the potential window, and improving the rate ability and long-term stability of V2O5. Furthermore, the mechanism for the performance improvement is explained in detail by combining theoretical calculation and experiments. The results indicated that, compared with that of undoped V2O5, the larger interplanar spacing, better electrical conductivity, a larger proportion of V3+/V4+, and more abundant oxygen vacancies result in an improved capacitive performance. Our proposed Cu-doped V2O5(Cu-V2O5) can be used as both a positive electrode and a negative electrode for the assembly of the symmetric supercapacitor, which can be used as an energy storage device for light emitting diode lamps.

Published
2024
Full Text
View/download PDF

46. Additional file 1 of Quantifying the spatial spillover effects of non-pharmaceutical interventions on pandemic risk

Author

Wang, Keli, Han, Xiaoyi, Dong, Lei, Chen, Xiao-Jian, Xiu, Gezhi, Kwan, Mei-po, and Liu, Yu

Abstract

Additional file 1: Table S1. Best-fit model posterior estimates of key epidemiological parameters for January 6 to March 9, 2020. Table S2. Gamma distributions for monthly reporting delay (Dr). Table S3. Comparison of the results of DIC based on different spatial weight matrices. Table S4. The contribution of direct effects and SSE of all NPIs on COVID-19 cases in 48 regions and Washington D.C. Table S5. The coefficients of different NPI estimates under the dependent variable multiplied by 100. Table S6. The coefficients estimate of direct effects under the dependent variable multiplied by 100. Table S7. The coefficients estimate of spatial spillover effects under the dependent variable multiplied by 100. Table S8. The coefficients estimate of direct effects from spatial panel durbin model. Table S9. The coefficients estimate of spatial spillover effects from spatial panel durbin model. Table S10. The coefficient estimated by our spatial panel model with different weight matrix during Phase A. Table S11. The coefficient estimated by our spatial panel model with different weight matrix during Phase B. Figure S1. The connectivity structure of different spatial weight matrix. Figure S2. Comparison of the change of human mobility estimated by our spatial panel model versus the factual human mobility change of 48 states and Washington D.C. Figure S3. The scatterplot of the change of human mobility estimated by our spatial panel model versus the factual human mobility change in all states. Figure S4. The direct and spatial spillover effect estimates of NPIs in 48 states and Washington D.C. from the spatial panel model. Figure S5. Model fitting to daily case numbers (blue dots) by S-SEIR model in 48 states and Washington D.C. Figure S6. The estimated cases by different model in 48 states (regions) and Washington D.C. Notes: S-SEIR (Direct effects): similar to the S-SEIR model, but only the direct effects of NPIs are considered. S-SEIR (SSE): similar to the S-SEIR model, but only the spatial spillover effects of NPIs are considered. S-SEIR: both the direct and spatial spillover effects are considered. The line and shaded area represent the median and 95% CI, respectively. Figure S7. The estimated national cumulative confirmed cases by changing the intensity of all NPIs in different regions. Figure S8. The national cumulative confirmed cases estimated by changing the intensity of diffferent NPIs in different regions. Figure S9. The estimated national cumulative confirmed cases by restricting the flows of r2, r4, or r5, including all travel flow with these targeted regions (denoted as ’OD’), travel flow from these targeted regions (denoted as ’O’) and travel flow to these targeted regions (denoted as ‘D’). Figure S10. The median differences in national cumulative confirmed cases at t = 30 week are estimated by the complete lockdown (OD), origin-based lockdown (O), destination-based lockdown (D) in r4 or r5 and baseline scenario. Figure S11. The change of intrastate human mobility in each state. Figure S12. The distribution of the dependent variables in different weeks. Figure S13. Estimation results of the change of intrastate human mobility and spatial spillover effects of NPIs by panel SAR model. The matrix during introduction phase is flow weight (8 neighbors) matrix, during lifting phase is the inverse distance matrix with a cut-off distance of 600 km. Figure S14. Estimation results of the change of intrastate human mobility and spatial spillover effects of NPIs by panel SAR model. The matrix during introduction phase is the inverse distance matrix with a cut-off distance of 600 km, and during lifting phase is flow weight (8 neighbors). Figure S15. Estimation results of the change of intrastate human mobility and spatial spillover effects of NPIs by panel SAR model. The matrix during introduction phase is the inverse distance matrix with a cut-off distance of 600 km, and during lifting phase is the inverse distance (the max distance is 800 km). Figure S16. Estimation results from the S-SEIR model, taking changes in actual human mobility as intervention effects. Figure S17. Estimation results from the Sp-SEIR model, the initial seed is the number of cases in the five days from T0 to T0+4. Figure S18. Estimation results from the Sp-SEIR model. During initialization, symptomatic patients are not assigned an initial value. Figure S19. Estimation results from the Sp-SEIR model, the range of movement fixed factor (θ) is from 1 to 10. Figure S20. Estimation results from the Sp-SEIR model. We assume that the infection of asymptomatic patients relative to symptomatic patients also varies over time and region.

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results