DingMing Liu

Dr. DingMing Liu

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 June 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
DLDingMing Liu
DingMing Liu
LWLizheng Wang
Lizheng Wang
YZYanling Zhong
Yanling Zhong
YDYi Dong
Yi Dong
JKJinling Kong
Jinling Kong
Personalized travel itinerary recommendation algorithms are the focus of research in smart tourism and tourism GIS. Aiming to address issues present in travel itinerary recommendations for the increasingly popular “self-drive tour” mode, this study proposes an algorithm based on comprehensive tourist satisfaction to mitigate problems such as the neglect of important relevant factors and low degree of personalization. First, we construct a model of tourist satisfaction for travel itineraries by comprehensively considering factors including time utilization, the attractiveness of attractions, itinerary feasibility, and the diversity of attraction types. Unlike previous studies, we consider dining and accommodation time during the itinerary, the physical condition of tourists, and the diversity of attraction types, and establish penalty functions to flexibly constrain deviations from the expected conditions in itinerary planning. Then, with the optimization of comprehensive tourist satisfaction as the objective, we design a new algorithm to address the itinerary recommendation problem, supporting tourists in selecting must-visit attractions, restaurants, and hotels, as well as personalized preferences such as the sightseeing sequence. The experimental results demonstrate that our proposed algorithm outperforms two baseline algorithms, providing higher comprehensive tourist satisfaction while also exhibiting greater feasibility in itinerary planning. The proposed algorithm effectively addresses the issue of personalized travel itinerary recommendation, presenting an efficient, feasible, and practical solution.

Verified
Journal Article
Open access
PDF file
杨心杨 心成
杨 心成
XXXukang Xie
Xukang Xie
DLDingMing Liu
DingMing Liu
Building clustering is a key challenge in cartographic generalization, where the goal is to group spatially related buildings into semantically coherent clusters while preserving the true distribution patterns of urban structures. Existing methods often rely on either spatial distance or building feature similarity alone, leading to clusters that sacrifice either accuracy or spatial continuity. Moreover, most deep learning-based approaches, including graph attention networks (GATs), fail to explicitly incorporate spatial distance constraints and typically restrict message passing to first-order neighborhoods, limiting their ability to capture long-range structural dependencies. To address these issues, this paper proposes LA-GATs, a multi-feature constrained and spatially adaptive building clustering network. First, a Delaunay triangulation is constructed based on nearest-neighbor distances to represent spatial topology, and a heterogeneous feature matrix is built by integrating architectural spatial features, including compactness, orientation, color, and height. Then, a spatial distance-constrained attention mechanism is designed, where attention weights are adjusted using a distance decay function to enhance local spatial correlation. A second-order neighborhood aggregation strategy is further introduced to extend message propagation and mitigate the impact of triangulation errors. Finally, spectral clustering is performed on the learned similarity matrix. Comprehensive experimental validation on real-world datasets from Xi’an and Beijing, showing that LA-GATs outperforms existing clustering methods in both compactness, silhouette coefficient and adjusted rand index, with up to about 21% improvement in residential clustering accuracy.

31 August 2024 in Toxics
Verified
Journal Article
Open access
PDF file
YZYanling Zhong
Yanling Zhong
YGYong Guo
Yong Guo
DLDingMing Liu
DingMing Liu
QZQiutong Zhang
Qiutong Zhang
LWLizheng Wang
Lizheng Wang
Long-term exposure to PM2.5 pollution increases the risk of cardiovascular diseases, particularly ischemic heart disease (IHD). Current assessments of the health effects related to PM2.5 exposure are limited by sparse ground monitoring stations and applicable disease research cohorts, making accurate health effect evaluations challenging. Using satellite-observed aerosol optical depth (AOD) data and the XGBoost-PM25 model, we obtained 1 km scale PM2.5 exposure levels across China. We quantified the premature mortality caused by PM2.5-exposure-induced IHD using the Global Exposure Mortality Model (GEMM) and baseline mortality data. Furthermore, we employed the Gini coefficient, a measure from economics to quantify inequality, to evaluate the distribution differences in health impacts due to PM2.5 exposure under varying socioeconomic conditions. The results indicate that PM2.5 concentrations in China are higher in the central and eastern regions. From 2007 to 2022, the national overall level showed a decreasing trend, dropping from 47.41 μg/m3 to 25.16 μg/m3. The number of premature deaths attributable to PM2.5 exposure increased from 819 thousand in 2007 to 870 thousand in 2022, with fluctuations in certain regions. This increase is linked to population growth and aging because PM2.5 levels have decreased. The results also indicate disparities in premature mortality from IHD among different economic groups in China from 2007 to 2022, with middle-income groups having a higher cumulative proportion of IHD-related premature deaths compared with high- and low-income groups. Despite narrowing GDP gaps across regions from 2007 to 2022, IHD consistently “favored” the middle-income groups. The highest Gini coefficient was observed in the Northwest (0.035), and the lowest was in the South (0.019). Targeted policy interventions are essential to establish a more equitable atmospheric environment.