Publications

You can also find my articles on my Google Scholar profile.

Journal Articles

Dual Attention with Self-Adaptive Negative Sampling for Session-Based Recommendation

Published in ACM Transactions on Management Information Systems, 2025

In the rapidly evolving landscape of online platforms and e-commerce websites, personalized recommendation plays a crucial role in optimizing user experience. Unlike traditional recommender systems that rely on users’ historical records, session-based recommendation systems aim to provide recommendations to anonymous users without requiring them to log in or expose their personal information. Recently, Graph Neural Networks (GNNs) have gained considerable attention in this field due to their effectiveness and remarkable performance. These graph-based methods often adopt a two-stage process that involves learning static item embeddings in the first stage and using these fixed embeddings to generate user intent in the second stage. However, this two-stage approach failed to leverage the hidden states of each item during information propagation, neglecting the item transitions within varying receptive fields. In this study, we propose a novel framework named Dual Attention with Self-Adaptive Negative Sampling for Session-based Recommendation (DANSeR). Our model employs a dual attention mechanism to capture complex user intent within different receptive fields. Also, we strengthen the discriminative power of our model by adaptively selecting the hard negatives for each session. Furthermore, we incorporate the user behavior on the item set for normalization to achieve personalized recommendations. Extensive experiments on five benchmark datasets demonstrate the superiority of DANSeR over existing session-based recommendation models.

Recommended citation: Yu-Chen Chen*, Pei-Xuan Li*, Hsun-Ping Hsieh, and Chris Shei. 2025. Dual Attention with Self-Adaptive Negative Sampling for Session-Based Recommendation. ACM Trans. Manage. Inf. Syst. Just Accepted (June 2025). https://doi.org/10.1145/3744348 (* Equal contribution)
Download Paper | Download Slides

Prediction for Sensor-less Locations Using Multi-View Graph Fusion Approach with Approximation Module: A Case Study on Dengue Fever Risk Sensor

Published in ACM Transactions on Intelligent Systems and Technology, 2025

Dengue fever is an emergency disease spread by mosquitoes. The most direct way to prevent the disease is to predict risky areas and bolster mosquito preventive strategies. Risk is usually evaluated by monitoring the number of eggs in the ovitraps set up by the government. However, areas without sensors still need to be checked and managed for dengue risk. In this study, we focus on forecasting each region’s fine-grained dengue fever risk, especially in regions without sensor coverage. The paucity of historical data makes this endeavor challenging. Furthermore, determining how to effectively blend different features is another important research challenge and practical issue. We propose a Multi-View Graph Fusion Approach with Approximation Module (MVGAM) to address these two issues. For the regions that have no sensor coverage, MVGAM first uses a feature extractor to learn their representation based on their dynamic and static features. Then we use a graph constructor to formulate the relationship between sensors from different perspectives, and a multi-view graph fusion module to learn the embedding of sensors. Finally, we use an approximation module to deal with the lack of historical data. We conducted experiments using a real-world dataset from the urban area of Tainan, Taiwan. The results show that the proposed MVGAM outperforms the state-of-the-art methods and baselines. The ablation study also shows that every component in MVGAM has a significant impact on boosting the prediction effectiveness.

Recommended citation: Pei-Xuan Li and Hsun-Ping Hsieh. 2025. Prediction for Sensor-less Locations Using Multi-View Graph Fusion Approach with Approximation Module: A Case Study on Dengue Fever Risk Sensor. ACM Trans. Intell. Syst. Technol. Just Accepted (February 2025). https://doi.org/10.1145/3718094
Download Paper | Download Slides

Estimating Future Financial Development of Urban Areas for Deploying Bank Branches: A Local-Regional Interpretable Model

Published in ACM Transactions on Management Information Systems, 2024

Financial forecasting is an important task for urban development. In this article, we propose a novel deep learning framework to predict the future financial potential of urban spaces. To be more precise, our target is to infer the number of financial institutions in the future for any arbitrary location with environmental and geographical data. We propose a novel local-regional model, the Local-Regional Interpretable Multi-Attention (LIMA) model, which considers multiple aspects of a location—the place itself and its surroundings. Besides, our model offers three kinds of interpretability, providing a superior way for decision makers to understand how the model determines the prediction: critical rules learned from the tree-based module, surrounding locations that are highly correlated with the prediction, and critical regional features. Our module not only takes advantage of a tree-based model, which can effectively extract cross features, but also leverages convolutional neural networks to obtain more complex and inclusive features around the target location. Experimental results on real-world datasets demonstrate the superiority of our proposed LIMA model against existing state-of-the-art methods. The LIMA model has been deployed as a web system for assisting one of the largest bank companies in Taiwan to select locations for building new branches in major cities since 2020.

Recommended citation: Pei-Xuan Li, Yu-En Chang, Ming-Chun Wei, and Hsun-Ping Hsieh. 2024. Estimating Future Financial Development of Urban Areas for Deploying Bank Branches: A Local-Regional Interpretable Model. ACM Trans. Manage. Inf. Syst. 15, 2, Article 8 (June 2024), 26 pages. https://doi.org/10.1145/3656479
Download Paper | Download Slides

Enhancing Robust Liver Cancer Diagnosis: A Contrastive Multi-Modality Learner with Lightweight Fusion and Effective Data Augmentation

Published in ACM Transactions on Computing for Healthcare, 2024

This article explores the application of self-supervised contrastive learning in the medical domain, focusing on classification of multi-modality Magnetic Resonance (MR) images. To address the challenges of limited and hard-to-annotate medical data, we introduce multi-modality data augmentation (MDA) and cross-modality group convolution (CGC). In the pre-training phase, we leverage Simple Siamese networks to maximize the similarity between two augmented MR images from a patient, without a handcrafted pretext task. Our approach also combines 3D and 2D group convolution with a channel shuffle operation to efficiently incorporate different modalities of image features. Evaluation on liver MR images from a well-known hospital in Taiwan demonstrates a significant improvement over previous methods. This work contributes to advancing multi-modality contrastive learning, particularly in the context of medical imaging, offering enhanced tools for analyzing complex image data.

Recommended citation: Pei-Xuan Li, Hsun-Ping Hsieh, Yang Fan-Chiang, Ding-You Wu, and Ching-Chung Ko. 2024. Enhancing Robust Liver Cancer Diagnosis: A Contrastive Multi-Modality Learner with Lightweight Fusion and Effective Data Augmentation. ACM Trans. Comput. Healthcare 5, 2, Article 6 (April 2024), 13 pages. https://doi.org/10.1145/3639414
Download Paper | Download Slides

ExoFIA: Deep Exogenous Assistance in the Prediction of the Influence of Fake News with Social Media Explainability

Published in Applied Sciences, 2023

The growth of social platforms has lowered the barrier of entry into the media sector, allowing for the spread of false information and putting democratic politics and social security at peril. Preliminary analysis shows that posts sharing real news and fake news are disseminated on social media. Moreover, posts pointing to fake news spread faster, so this paper aims to predict the impact of posts citing fake news on social platforms. In this study, we take into account that exogenous factors, in addition to endogenous factors, can potentially determine how influential a post is. For example, the occurrence of social events can generate public resonance and discussion, thereby increasing the impact of relevant posts. Given that Google Trends can obtain search trends that reflect social popularity, this work aims to use Google Trends as the source of our exogenous factors. We propose a deep learning model called the deep exogenous aid in fake news (ExoFIA) model, which combines multi-modal features and utilizes an attention mechanism to provide model interpretability and analyze the influencing factors. Applying the model to real-world data from Twitter demonstrates that our model outperforms existing diffusion models. Furthermore, further examination of the relevant aspects of true and fake news reveals that the two are influenced by distinct variables.

Recommended citation: Pei-Xuan Li, Yu-Yun Huang, Chris Shei, Hsun-Ping Hsieh*. (2023) “ExoFIA: Deep Exogenous Assistance in the Fake News Influence Predictor with Social Media Explanability.” Applied Sciences, 2023.
Download Paper | Download Slides

Conference Papers

A Two-Stage Anomaly-Aware Framework for Robust Traffic Forecasting with Memory-Guided GNNs

Published in Proceedings of the 19th ACM International Conference on Web Search and Data Mining, 2026

Accurate traffic forecasting plays a critical role in modern transportation systems by enabling effective congestion management and route optimization. Although recent deep learning-based models have shown substantial progress in modeling complex spatiotemporal dependencies, most existing methods overlook the challenges posed by anomalous traffic conditions. To address this gap, we propose a two-stage anomaly-aware forecasting correction framework. The first stage employs an unsupervised auto-regressive anomaly detector that combines representation learning and spatiotemporal attention to capture normal traffic patterns and filter anomalous inputs through error thresholding. In the second stage, an anomaly optimization predictor leverages a memory module and a sparsity regularization learning strategy to enhance the representation of normal patterns and suppress noise. It models spatiotemporal dependencies using an information propagation layer composed of sequential small-kernel temporal convolutions and memory-guided graph convolutions. Extensive experiments on real-world traffic flow datasets demonstrate that our framework outperforms state-of-the-art models under anomalous conditions.

Recommended citation: Pei-Xuan Li, Cheng-Ru Chou, Jhe-Wei Tsai, Hsun-Ping Hsieh. A Two-Stage Anomaly-Aware Framework for Robust Traffic Forecasting with Memory-Guided GNNs. In Proceedings of the 19th ACM International Conference on Web Search and Data Mining (WSDM '26).
Download Paper | Download Slides

Multi-modal Spatio-temporal Forecasting in Sensor-less Regions: A Dual-stage Graph Approach from Disease to Crime

Published in Proceedings of the 33st ACM International Conference on Advances in Geographic Information Systems, 2025

Spatio-temporal forecasting is critical for urban applications such as epidemic control and crime prevention, yet many existing methods assume dense and consistent sensor data, which is often unavailable due to infrastructural or cost constraints. This work explores the challenge of forecasting in sensor-less regions, where direct temporal observations are missing. Building on two prior studies: Multi-View Graph Fusion Approach with Approximation Module (MVGAM) for disease risk prediction and Graph Disentangler with POI Weighted Module (GDPW), a contrastive learning framework for enhancing POI embedding, we outline a new research direction. Our framework integrates large language models (LLMs), gated recurrent units (GRUs) and multi-layer perceptron (MLP) to encode multi-modal signals, with contrastive learning aligning heterogeneous representations. A dual-stage graph propagation mechanism consolidates knowledge in sensor-rich areas and transfer it to sensor-less regions via localized subgraphs. We anticipate using crime forecasting in Chicago as a case study, this work lays the foundation for robust and interpretable forecasting in data-scarce urban settings.

Recommended citation: Pei-Xuan Li, Hsun-Ping Hsieh. Multi-modal Spatio-temporal Forecasting in Sensor-less Regions: A Dual-stage Graph Approach from Disease to Crime. In Proceedings of the 33st ACM International Conference on Advances in Geographic Information Systems (SIGSPATIAL '25).
Download Paper | Download Slides

MAC: A Multi-Agent Reinforcement Learning Framework with Correctable Strategies for Portfolio Management

Published in 8th International Conference on Knowledge Innovation and Invention, 2025

Portfolio Management (PM) is a broad investment strategy aimed at risk mitigation through diversified financial product investments. Acknowledging the significance of dynamic adjustments after establishing a portfolio to enhance stability and returns, we propose employing Reinforcement Learning (RL) to address dynamic decision-making challenges. However, traditional RL methods often struggle to adapt to significant market volatility, primarily focusing on adjusting existing asset weights. Unlike traditional RL methods, our proposed Multi-Agent Reinforcement Learning Correctable Strategy (MAC) detects and replaces potentially harmful assets with familiar alternatives, ensuring a more resilient response to market crises. Utilizing the Multi-Agent Reinforcement Learning (MARL) model, MAC empowers individual agents to maximize portfolio returns and minimize risk separately. During training, MAC strategically replaces assets to simulate market changes, allowing agents to learn risk identification through uncertainty estima-tion. During testing, MAC dynamically detects potentially harmful assets and replaces them with more confident alternatives, enhancing portfolio stability. Experiments con-ducted on a real-world US Exchange-Traded Fund (ETF) market dataset demonstrate MAC’s superiority over standard RL-based PM methods and other baselines, underscor-ing its practical efficacy for real-world applications.

Recommended citation: Kuang-Da Wang*, Pei-Xuan Li*, Hsun-Ping Hsieh, Wen-Chih Peng. MAC: A Multi-Agent Reinforcement Learning Framework with Correctable Strategies for Portfolio Management. IEEE International Conference on Knowledge Innovation and Invention (ICKII) 2025 (Best paper award, * Equal contribution)
Download Paper | Download Slides

Session-based Recommendation with Multi-granularity User Intent and Dual-channel Sparse Graph Attention Networks

Published in Proceedings of the 2025 Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2025

The session-based recommendation predicts a user’s next action based on their behaviors within the current session. Unlike traditional recommender systems that rely on long-term user data, this system treats users anonymously, without historical information. The main challenge is to extract relevant user preferences from the session while minimizing noise. To address these challenges, we propose Multi-granularity User Intent and Dual-channel Sparse Graph Attention Networks (MG-DSGAT). This model employs a dual-channel sparse attention network to capture both inner-session and cross-session information, enhancing accuracy and reducing the impact of irrelevant items. An additional attention network extracts multi-granularity user information, and a fusion module integrates inner-session, cross-session, and multi-granularity intent to better model complex user behaviors. Extensive experiments on three real-world datasets demonstrate that MG-DSGAT outperforms state-of-the-art session-based recommendation models.

Recommended citation: Pei-Xuan Li, Chia-Lung Lin, Hsun-Ping Hsieh*. Session-based Recommendation with Multi-granularity User Intent and Dual-channel Sparse Graph Attention Networks. In Proceedings of the 2025 Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2025 (PAKDD '25)
Download Paper | Download Slides

FOG: Feature-Oriented Graph Neural Networks for Tabular Data

Published in Proceedings of the 2025 Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2025

Recent advancements in graph neural networks (GNNs) have highlighted their potential for addressing challenges in tabular data prediction by capturing complex inter-sample relationships and relaxing the traditional independent and identically distributed (i.i.d.) assumption. In this work, we present a novel GNN architecture, Feature-Oriented Graph Neural Networks (FOG), specifically designed for tabular data prediction. The FOG model transforms tabular data into feature-oriented graphs and incorporates a feature importance learner to identify distinct feature importance patterns across different samples, enabling it to effectively capture intricate sample interactions. Experimental results demonstrate that FOG achieves state-of-the-art performance on various real-world and synthetic datasets. It accurately identifies key features and delivers feature importance assessments that are highly consistent with those produced by traditional interpretable tree-based models. Additionally, experiments reveal that FOG effectively identifies distinct feature importance patterns across different samples, further enhancing its ability to capture intricate sample interactions. The FOG model aligns with recent trends in GNN-based tabular data prediction, offering an innovative solution that combines enhanced predictive performance with improved interpretability.

Recommended citation: Teng-Yuan Tsou, Pei-Xuan Li, Fandel Lin, Hsun-Ping Hsieh*. FOG: Feature-Oriented Graph Neural Networks for Tabular Data. In Proceedings of the 2025 Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2025 (PAKDD '25)
Download Paper | Download Slides

ACCEPT: A Context-Sensitive, Configurable, and Extensible Prediction Tool using Grid-based Data Processing and Neural Networks in Geospatial Decision Support

Published in Proceedings of the 32nd ACM International Conference on Advances in Geographic Information Systems, 2024

We introduce ACCEPT, a geospatial decision support system that merges robust, intuitive visualization with grid-based data processing and neural networks to enhance spatial data analysis and interpretation in context-sensitive scenarios. It offers versatile machine learning modules with multiple prediction models, tailored to specific requirements with user-defined configurable parameters and flexible predictive target selection. The system serves as an accessible introduction to geographic information systems (GIS) for the general public. The system maps Points of Interest (POIs) to grids, simplifying processes like weighting, intersection, and interpolation, enhancing data accessibility and manipulation. Our case studies show effective handling of spatial data, reflecting similar distribution patterns of POIs, spatial separation, local feature sensitivity, and proximity to infrastructure and kernel size affect evaluations. The extensible and user-friendly web interface includes geospatial data inquiries, overlay, import/export, statistic, and multiple map views, facilitating informed decisions in resource distribution and urban planning. It supports urban planners, analysts, and policymakers in achieving equitable resource distribution and enhancing residential justice, while also providing non-experts an introduction to advanced geospatial analyses, promoting wider engagement and understanding in spatial decision-making.

Recommended citation: Teng-Yuan Tsou, Shih-Yu Lai, Hsuan-Ching Chen, Jung-Tsang Yeh, Pei-Xuan Li, Tzu-Chang Lee, and Hsun-Ping Hsieh. 2024. ACCEPT: A Context-Sensitive, Configurable, and Extensible Prediction Tool using Grid-based Data Processing and Neural Networks in Geospatial Decision Support. In Proceedings of the 32nd ACM International Conference on Advances in Geographic Information Systems (SIGSPATIAL '24). Association for Computing Machinery, New York, NY, USA, 669–672. https://doi.org/10.1145/3678717.3691275
Download Paper | Download Slides

LINKin-PARK: Land Valuation Information and Knowledge in Predictive Analysis and Reporting Kit via Dual Attention-DCCNN

Published in Proceedings of the 33rd ACM International Conference on Information and Knowledge Management, 2024

We present LINKin-PARK, an innovative system that seamlessly merges geographic visualization with an advanced Dual Attention Double Channel Convolutional Neural Network with Multilayer Perceptron (Dual Attention-DCCNN+MLP) to facilitate the efficient analysis of land valuation. LINKin-PARK provides robust visualization capabilities for intuitive comprehension. Our model outperforms traditional methods, e.g., linear regression, multilayer perceptron (MLP), Extreme Gradient Boosting (XGBoost), and the combination of CNN (Convolutional Neural Network) with MLP. An ablation study further evaluates the influence of specific components within the model, revealing that spatial and channel-wise attention mechanisms and the integration of DCCNN and skip connections are crucial for capturing spatial details and improving prediction accuracy. Users have the flexibility to explore and predict developable land valuation based on their specific requirements and provide their feedback to minimize errors in model prediction. For instance, this system can forecast future development potential and market demand for everywhere in an urban space, enabling users to make informed decisions before purchasing a property. Similarly, retailers can anticipate future revenues to aid in strategic decisions, such as selecting optimal locations for establishing new retail outlets. In summary, LINKin-PARK effectively combines geographic visualization and Dual Attention-DCCNN+MLP to assist users in analyzing and predicting land valuation and other scenarios.

Recommended citation: Teng-Yuan Tsou, Shih-Yu Lai, Hsuan-Ching Chen, Jung-Tsang Yeh, Pei-Xuan Li, Tzu-Chang Lee, and Hsun-Ping Hsieh. 2024. LINKin-PARK: Land Valuation Information and Knowledge in Predictive Analysis and Reporting Kit via Dual Attention-DCCNN. In Proceedings of the 33rd ACM International Conference on Information and Knowledge Management (CIKM '24). Association for Computing Machinery, New York, NY, USA, 5289–5293. https://doi.org/10.1145/3627673.3679239
Download Paper | Download Slides

Dengue Risk Detection and Observation System

Published in 7th International Conference on Knowledge Innovation and Invention, 2024

This study aims to develop a dedicated dengue fever prediction and monitoring system for the Tainan City Government to predict dengue fever outbreaks using advanced AI technologies. We compared statistical models, linear models, machine learning (ML), and deep learning (DL) models to construct the system. We found that the Graph WaveNet (Gwinnet) model which is based on graph neural networks, performed best for predicting the total egg count. In contrast, the gradient boosting machine learning algorithm (XGBoost) was most effective for predicting the positivity rate. Using these optimal models, we successfully forecasted the total egg count and positivity rate in all regions of Tainan. The system provides a clear and user-friendly interface for the government to quickly view the relationship between the risk areas and spatially influenced factors of dengue. Using the developed real-time risk warning and monitoring system, the efficiency and effectiveness of dengue fever prevention are improved. The potential of AI technology in public health is confirmed, and the system provides a reference for future epidemic prevention efforts.

Recommended citation: Hung Wei Lee, Hsun-Ping Hsieh, Pei-Xuan Li, Chih Ching Tsao, Ally Chang, Po-Jui Lai, Zheng Lu. Dengue Risk Detection and Observation System. IEEE International Conference on Knowledge Innovation and Invention (ICKII) 2024 (Best paper award)
Download Paper | Download Slides

Exploring Feature Fusion from A Contrastive Multi-Modality Learner for Liver Cancer Diagnosis

Published in Proceedings of the 5th ACM International Conference on Multimedia in Asia, 2023

Self-supervised contrastive learning has achieved promising results in computer vision, and recently it also received attention in the medical domain. In practice, medical data is hard to collect and even harder to annotate, but leveraging multi-modality medical images to make up for small datasets has proved to be helpful. In this work, we focus on mining multi-modality Magnetic Resonance (MR) images to learn multi-modality contrastive representations. We first present multi-modality data augmentation (MDA) to adapt contrastive learning to multi-modality learning. Then, the proposed cross-modality group convolution (CGC) is used for multi-modality features in the downstream fine-tune task. Specifically, in the pre-training stage, considering different behaviors from each MRI modality with the same anatomic structure, yet without designing a handcrafted pretext task, we select two augmented MR images from a patient as a positive pair, and then directly maximize the similarity between positive pairs using Simple Siamese networks. To further exploit multi-modality representation, we combine 3D and 2D group convolution with a channel shuffle operation to efficiently incorporate different modalities of image features. We evaluate our proposed methods on liver MR images collected from a well-known hospital in Taiwan. Experiments show our framework has significantly improved from previous methods.

Recommended citation: Yang Fan Chiang, Pei-Xuan Li, Ding-You Wu, Hsun-Ping Hsieh, and Ching-Chung Ko. 2024. Exploring Feature Fusion from A Contrastive Multi-Modality Learner for Liver Cancer Diagnosis. In Proceedings of the 5th ACM International Conference on Multimedia in Asia (MMAsia '23). Association for Computing Machinery, New York, NY, USA, Article 14, 1–7. https://doi.org/10.1145/3595916.3626383
Download Paper | Download Slides

Forecasting Dengue Fever Risk in Regions without Sensors Using Multi-View Graph Fusion Recurrent Neural Network

Published in Proceedings of the 31st ACM International Conference on Advances in Geographic Information Systems, 2023

Dengue fever is an emergency disease spread by mosquitoes. The most direct way to prevent the disease is to predict risky areas and increase mosquito preventive strategies. Risk is evaluated by monitoring the sensors set up by the government. However, areas without sensors still need to be managed for dengue risk. In this study, we focus on forecasting each region’s fine-grained dengue fever risk, especially in regions without sensor coverage. The lack of historical data makes this endeavor challenging. Furthermore, determining how to effectively blend different features is another challenge. We propose a Multi-View Graph Fusion Recurrent Neural Network (MVGFRNN), which consists of a multi-view graph constructor, graph fusion module, and an approximation module to address these two issues. We conducted experiments using a real-world dataset from the urban area of Tainan, Taiwan. The results show that MVGFRNN outperforms state-of-the-art methods.

Recommended citation: Pei-Xuan Li and Hsun-Ping Hsieh. 2023. Forecasting Dengue Fever Risk in Regions without Sensors Using Multi-View Graph Fusion Recurrent Neural Network. In Proceedings of the 31st ACM International Conference on Advances in Geographic Information Systems (SIGSPATIAL '23). Association for Computing Machinery, New York, NY, USA, Article 86, 1–4. https://doi.org/10.1145/3589132.3625636
Download Paper | Download Slides

ParkFlow: Intelligent Dispersal for Mitigating Parking Shortages Using Multi-Granular Spatial-Temporal Analysis

Published in Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, 2023

Parking behaviors near popular destinations often exhibit a preference for proximity, resulting in poor habits, limited parking availability, and a range of consequential issues such as traffic chaos, economic challenges due to congestion, and imbalanced parking utilization. Taiwan has also faced escalating challenges in this regard. To effectively address these issues, the Government of Taiwan has initiated the Smart City program, encompassing various initiatives to enhance urban functionality. One notable solution implemented under this program is the Smart Parking Meter System (SPMS), designed to enhance the overall parking experience. The SPMS incorporates intelligent billing and secure parking data transmission, ensuring a safer and improved parking environment. In this paper, we propose ParkFlow, a comprehensive software-based solution that seamlessly integrates with smart parking hardware, presenting a holistic approach to tackling these challenges. ParkFlow intelligently disperses parking shortages in highly frequented areas and addresses the problem from multiple perspectives, including user, engineering, and government scenarios. By exploring and addressing these scenarios, we aim to provide valuable insights and inspiration to regions worldwide grappling with similar parking-related difficulties. Based on historical data analysis, the implementation of ParkFlow in resolving the parking imbalance problem is anticipated to lead to a significant increase of up to 10% to 20% in available parking hours in popular areas of Tainan, Taiwan. ParkFlow is in the process of being integrated into the Tainan City Government’s Parking application, indicating its potential to address real-world parking challenges.

Recommended citation: Yang Fan Chiang, Chun-Wei Shen, Jhe-Wei Tsai, Pei-Xuan Li, Tzu-Chang Lee, and Hsun-Ping Hsieh. 2023. ParkFlow: Intelligent Dispersal for Mitigating Parking Shortages Using Multi-Granular Spatial-Temporal Analysis. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management (CIKM '23). Association for Computing Machinery, New York, NY, USA, 5036–5040. https://doi.org/10.1145/3583780.3614751
Download Paper | Download Slides