2023 ICDM 时空数据&时间序列论文总结
ICDM（IEEE International Conference on Data Mining）即IEEE国际数据挖掘会议，是CCF B类会议,是数据挖掘领域的顶级国际学术会议，。
2023 ICDM目前正在上海召开（2023年12月1日-4日，中国上海）。
本文总结了2023 ICDM上有关时空数据（spatial-temporal）包括交通预测，轨迹预测、聚类、表示学习，时空数据管理（查询优化）和不确定性量化等工作以及时间序列（time series）数据包括时间序列预测，因果分析，反事实等相关论文。论文分为regular（展示20min）和short（展示15min）两部分。
2023 ICDM专门为时空数据和时序数据设置了4个session
时空数据（spatial-temporal）
Regular paper ®
论文：IE-Evo: Internal and External Evolution-Enhanced Temporal Knowledge Graph Forecasting
作者：Kangzheng Liu, Feng Zhao, Guandong Xu, and Shiqing Wu
关键词：时序知识图谱（TKG）预测
论文：CAC: Enabling Customer-Centered Passenger-Seeking for Self-Driving Ride Service with Conservative Actor-Critic
作者：Palawat Busaranuvong, Xin Zhang, Yanhua Li, Xun Zhou, and Jun Luo
**关键词：**自动驾驶，离线强化学习。actor-critic
论文： MoonKV: Optimizing Update-Intensive Workloads for NVM-Based Key-Value Stores
作者：Zhenghong Luo, Qian Wang, Tianshan Qu, Haomai Wang, Li Meng, Rong Gu, and Haipeng Dai
关键词：新型非易失性存储器(NVM)，存储优化
论文：Efficient Cardinality and Cost Estimation with Bidirectional Compressor-based Ensemble Learning
作者：Zibo Liang, Xu Chen, Yan Zhao, Jiandong Xie, Kai Zeng, and Kai Zheng
**关键词：**基数估计，查询优化，集成学习
论文：Meteorology-Assisted Spatio-Temporal Graph Network for Uncivilized Urban Event Prediction
作者：Yang Luo, Zehao Gu, Shiyang Zhou, Yun Xiong, and Xiaofeng Gao
关键词：城市时间预测，时空图神经网络
论文：Koopman Invertible Autoencoder: Leveraging Forward and Backward Dynamics for Temporal Modeling
作者：Kshitij Tayal, Arvind Renganathan, Rahul Ghosh, Xiaowei Jia, and Vipin Kumar
**关键词：**长时时空预测
论文：Self-supervised Pre-Training for Robust and Generic Spatial-Temporal Representations
作者：Ruifeng Wang, Yuansheng Liu, Yongshun Gong, Wei Liu, Meng Chen, Yilong Yin, and Yu Zheng
关键词：轨迹表示学习，预训练，自监督
Short paper (S)
论文：Fine-grained Urban Flow Inference with Unobservable Data via Space-Time Attraction Learning
作者：Ruifeng Wang, Yuansheng Liu, Yongshun Gong, Wei Liu, Meng Chen, Yilong Yin, and Yu Zheng
关键词：交通预测，注意力机制
论文：Prediction in Long-Term Evolution: Exploiting the Interaction Between Urban Crowd Flow Variation and POI Transition Patterns
作者：Zhe Zheng, Jingjing Gu, Qiang Zhou, and Xinjiang Lu
关键词：兴趣点（POI），长期演化预测
论文 ：Fed-mSSA: A Federated Approach for Spatio-Temporal Data Modeling Using Multivariate Singular Spectrum Analysis
作者：Jiayu He, Matloob Khushi, Tung Anh Nguyen, and Nguyen H. Tran
关键词：谱分析，时空建模
论文 ：Early Spatiotemporal Event Prediction via Adaptive Controller and Spatiotemporal Embedding
作者：Wei Shao, Ziyan Peng, Yufan Kang, Xiao Xiao, and Zhiling Jin
关键词：早期时空事件预测，时空嵌入
论文：Distribution-Based Trajectory Clustering
作者：Zi Jing Wang, Ye Zhu, and Kai Ming Ting
**关键词：**轨迹聚类，轨迹相似性度量
论文：Continual Trajectory Prediction with Uncertainty-Aware Generative Memory Replay
作者：Xiushi Feng, Shuncheng Liu, Haitian Chen, and Kai Zheng
关键词：轨迹预测，不确定性（感知）
论文：Efficient Multi-source Contact Event Query Processing for Moving Objects
作者：Pengyue Li, Hua Dai, Yu Chen, Bohan Li, and Geng Yang
**关键词：**查询优化
论文：STSD: Modeling Spatial Temporal Staticity and Dynamicity in Traffic Forecasting
作者：Guanghui Zhu, Haojun Hou, Peiliang Wang, Chunfeng Yuan, and Yihua Huang
关键词：时空动态性和静态性，交通预测
论文：Uncertainty-Aware Traffic Prediction under Missing Data
作者：Hao Mei, Junxian Li, Zhiming Liang, Guanjie Zheng, Bin Shi, and Hua Wei
关键词：交通预测，不确定性量化
论文：GeoMixer: The MLP-Based Sequential POI Recommender with Travel Routing Modelling
作者：Tianxing Wang, Can Wang, Hui Tian, and Hong Shen
关键词：MLP（没错又是MLP），POI推荐，路线建模
时间序列（time series）
Regular paper ®
论文：PatternRCA: A Pattern-aware Root Cause Analysis Framework For Multi-Dimensional Time Series
作者：Cheng HE, Fulong Tian, Peijiao Xue, Yuan Wu, Yao Li, Jiajia Li, Zikai Wang, Feng Tan, Hongyang Chen, and Linghe Kong
关键词：因果分析，多维时间序列
论文：PatternRCA: A Pattern-aware Root Cause Analysis Framework For Multi-Dimensional Time Series
作者：Cheng HE, Fulong Tian, Peijiao Xue, Yuan Wu, Yao Li, Jiajia Li, Zikai Wang, Feng Tan, Hongyang Chen, and Linghe Kong
关键词：因果分析，多维时间序列
论文： Explainable Adaptive Tree-Based Model Selection for Time-Series Forecasting
作者：Matthias Jakobs and Amal Saadallah
关键词：可解释性，自适应，时序预测
Short paper (S)
论文：MTT-DynGL: Towards Multidimensional Topology-Oriented Time-Series Dynamic Graphs Learning Model
作者：Chen Shi, Yujie Mao, Yiding Shen, Wenli Xiong, Feng Liu, Chenhui Li, and Changbo Wang
关键词：动态图
论文：Rethinking Temporal Dependencies in Multiple Time Series: A Use Case in Financial Data
作者：Patrick Asante Owusu, Etienne Gael Tajeuna, Jean-Marc Patenaude, Armelle Brun, and Shengrui Wang
**关键词：**时间独立性，多元时间序列，金融数据
论文：Compatible Transformer for Irregularly Sampled Multivariate Time Series
作者：Yuxi Wei, Juntong Peng, Tong He, Chenxin Xu, Jian Zhang, Shirui Pan, and Siheng Chen
关键词：多元时间序列，不规则采样
论文：Rethinking Temporal Dependencies in Multiple Time Series: A Use Case in Financial Data
作者：Patrick Asante Owusu, Etienne Gael Tajeuna, Jean-Marc Patenaude, Armelle Brun, and Shengrui Wang
关键词：时间独立性，多元时间序列，金融数据
论文：GAFNO: Gated Adaptive Fourier Neural Operator for Task-Agnostic Time Series Modeling
作者：Xin-Yi Li and Yu-Bin Yang
关键词：自适应，傅里叶变换
论文：A Symbolic Representation of Two-Dimensional Time Series for Arbitrary Length DTW Motif
作者：Makoto Imamura and Takaaki Nakamura
关键词：DTW算法，二维时间序列
论文： Counterfactual Explanations for Time Series Forecasting
作者：Zhendong Wang, Ioanna Miliou, Isak Samsten, and Panagiotis Papapetrou
关键词：反事实，时间序列预测
Tutorial
今年ICDM还有一个阿里巴巴达摩院关于时间序列的tutorial：
稳健的时间序列分析和应用：跨学科方法：Robust Time Series Analysis and Applications: A Interdisciplinary Approach
相关链接
**ICDM 2023 论文（program schedule）：**https://cloud-conf.net/icdm2023/schedule.html
NeurIPS 2023 时间序列(Time Series)论文总结
759
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
