Filters








2,525 Hits in 3.4 sec

Automatically Inferring Data Quality for Spatiotemporal Forecasting

Sungyong Seo, Arash Mohegh, George Ban-Weiss, Yan Liu
2018 International Conference on Learning Representations  
In this paper, we propose a novel solution that can automatically infer data quality levels of different sources through local variations of spatiotemporal signals without explicit labels.  ...  Spatiotemporal forecasting has become an increasingly important prediction task in machine learning and statistics due to its vast applications, such as climate modeling, traffic prediction, video caching  ...  We thank Michael Tsang for discussion and his comments that greatly improved the manuscript.  ... 
dblp:conf/iclr/SeoMBL18 fatcat:riwjkdrp2nh65ihs3h2s3rjcty

Spatiotemporal deep learning model for citywide air pollution interpolation and prediction [article]

Van-Duc Le, Tien-Cuong Bui, Sang Kyun Cha
2019 arXiv   pre-print
Recently, air pollution is one of the most concerns for big cities. Predicting air quality for any regions and at any time is a critical requirement of urban citizens.  ...  Specially, we introduce how to transform the air pollution data into sequences of images which leverages the using of ConvLSTM model to interpolate and predict air quality for the entire city at the same  ...  If the error is small then we can infer that other spatiotemporal data has a remarkable effect on air pollution interpolation.  ... 
arXiv:1911.12919v1 fatcat:tjp3qmm3kbfnddny4s44wa3nhq

How events unfold

Ting Hua, Liang Zhao, Feng Chen, Chang-Tien Lu, Naren Ramakrishnan
2016 SIGSPATIAL Special  
event forecasting.  ...  There has been significant recent interest in the application of social media analytics for spatiotemporal event mining. However, no structured survey exists to capture developments in this space.  ...  How to utilize social media data for forecasting is an active research topic of current interest.  ... 
doi:10.1145/2876480.2876485 fatcat:34dx66my65b4pobofkioeaqknm

Preface: special issue on geo-social media analytics

Feng Chen, Arnold Boedihardjo, Chang-Tien Lu
2018 Geoinformatica  
Exploiting the open source data in conjunction with their spatiotemporal contexts can enhance our understanding of the physical environment, societal condition, and the dynamic and complex relationships  ...  For example, in the context of disaster response, Twitter feeds and Flikr imageries can provide a rich and valuable avenue for monitoring the spatial distribution of affected areas and population sentiments  ...  addressing the challenges of highly skewed data availability and distributions, optimizing analytical models to handle continuously evolving data, inferring and extracting implicit and ambiguous spatiotemporal  ... 
doi:10.1007/s10707-018-0324-7 fatcat:j7gww2mcxrcofpbuoss5qm4z2u

Building Autocorrelation-Aware Representations for Fine-Scale Spatiotemporal Prediction [article]

Yijun Lin, Yao-Yi Chiang, Meredith Franklin, Sandrah P. Eckel, José Luis Ambite
2021 arXiv   pre-print
We conduct a demonstration of DeepLATTE using publicly available data for an important public health topic, air quality prediction, in a well-studied, complex physical environment - Los Angeles.  ...  This paper presents a novel deep learning architecture, Deep learning predictions for LocATion-dependent Time-sEries data (DeepLATTE), that explicitly incorporates theories of spatial statistics into neural  ...  Finally, the refined spatiotemporal embeddings are fed to fully connected layers to infer air quality prediction for each cell. E.  ... 
arXiv:2112.05313v1 fatcat:6hgnz3p32je2bmtko5og75xb7y

PM2.5 Prediction with a Novel Multi-Step-Ahead Forecasting Model Based on Dynamic Wind Field Distance

Mei Yang, Hong Fan, Kang Zhao
2019 International Journal of Environmental Research and Public Health  
A local stateful LSTM model was employed to capture temporal correlations in historical air quality and meteorological data for each related site.  ...  Then, these temporal features were integrated as a spatiotemporal matrix, and input into CNN for extracting spatiotemporal correlation features.  ...  Acknowledgments: We are very grateful for the comments on this paper from Teacher Steve and. Mao that significantly increase the clarity of this work.  ... 
doi:10.3390/ijerph16224482 fatcat:vvgv43lkibf7na4j5wdhl7ibwe

Poster Abstract: A Dynamic Data-Driven Prediction Model for Sparse Collaborative Sensing Applications [article]

Daniel Zhang, Yang Zhang, Dong Wang
2019 arXiv   pre-print
However, due to the resource constraints, collaborative sensing applications normally only collect measurements from a subset of physical locations and predict the measurements for the rest of locations  ...  This poster presents a new data-driven solution for sparse collaborative sensing applications.  ...  The dynamic spatial weight matrix W t i allows the model to infer unseen data from spatial neighbors.  ... 
arXiv:1909.04111v1 fatcat:wrdpn2ibojhmjoknxrceamog7y

On how to incorporate public sources of situational context in descriptive and predictive models of traffic data

Sofia Cerqueira, Elisabete Arsenio, Rui Henriques
2021 European Transport Research Review  
spatiotemporal traffic data structures: i) georeferenced time series data; ii) origin-destination tensor data; iii) raw traffic event data.  ...  , triggering new opportunities for context-aware traffic data analysis.  ...  Acknowledgements The authors thank the support of CARRIS, METRO and Câmara Municipal de Lisboa) (particularly, Gabinete de Mobilidade and Centro de Operações Integrado) for the data provision and valuable  ... 
doi:10.1186/s12544-021-00519-w fatcat:u3y4vk33orbsjeualot2kqawjm

Deep-learning Architecture for Short-term Passenger Flow Forecasting in Urban Rail Transit [article]

Jinlei Zhang, Feng Chen, Yadi Zhu, Yinan Guo
2020 arXiv   pre-print
Model architecture includes four branches for inflow, outflow, graph-network topology, as well as weather conditions and air quality.  ...  Three time granularities (10, 15, and 30 min) are chosen to conduct short-term passenger flow forecasting.  ...  w represents 11 indicators for weather-condition and air-quality data.  ... 
arXiv:1912.12563v2 fatcat:f7xqbfv3zzgbhmiucobjhharau

A Deep Hybrid Model for Weather Forecasting

Aditya Grover, Ashish Kapoor, Eric Horvitz
2015 Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '15  
We also derive an efficient learning and inference procedure that allows for large scale optimization of the model parameters.  ...  We explore new directions with forecasting weather as a dataintensive challenge that involves inferences across space and time.  ...  Acknowledgments We are grateful to Imke Durre for answering our queries concerning the IGRA dataset.  ... 
doi:10.1145/2783258.2783275 dblp:conf/kdd/GroverKH15 fatcat:62rqdnmt2va3jbcgoc7oeizn4a

Spatiotemporal Feature Learning Based Hour-Ahead Load Forecasting for Energy Internet

Liufeng Du, Linghua Zhang, Xu Wang
2020 Electronics  
In this paper, we analyze the characteristics of the load forecasting task in the Energy Internet context and the deficiencies of existing methods and then propose a data driven approach for one-hour-ahead  ...  for multiple consecutive time points before the forecasted hour.  ...  Consequently, for a given forecast zone, making full use of spatiotemporally related information will be beneficial for forecasting tasks.  ... 
doi:10.3390/electronics9010196 fatcat:tg7a25ya75cplhq3j7pyjhdg4e

An Ontology-Driven Approach for Integrating Intelligence to Manage Human and Ecological Health Risks in the Geospatial Sensor Web

Xiaoliang Meng, Feng Wang, Yichun Xie, Guoqiang Song, Shifa Ma, Shiyuan Hu, Junming Bai, Yiming Yang
2018 Sensors  
We demonstrate this intelligent system through a case study of automatic prediction of air quality and related health risk.  ...  However, there are rarely specific geospatial sensor web (GSW) applications for certain ecological public health questions.  ...  Special thanks to editors and reviewers for providing valuable insight into this article. Conflicts of Interest: The authors declare that they have no conflicts of interest.  ... 
doi:10.3390/s18113619 pmid:30366399 fatcat:msokkn4o3rcklom7ryzig6wndq

Non-Intrusive Reduced Models based on Operator Inference for Chaotic Systems [article]

João Lucas de Sousa Almeida, Arthur Cancellieri Pires, Klaus Feine Vaz Cid, Alberto Costa Nogueira Junior
2022 arXiv   pre-print
This work explores the physics-driven machine learning technique Operator Inference (OpInf) for predicting the state of chaotic dynamical systems.  ...  The quality of the OpInf predictions is assessed via the Normalized Root Mean Squared Error (NRMSE) metric from which the Valid Prediction Time (VPT) is computed.  ...  Acknowledgments ACNJ acknowledges the Brazil's IBM Research laboratory for supporting this work.  ... 
arXiv:2206.01604v2 fatcat:57eksjhamvgnzmaf7o3ad2cddm

Fast Kronecker Inference in Gaussian Processes with non-Gaussian Likelihoods

Seth R. Flaxman, Andrew Gordon Wilson, Daniel B. Neill, Hannes Nickisch, Alexander J. Smola
2015 International Conference on Machine Learning  
We propose new scalable Kronecker methods for Gaussian processes with non-Gaussian likelihoods, using a Laplace approximation which involves linear conjugate gradients for inference, and a lower bound  ...  Our approach has near linear scaling, requiring O(Dn D+1 D ) operations and O(Dn 2 D ) storage, for n training data-points on a dense D > 1 dimensional grid.  ...  Our kernel learning automatically discovered multiscale seasonal trends and our inference generated highly accurate long-range forecasts, with accurate uncertainty intervals. instigated their Proceedings  ... 
dblp:conf/icml/FlaxmanWNNS15 fatcat:cygsk6bobvcjhfe57wxxwibgt4

A Multivariate Frequency-Domain Approach to Long-Lead Climatic Forecasting*

Balaji Rajagopalan, Michael E. Mann, Upmanu Lall
1998 Weather and forecasting  
, and an associated forecasting methodology is introduced.  ...  This combined signal detection/forecasting scheme exhibits significantly greater skill than conventional forecasting approaches in the context of a synthetic example consistent with the adopted paradigm  ...  Formal techniques for automatically detecting and correcting for such errors still need to be developed.  ... 
doi:10.1175/1520-0434(1998)013<0058:amfdat>2.0.co;2 fatcat:fqpcgdrgcvasnocruiht73jtzy
« Previous Showing results 1 — 15 out of 2,525 results