GeoMAN: Multi-level Attention Networks for Geo-sensory Time Series Prediction

By Yuxuan Liang, Songyu Ke, Junbo Zhang, Xiuwen Yi, Yu Zheng. International Joint Conference on Artificial Intelligence, 2018

527 citations recorded by the source corpus.

Numerous sensors have been deployed in different geospatial locations to continuously and cooperatively monitor the surrounding environment, such as the air quality. These sensors generate multiple geo-sensory time series, with spatial correlations between their readings. Forecasting geo-sensory time series is of great importance yet very challenging as it is affected by many complex factors, i.e., dynamic spatio-temporal correlations and external factors. In this paper, we predict the readings of a geo-sensor over several future hours by using a multi-level attention-based recurrent neural network that considers multiple sensors' readings, meteorological data, and spatial data. More specifically, our model consists of two major parts: 1) a multi-level attention mechanism to model the dynamic spatio-temporal dependencies. 2) a general fusion module to incorporate the external factors from different domains. Experiments on two types of real-world datasets, viz., air quality data and water quality data, demonstrate that our method outperforms nine baseline methods.

Subfields: Artificial Intelligence, Environmental Engineering, Signal Processing

View Paper: https://www.semanticscholar.org/paper/134b77182981451c456ad0a02c1d61f67906f892

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Philip S. Yu
2. Yu Zheng
3. Zezhi Shao
4. Senzhang Wang
5. Jiannong Cao
6. Junbo Zhang
7. Youfang Lin
8. Shengna Guo

In Environmental Engineering

1. Philip S. Yu
2. Senzhang Wang
3. Jiannong Cao
4. Junbo Zhang
5. Yu Zheng
6. Henry Leung

In Signal Processing

1. Junbo Zhang
2. Philip S. Yu
3. Senzhang Wang
4. Lijun Sun
5. Jiannong Cao
6. Henry Leung
7. Zezhi Shao
8. Xinyu Chen

Citations in the corpus

• Bayesian Temporal Factorization for Multidimensional Time Series Prediction (2019, background): “Third, we would like to integrate recent advances in deep learning to better capture the complex and non-linear dynamics in modern time series data [39], [40], [41], [42], [43], [44], [45], [46].”
• Deep Learning for Spatio-Temporal Data Mining: A Survey (2019, methodology): “Although attention mechanisms are used in some previous works to increase the model interpretability such as periodicity and local spatial dependency [19], [50], [80], how to build a more interpretable deep learning model for STDM tasks is still not well studied and remains an open problem.”
• Spatio-Temporal Graph Neural Networks for Predictive Learning in Urban Computing: A Survey (2023, background): “Due to the fluidity of the air, monitoring stations that are geospatially close or sharing the same wind direction may collect correlated results [110], [111], [112].”
• A Review of Deep Learning Models for Time Series Prediction (2019, methodology): “Besides, the real-world examples of using LSTM and GRU for time series analysis are also reported in the application of diagnosis of neurodegenerative diseases [101], hydrologic prediction [102], multi-sensors fusion [98], [103], remote sensing [104]–[106], etc.”
• EA-LSTM: Evolutionary Attention-based LSTM for Time Series Prediction (2018, background): “Additionally, in recent years, some related work [Qin et al. , 2017; Liang et al. , 2018] on time series prediction is improved usually by introducing attention layers into the encoding-decoding framework.”
• Parallel spatio-temporal attention-based TCN for multivariate time series prediction (2021, background): “Our explorations into the domain of sequence modeling to generate these forecasts, i.e., time-series predictions, have taken us from statistical engines to multi-layer perceptrons (MLPs) to recursive models [26], [27].”
• Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting (2019, background): “• GeoMAN (Liang et al. 2018): A multi-level attention-based recurrent neural network model proposed for the geo-sensory time series prediction problem.”