Spatio-temporal dependence-based tensor fusion for thermocline analysis in Argo data

2020 ◽  
pp. 095965182093373
Author(s):  
Yu Jiang ◽  
Shenggeng Lin ◽  
Jinjian Ruan ◽  
Hong Qi
Keyword(s):  
Short Term Memory ◽  
Spatial Dimension ◽  
Future Trend ◽  
Time Dimension ◽  
Temporal Dependence ◽  
Argo Data ◽  
Important Research Topic ◽  
Long Short Term Memory ◽  
Spatio Temporal ◽  
Argo Project

As the ocean data acquired by the Argo project is increasingly huge, how to use artificial intelligence to analyze it so as to discover the distribution and variation of ocean temperature with space and time becomes an important research topic in the world. In this article, a spatio-temporal dependence-based tensor fusion method is proposed, which can be used to determine and analyze the thermocline. In the time dimension, long short-term memory is used to predict the temperature of seawater; in the spatial dimension, the thermocline is found incrementally by using tensor analysis. Experiments on BOA Argo data from 2004 to 2016 show that the proposed method can accurately determine the boundary of the thermocline and predict the future trend of the thermocline.

scholarly journals VN-INDEX TREND PREDICTION USING LONG-SHORT TERM MEMORY NEURAL NETWORKS

2019 ◽  
Vol 17 (12.2) ◽  
pp. 61
Author(s):  
Nguyen Ngoc Tra ◽  
Ho Phuoc Tien ◽  
Nguyen Thanh Dat ◽  
Nguyen Ngoc Vu
Keyword(s):  
Time Series Data ◽  
Short Term Memory ◽  
Moving Average ◽  
Future Trend ◽  
Series Data ◽  
Short Term ◽  
Trend Prediction ◽  
Temporal Dependence ◽  
Term Memory ◽  
Long Short Term Memory

The paper attemps to forecast the future trend of Vietnam index (VN-index) by using long-short term memory (LSTM) networks. In particular, an LSTM-based neural network is employed to study the temporal dependence in time-series data of past and present VN index values. Empirical forecasting results show that LSTM-based stock trend prediction offers an accuracy of about 60% which outperforms moving-average-based prediction.

scholarly journals Dam Deformation Interpretation and Prediction Based on a Long Short-Term Memory Model Coupled with an Attention Mechanism

2021 ◽  
Vol 11 (14) ◽  
pp. 6625
Author(s):  
Yan Su ◽  
Kailiang Weng ◽  
Chuan Lin ◽  
Zeqin Chen
Keyword(s):  
Short Term Memory ◽  
Attention Mechanism ◽  
Impact Factors ◽  
Nonlinear Prediction ◽  
Time Dimension ◽  
Short Term ◽  
Deformation Prediction ◽  
Term Memory ◽  
Long Short Term Memory ◽  
Dam Deformation

An accurate dam deformation prediction model is vital to a dam safety monitoring system, as it helps assess and manage dam risks. Most traditional dam deformation prediction algorithms ignore the interpretation and evaluation of variables and lack qualitative measures. This paper proposes a data processing framework that uses a long short-term memory (LSTM) model coupled with an attention mechanism to predict the deformation response of a dam structure. First, the random forest (RF) model is introduced to assess the relative importance of impact factors and screen input variables. Secondly, the density-based spatial clustering of applications with noise (DBSCAN) method is used to identify and filter the equipment based abnormal values to reduce the random error in the measurements. Finally, the coupled model is used to focus on important factors in the time dimension in order to obtain more accurate nonlinear prediction results. The results of the case study show that, of all tested methods, the proposed coupled method performed best. In addition, it was found that temperature and water level both have significant impacts on dam deformation and can serve as reliable metrics for dam management.

scholarly journals Earthquake trend prediction using long short-term memory RNN

2019 ◽  
Vol 9 (2) ◽  
pp. 1304 ◽  
Author(s):  
Tanvi Bhandarkar ◽  
Vardaan K ◽  
Nikhil Satish ◽  
S. Sridhar ◽  
R. Sivakumar ◽  
...  
Keyword(s):  
Neural Network ◽  
Short Term Memory ◽  
Future Trend ◽  
Earthquake Forecasting ◽  
Short Term ◽  
Feed Forward Neural Network ◽  
Trend Prediction ◽  
Term Memory ◽  
Area Of Interest ◽  
Long Short Term Memory

<p>The prediction of a natural calamity such as earthquakes has been an area of interest for a long time but accurate results in earthquake forecasting have evaded scientists, even leading some to deem it intrinsically impossible to forecast them accurately. In this paper an attempt to forecast earthquakes and trends using a data of a series of past earthquakes. A type of recurrent neural network called Long Short-Term Memory (LSTM) is used to model the sequence of earthquakes. The trained model is then used to predict the future trend of earthquakes. An ordinary Feed Forward Neural Network (FFNN) solution for the same problem was done for comparison. The LSTM neural network was found to outperform the FFNN. The R^2 score of the LSTM is better than the FFNN’s by 59%.</p>

scholarly journals Spatio-temporal topological relationships between land parcels in cadastral database

2014 ◽  
Vol XL-6 ◽  
pp. 89-92
Author(s):  
W. Song ◽  
F. Zhang
Keyword(s):  
Spatial Dimension ◽  
Temporal Data ◽  
Time Dimension ◽  
3 Dimensional ◽  
Topological Relationships ◽  
Temporal Space ◽  
Temporal Relationships ◽  
Temporal Data Models ◽  
Spatio Temporal ◽  
Spatiotemporal Relationships

There are complex spatio-temporal relationships among cadastral entities. Cadastral spatio-temporal data model should not only describe the data structure of cadastral objects, but also express cadastral spatio-temporal relationships between cadastral objects. In the past, many experts and scholars have proposed a variety of cadastral spatio-temporal data models, but few of them concentrated on the representation of spatiotemporal relationships and few of them make systematic studies on spatiotemporal relationships between cadastral objects. The studies on spatio-temporal topological relationships are not abundant. In the paper, we initially review current approaches to the studies of spatio-temporal topological relationships, and argue that spatio-temporal topological relation is the combination of temporal topology on the time dimension and spatial topology on the spatial dimension. Subsequently, we discuss and develop an integrated representation of spatio-temporal topological relationships within a 3-dimensional temporal space. In the end, based on the semantics of spatiotemporal changes between land parcels, we conclude the possible spatio-temporal topological relations between land parcels, which provide the theoretical basis for creating, updating and maintaining of land parcels in the cadastral database.

scholarly journals Spatio-temporal analysis of the development of the COVID-19 epidemic (pandemic) in Poland: First phase of development

2021 ◽  
Vol 94 (3) ◽  
pp. 325-354
Author(s):  
Jerzy Parysek ◽  
Lidia Mierzejewska
Keyword(s):  
Development Process ◽  
Temporal Analysis ◽  
Spatial Dimension ◽  
Regional Differentiation ◽  
Retirement Age ◽  
Material Base ◽  
Time Dimension ◽  
Spatio Temporal ◽  
Degree Of Similarity

The purpose of this study is to present a description of the course of the COVID-19 epidemic in Poland in the space-time dimension in the period from March 15th to August 8th 2020. The result of the conducted research is a presentation of the regional differentiation of the course of the epidemic in Poland, the comparison of the intensity of SARS-CoV-2 infections in particular voivodeships, the determination of the degree of similarity in the course of the pandemic development process in individual regions (voivodeships) of the country, and also the indication of the factors which could be taken into account when attempting to explain the interregional differences in the course of the epidemic. The conducted research shows, among other things, that: (1) in terms of time, the development of the epidemic was generally monotonic, however the increase in new infections was rather cyclical, (2) in the spatial dimension, the development of the epidemic was rather random, although the greatest number of infections was characteristic of the most populated regions of the country, (3) the level of infections in Poland was mainly positively influenced by: population density, working in industry, people beyond retirement, age as well as a poorly developed material base of inpatient care.

scholarly journals Predicting of air pollutant concentrations based on spatio-temporal attention convolutional LSTM networks

2021 ◽  
Author(s):  
P. Jiang ◽  
I. Bychkov ◽  
J. Liu ◽  
A. Hmelnov
Keyword(s):  
Neural Networks ◽  
Short Term Memory ◽  
Air Pollutant ◽  
Pollutant Concentration ◽  
Support Vector ◽  
Temporal Attention ◽  
Short Term ◽  
Term Memory ◽  
Long Short Term Memory ◽  
Spatio Temporal

Forecasting of air pollutant concentration, which is influenced by air pollution accumulation, traffic flow and industrial emissions, has attracted extensive attention for decades. In this paper, we propose a spatio-temporal attention convolutional long short term memory neural networks (Attention-CNN-LSTM) for air pollutant concentration forecasting. Firstly, we analyze the Granger causalities between different stations and establish a hyperparametric Gaussian vector weight function to determine spatial autocorrelation variables, which is used as part of the input feature. Secondly, convolutional neural networks (CNN) is employed to extract the temporal dependence and spatial correlation of the input, while feature maps and channels are weighted by attention mechanism, so as to improve the effectiveness of the features. Finally, a depth long short term memory (LSTM) based time series predictor is established for learning the long-term and short-term dependence of pollutant concentration. In order to reduce the effect of diverse complex factors on LSTM, inherent features are extracted from historical air pollutant concentration data meteorological data and timestamp information are incorporated into the proposed model. Extensive experiments were performed using the Attention-CNNLSTM, autoregressive integrated moving average (ARIMA), support vector regression (SVR), traditional LSTM and CNN, respectively. The results demonstrated that the feasibility and practicability of Attention-CNN-LSTM on estimating CO and NO concentration.

Sign in / Sign up

Export Citation Format

Share Document