Spatial-Temporal Correlation-Based LSTM Algorithm and Its Application in PM2.5 Prediction

Zhao, Yanming

doi:10.18280/ria.340104

Cited by 8 publications

(9 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Long short-term memory (LSTM) network [33,34], each module contains various learning paths, owing to the longterm dependency of structured learning [11].…”

Section: A Teaching Notion Retrievalmentioning

confidence: 99%

A Knowledge Image Construction Method for Effective Information Filtering and Mining From Education Big Data

et al. 2021

View full text Add to dashboard Cite

Traditionally, education resources are shared insufficiently, and updated slowly; the education data are not utilized adequately. What is worse, the conventional information filtering method cannot effectively mine desired information, if the big data has a heavy noise. This article presents an information mining method from education big data, on the basis of support vector machine (SVM), and cleans the sampled abnormal data through data integration and conversion. Besides, the authors presented a method that automatically builds education knowledge image. Based on the filtered and mined education data, a neural network was designed to retrieve the themes of classroom knowledge, and the education correlations between these notions were recognized from the evaluation data by possibility correlation rules. The results show our method achieved excellent results on teaching notion retrieval and education correlation recognition.

show abstract

“…In Long short-term memory (LSTM) network [33,34], each module contains various learning paths, owing to the longterm dependency of structured learning [11].…”

Section: A Teaching Notion Retrievalmentioning

confidence: 99%

A Knowledge Image Construction Method for Effective Information Filtering and Mining From Education Big Data

et al. 2021

View full text Add to dashboard Cite

show abstract

“…If there is too much noise, the generated basis functions will not have visual selection consistency continuity, which is not worth studying. Therefore, the noise intensity was set at [0, 1,3,5,7,9,11,13,15] in the test, with % as the unit. Based on the training set LSVRC201_train, by the random sampling method and with the sampling parameter L being greater than 80% of the number of elements in the basis function family and the pruning control parameter ε≥0.12, the results of antinoise capacity test was displayed as the Table 6.…”

Section: Comparison Of the Recognition Performance Of Different Algorithmsmentioning

confidence: 99%

“…In 2020, Zhao [14,15] proposed an improved multi-layer LSTM algorithm based on spatial-temporal correlation, and applied it in the dynamic prediction of PM25, which achieved good results. The study proved that: in time series analysis, the improved LSTM algorithm integrating spatial features has better long-term and short-term prediction capabilities, and can realize the spatio-temporal fusion of the prediction results, which solved the problem that the traditional LSTM algorithm could not achieve the fusion of spatial features.…”

Section: Introductionmentioning

confidence: 99%

Slow Feature Extraction Algorithm Based on Visual Selection Consistency Continuity and Its Application

Yang¹,

Zhao²,

Su³

et al. 2021

Self Cite

View full text Add to dashboard Cite

The conventional slow feature analysis (SFA) algorithm has no support of computational theory of vision for primates, nor does it have the ability to learn the global features with visual selection consistency continuity. And what is more, the algorithm is highly complex. Based on this, Slow Feature Extraction Algorithm Based on Visual selection consistency continuity and Its Application was proposed. Inspired by the visual selection consistency continuity theory for primates, this paper replaced the principal component analysis (PCA) method of the conventional SFA algorithm with the myTICA method, extracted the Gabor basis functions of natural images, initialized the basis function family; it used the feature basis expansion algorithm based on visual selection consistency continuity (the VSCC_FBEA algorithm) to replace the polynomial expansion method in the original SFA algorithm to generates the Gabor basis functions of features with long and short-term visual selectivity in the family of basis functions, which solved the drawbacks of the polynomial prediction algorithm; it also designed the Lipschitz consistency constraint, and proposed the Lipschitz-Orthogonal-Pruning-Method (LOPM algorithm) to optimize the basis function family into an over-complete family of basis functions. In addition, this paper used the feature expression method based on visual invariance theory (visual invariance theory -FEM) to establish the set of features of natural images with visual selection consistency continuity. Subsequently, it adopted three error evaluation methods and mySFA classification method to evaluate the proposed algorithm. According to the experimental results, the proposed algorithm showed good prediction performance with respect to the LSVRC2012 data set; compared with the SFA, GSFA, TICA, myICA and mySFA algorithms, the proposed algorithm is correct and feasible; when the classification threshold of the algorithm was set at 8.0, the recognition rate of the proposed algorithm reached 99.66%, and neither of the false recognition rate and the false rejection rate was higher than 0.33%. The proposed algorithm has good performance in prediction and classification, and also shows good anti-noise capacity under limited noise conditions.

show abstract

“…Since the fine PM2.5 has a smaller diameter, it can reach the alveoli and even the bronchioles more quickly, interrupting lung gas interchange. Long-term submission of PM in the air raised the risk of cardiovascular illness, respiratory problems, and lung cancer [1,2]. AQ monitoring systems have been set up in various localities in response to rising public health awareness.…”

Section: Introductionmentioning

confidence: 99%

Exploitation of Advanced Deep Learning Methods and Feature Modeling for Air Quality Prediction

Parthiban¹,

Amudha²,

Sivakumari³

2022

RIA

View full text Add to dashboard Cite

Air pollution is a major issue because Particulate Matter (PM) has a substantially higher effect on human health than other pollutants. Air Quality (AQ) prediction has become critical recently to take action to reduce pollution. This research introduces a unique methodology for assessing the effectiveness of PM10 and PM2.5. Enhanced spatial, temporal sequence-Improved Sparse Auto Encoder with Deep Learning (EISAE-DL) has been proposed to predict AQ affected by the prolonged dependency of air pollution congregation. However, Long Short-Term Memory (LSTM) used in EISAE-DL has suffered from the learning of a long-term dependent sequence of the training dataset. In addition, it is hard to create very reliable AQ forecasts at higher periodic frequencies, such as daily, weekly, or even monthly. This paper proposes Transfer learning (TL) in a Stacked Bidirectional and Unidirectional LSTM to solve the learning issue in LSTM for long-term dependencies. So, EISAE-DL with TL and modified LSTM model is named as EISAE-Deep Transfer Learning (EISAE-DTL). TL with a modified structure can handle large-size datasets effectively. However, training time is increased more than twice for non-transfer learning way of modeling due to TL, Wasserstein Distance-based adversarial learning is proposed in EISAE-DTL to decrease the variances among AQ data collected from any two sites. The proposed work is named EISAE- Enhanced DTL (EISAE-EDTL). The developed EISAE-DTL and EISAE-EDTL models are compared and analyzed with the performance of existing algorithms EISAE-DL, ISAE-DL, TL-BLSTM, MMSL, and ST-DNN. The experimental findings demonstrate the accuracy, precision, sensitivity, specificity, Area Under Curve (AUC), and Matthew's correlation coefficient of the proposed model performs admirably and improves present condition approaches.

show abstract

Spatial-Temporal Correlation-Based LSTM Algorithm and Its Application in PM2.5 Prediction

Cited by 8 publications

References 24 publications

A Knowledge Image Construction Method for Effective Information Filtering and Mining From Education Big Data

A Knowledge Image Construction Method for Effective Information Filtering and Mining From Education Big Data

Slow Feature Extraction Algorithm Based on Visual Selection Consistency Continuity and Its Application

Exploitation of Advanced Deep Learning Methods and Feature Modeling for Air Quality Prediction

Contact Info

Product

Resources

About