Search

China's Ecological Protection Red Lines (ERLs) policy has proven effective in constructing regional ecological security patterns and protecting ecological space. However, the existing methods for the identification of high conservation value areas (HCVAs) usually use physical models, whose parameters and processes are complex and only for a single service, affecting the ERL delineation. In this study, the data-driven machine learning (ML) models were innovatively applied to construct a framework for ERL identification. First, the One-Class Support Vector Machine (OC-SVM) was used to generate negative samples from natural reserves and ecological factors. Second, the supervised ML models were applied to predict the HCVAs by using samples. Third, by applying the same ecological factors, the traditional physical models were used to assess the ecological services of the study area for reference and comparison. Take Chengdu–Chongqing Urban Agglomeration (CY) as a case study, wherein data from 11 factors and 1822 nature reserve samples were prepared for feasibility verification of the proposed framework. The results showed that the area under the receiver operating characteristic curve (AUC) of all ML models was more than 97%, and random forest (RF) achieved the best performance at 99.57%. Furthermore, the land cover had great contributions to the HCVAs prediction, which is consistent with the land use pattern of CY. High-value areas are distributed in the surrounding mountains of CY, with lush vegetation. All of the above results indicated that the proposed framework can accurately identify HCVAs, and that it is more suitable and simpler than the traditional physical model. It can help improve the effectiveness of ERL delimitation and promote the implementation of ERL policies. [ABSTRACT FROM AUTHOR]

Acute type A aortic dissection (AAAD) has a high probability of postoperative adverse outcomes (PAO) after emergency surgery, so exploring the risk factors for PAO during hospitalization is key to reducing postoperative mortality and improving prognosis. An artificial intelligence approach was used to build a predictive model of PAO by clinical data‐driven machine learning to predict the incidence of PAO after total arch repair for AAAD. This study included 380 patients with AAAD. The clinical features that are associated with PAO were selected using the LASSO regression analysis. Six different machine learning algorithms were tried for modeling, and the performance of each model was analyzed comprehensively using receiver operating characteristic curves, calibration curve, precision recall curve, and decision analysis curves. Explain the optimal model through Shapley Additive Explanation (SHAP) and perform an individualized risk assessment. After comprehensive analysis, the authors believe that the extreme gradient boosting (XGBoost) model is the optimal model, with better performance than other models. The authors successfully built a prediction model for PAO in AAAD patients based on the XGBoost algorithm and interpreted the model with the SHAP method, which helps to identify high‐risk AAAD patients at an early stage and to adjust individual patient‐related clinical treatment plans in a timely manner. [ABSTRACT FROM AUTHOR]

In brain–computer interface (BCI) systems, motor imagery electroencephalography (MI-EEG) signals are commonly used to detect participant intent. Many factors, including low signal-to-noise ratios and few high-quality samples, make MI classification difficult. In order for BCI systems to function, MI-EEG signals must be studied. In pattern recognition and other fields, deep learning approaches have recently been successfully applied. In contrast, few effective deep learning algorithms have been applied to BCI systems, especially MI-based systems. In this paper, we address these problems from two aspects based on the characteristics of EEG signals: first, we proposed a combined time–frequency domain data enhancement method. This method guarantees that the size of the training data is effectively increased while maintaining the intrinsic composition of the data. Second, our design consists of a parallel CNN that takes both raw EEG images and images transformed through continuous wavelet transform (CWT) as inputs. We conducted classification experiments on a public data set to verify the effectiveness of the algorithm. According to experimental results based on the BCI Competition IV Dataset2a, the average classification accuracy is 97.61%. A comparison of the proposed algorithm with other algorithms shows that it performs better in classification. The algorithm can be used to improve the classification performance of MI-based BCIs and BCI systems created for people with disabilities. [ABSTRACT FROM AUTHOR]

Scour around monopile foundations is one of the most important phenomena that causes instability of marine structures. Machine learning methods are currently widely used to describe the complex physics of flow-sediment-structure interactions, but their application to local scour around piers and monopile foundations remains a challenge. In this study, three machine learning algorithms, including backpropagation neural networks (BPNN), radial basis function neural network (RBFNN), and support vector machines (SVM), are used to solve the problem of predicting local scour depth around pile foundations for offshore structures in the marine environment. Both the dimensional and dimensionless data combinations are considered to build the models. After the optimal configuration is determined, the models are used to predict the scour depth under pure-wave and combined wave and current conditions, and the statistical indices show the higher accuracy of the SVM model under either data combination. Moreover, the thus obtained normalized results are compared with those computed using validated empirical equations. Results of the sensitivity analysis indicate that the ratio of velocities (U cw) is the most effective parameter for predicting the scour depth around monopile foundations in marine environment, and the findings on parameter investigations were in good agreement with the results from laboratory experiments. • The scour data under wave-only and combined wave and current conditions are jointly used to develop machine learning models. • The model developed with dimensionless combination has more advantages. • SVM model has strongest regression ability by comparing with BPNN, RBFNN and empirical equations. [ABSTRACT FROM AUTHOR]

This article reviews recent advances in causal inference relevant to sociology. We focus on a selective subset of contributions aligning with four broad topics: causal effect identification and estimation in general, causal effect heterogeneity, causal effect mediation, and temporal and spatial interference. We describe how machine learning, as an estimation strategy, can be effectively combined with causal inference, which has been traditionally concerned with identification. The incorporation of machine learning in causal inference enables researchers to better address potential biases in estimating causal effects and uncover heterogeneous causal effects. Uncovering sources of effect heterogeneity is key for generalizing to populations beyond those under study. While sociology has long emphasized the importance of causal mechanisms, historical and life-cycle variation, and social contexts involving network interactions, recent conceptual and computational advances facilitate more principled estimation of causal effects under these settings. We encourage sociologists to incorporate these insights into their empirical research. [ABSTRACT FROM AUTHOR]

Mangrove ecosystems play a dominant role in global, tropical, and subtropical coastal wetlands. Remote sensing plays a central role in mangrove conservation, as it is the preferred tool for monitoring changes in spatiotemporal distribution. To improve correlated estimation accuracies and explore the influencing mechanisms based on the mangrove ground survey, this study used a support vector machine (SVM) machine learning and Res-UNet deep learning algorithms to identify the land area of mangrove forests and the crown surface cover area of mangrove forests in the Hainan Island from 1991 to 2021. Both classification techniques were verified by a confusion matrix, which from 1991 to 2021, revealed overall accuracies of 93.11 ± 1.54% and 96.43 ± 1.15% for SVM and Res-UNet, respectively. Res-UNet was more accurate in identifying the crown surface cover area, whereas SVM was more suitable for obtaining the land area. Furthermore, based on the crown surface cover area of the mangrove forests on the Hainan Island, influencing mechanisms were analyzed through dynamic changes and landscape patterns. Since 1991, the Hainan Island mangrove forest area has increased, with the center of mass moving from coastal areas to the ocean and increasing the overall landscape fragmentation. Moreover, the change in the mangrove forests area was correlated with economic development and the increasingly urban population of the entire island. Altogether, the reliable assessment of the tropical mangrove forest land area and crown surface cover provides an important research foundation for the protection and restoration plans of tropical mangrove forests. [ABSTRACT FROM AUTHOR]

• We put forward a novel taxonomy of existing heterogeneous network representation learning algorithms, including path based algorithms, semantic unit based algorithms and other algorithms. • We introduce and compare the classical heterogeneous network representation learning techniques in detail. Apparently, this survey brings new perspectives to better understand the existing works. • To promote future research of this field for researchers, we summarize the research challenges and provide the future research directions in heterogeneous network representation learning. Heterogeneous information networks usually contain different kinds of nodes and distinguishing types of relations, which can preserve more information than homogeneous information networks. Heterogeneous network representation learning attempts to learn a low-dimensional representation for each node and capture rich semantic information of the given network. Most of existing surveys focus on heterogeneous information network analysis and homogeneous information network representation learning. Although considerable research efforts concentrate on heterogeneous network representation learning, there are few surveys that systematically review the state-of-the-art heterogeneous network representation learning techniques. Motivated by this, we propose a taxonomy of heterogeneous network representation learning algorithms according to different approaches of capturing semantic information in heterogeneous networks, including path based algorithms and semantic unit based algorithms. Moreover, we introduce the typical heterogeneous network representation learning techniques in detail and make a comparative analysis of these techniques. In addition, the research challenges in terms of semantics preserving, data sparsity and scalability are discussed. To tackle these challenges, several potential future research directions for heterogeneous network representation learning are pointed out, including semantic relations extraction, dynamic heterogeneous networks, very large heterogeneous networks and heterogeneous networks construction. [ABSTRACT FROM AUTHOR]

Detecting anomalies in videos is a fundamental issue in public security. The majority of existing deep learning methods often perform anomaly detection based on the behavior or the trajectory of a single target. However, due to the overlaps of the crowd and the low-resolution of monitoring images, the segmentation of population is hard to implement and the features cannot be learned thoroughly, which make the methods be easily disturbed by visual elements and thus may lead to false detection sometimes. To tackle these problems, we propose the influence-aware attention to learn the representative attributes of the whole crowd. Walking pedestrians can be divided into numbers of flows, and in this paper, we aim to measure the consistency of movement patterns in the same stream and the interactions between different streams. Meanwhile, great importance is given to the relation between pedestrians and the circumstance for certain anomalies occur as a result of environmental issues. Specifically, the influence-aware attention module is composed of the motion attention and the location attention, which is designed to quantify the relations in the scene from spatial and temporal aspects. For the lack of abnormal samples, we utilize a dual generator-based framework to learn interactions among normal scenes. Experimental results on six benchmarks verify the effectiveness and robustness of our proposed method. [ABSTRACT FROM AUTHOR]

• A prototype-guided domain-invariant feature representation method is proposed to avoid harmful knowledge transfer from source domain to the target domain. • A modified nearest class prototype (MNCP) method is proposed to predict the target sample in the target subspace, which can make better use of the structure information of the target domain. • A multi-stage adaptive label filtering method is proposed to iteratively optimize the model, which can alleviate the errors introduced by pseudo-labeling. • Extensive experiments demonstrate that our approach is competitive with the mainstream unsupervised domain adaptive approaches. Unsupervised Domain Adaptation transfers knowledge from the source domain to the target domain. It makes remarkable progress in alleviating the label-shortage problem in machine learning. Existing methods focus on aligning the two domain distributions directly. However, due to domain discrepancy, there may be some samples in the source domain being unnecessary or even harmful to the target tasks. Avoiding transferring knowledge from these samples is crucial. Existing researches are limited in this area. To this end, we propose a new unsupervised domain adaptation approach named the prototype-guided feature learning. The proposed method contains three main innovations. Firstly, we propose to utilize the more representative source-domain samples, class prototypes, to learn a domain-invariant subspace with the target samples. Secondly, the modified nearest class prototype method is proposed to predict the target samples by exploiting the structural information of the target domain efficiently. Thirdly, a multi-stage label filtering method is proposed to alleviate the mislabeling problem during training. Extensive experiments manifest that our method is competitive compared to the current mainstream unsupervised domain adaptive methods. [ABSTRACT FROM AUTHOR]

• A hierarchy of machine learning forecast systems is presented and compared with five persistence models. • Difference between hierarchical models enhance the interpretability of forecasts. • LSTM and XGBoost outperform five persistence models. • The introduction of cloud albedo and cloud fraction significantly improves solar forecasts. • Machine learning models can somewhat learn clear-sky irradiance, making the value of adding it limited. Machine learning (ML) models have been applied to forecast solar energy; however, they often lack clarity of interpretability and underlying physics. This work addresses such challenges by developing a hierarchy of ML models that gradually introduce predictors to improve the forecast accuracy based on a physics-based framework. Three ML models (ARIMA, LSTM, and XGBoost) are examined and compared with four physics-informed persistence models reported in Part I and the simple persistence model to assess the improvement of different models. The 7-year measurements at the U.S. Department of Energy's Atmospheric Radiation Measurement's Southern Great Plains Central Facility site are used for forecasts and evaluations. The results reveal that the step-by-step introduction of predictors leads to different improvements for models at different hierarchical levels. Comparison of the ML models with persistence models shows that LSTM and XGBoost outperform all the persistence models, with LSTM having the overall best performance; however, ARIMA underperforms the four physics-informed persistence models. This study demonstrates the importance and utility of incorporating physics into ML models in improving forecast accuracy by introducing a hierarchy of physics-based predictors, distinguishing predictor contributions, and enhancing the ML interpretability. The combined use of Global Horizontal Irradiance (GHI) and Direct Normal Irradiance (DNI) significantly improves the forecast accuracy compared to using individual irradiances alone because the pair contains more information on cloud-radiation interactions. [ABSTRACT FROM AUTHOR]

• A novel machine learning model is presented for remote sensing of cloud properties. • The machine learning model is guided using a physics-based radiative transfer model. • Resulting solar resource data is extensively validated against ground measurements. • Significant improvements are shown in the accuracy of the solar resource data. The National Solar Radiation Database (NSRDB) provides high-resolution spatiotemporal solar irradiance data for the entire globe. The NSRDB uses a two-step Physical Solar Model (PSM) to compute the effects of clouds and other atmospheric variables on the solar radiation reaching the surface of the Earth. Physical and optical cloud properties are fundamental inputs to the PSM and are derived from the National Oceanic and Atmospheric Administration's Geostationary Operational Environmental Satellites. This paper describes recent improvements to the NSRDB driven by physics-guided machine learning methods for cloud property retrieval. The impacts of these new methods on the NSRDB irradiance data are validated using an extensive set of ground measurement sites, showing significant improvement for all sites. On average, the mean absolute percentage error for global horizontal irradiance and direct normal irradiance show reductions of 2.16 and 3.95 percentage points respectively for all daylight conditions, 5.92 and 17.39 percentage points respectively for cloudy conditions, and 9.00 and 22.59 percentage points respectively for gap-filled cloudy conditions. These new methods will help improve the quality and accuracy of the irradiance and cloud data in the NSRDB. [ABSTRACT FROM AUTHOR]

• We propose a novel dynamic network embedding model with structural attention. • Our model can capture the evolving characteristic of dynamic networks. • The proposed method models the process of developing an open triad into a closed triad. • The learned representations can preserve both the first-order and second-order proximities. • The experimental results demonstrate the efficiency of our method. Network embedding aims to learn low-dimensional vector representations for each node in a network, which facilitates various learning tasks such as node classification, link prediction and so on. The majority of existing embedding methods mainly focus on static networks. However, many real-world networks are dynamic and change over time. Although a small number of very recent literatures have been developed for dynamic network embedding, they either need to be retrained without closed-form expression, or suffer high-time complexity. Additionally, a large number of real-world networks may be both large and noisy, presenting great challenges to effective network representation learning. In this paper, we propose a novel method named Dynamic Network Embedding via Structural Attention (DNESA). Specifically, we incorporate the attention mechanism into network embedding, which facilitates our method mainly concentrating on task-related parts of the given graph while avoiding or ignoring noisy parts of the network. Furthermore, we can capture the evolving characteristic of dynamic networks and learn embedding vectors of each node at different time steps by modeling the process of developing an open triad into a closed triad under the attention mechanism. Meanwhile, we carefully design an optimization function for preserving both the first-order and second-order proximities. Empirical experiments conducted on six real-world networks illustrate the efficiency of the proposed method, which outperforms state-of-the-art network embedding methods in applications including link prediction and node classification. [ABSTRACT FROM AUTHOR]