Search

Predicting photovoltaic (PV) power generation is a crucial task in the field of clean energy. Achieving high-accuracy PV power prediction requires addressing two challenges in current deep learning methods: (1) In photovoltaic power generation prediction, traditional deep learning methods often generate predictions for long sequences one by one, significantly impacting the efficiency of model predictions. As the scale of photovoltaic power stations expands and the demand for predictions increases, this sequential prediction approach may lead to slow prediction speeds, making it difficult to meet real-time prediction requirements. (2) Feature extraction is a crucial step in photovoltaic power generation prediction. However, traditional feature extraction methods often focus solely on surface features, and fail to capture the inherent relationships between various influencing factors in photovoltaic power generation data, such as light intensity, temperature, and more. To overcome these limitations, this paper proposes a mid-term PV power prediction model that combines Graph Convolutional Network (GCN) and Informer models. This fusion model leverages the multi-output capability of the Informer model to ensure the timely generation of predictions for long sequences. Additionally, it harnesses the feature extraction ability of the GCN model from nodes, utilizing graph convolutional modules to extract feature information from the ‘query’ and ‘key’ components within the attention mechanism. This approach provides more reliable feature information for mid-term PV power prediction, thereby ensuring the accuracy of long sequence predictions. Results demonstrate that the GCN–Informer model significantly reduces prediction errors while improving the precision of power generation forecasting compared to the original Informer model. Overall, this research enhances the prediction accuracy of PV power generation and contributes to advancing the field of clean energy.

The large flux of online products in today's world makes business reviews a valuable source for consumers for making sound decisions before making online purchases. Reviews are useful for readers in learning more about the product and gauge its quality. Fake reviews and reviewers form the bulk of the review corpus, making review spamming an open research challenge. These spam reviews require detection to nullify their contribution to product recommendations. In the past, researchers and communities have taken spam detection problems as a matter of serious concern. Yet, for all that, there is space for the performance of exploration on large-scale complex datasets. The work contributes towards robust feature selection with derived features that provide more details on malicious reviews and spammers. Ensemble and other standard machine learning techniques are trained and evaluated over optimal feature sets. In addition, the Metapath-based Graph Convolution Network (M-GCN) framework is proposed, which is an implicit knowledge extraction method to automatically capture the complex semantic meaning of reviews from the heterogeneous network. It makes analysis of triplet (users, reviews, and products) relationships in e-commerce sites through examination of Top-n feature sets in a mutually reinforcing manner. The proposed model is demonstrated on Yelp and Amazon benchmark datasets for evaluation of efficacy and it is shown outperforming state-of-the-art techniques with and without graph-utilization, providing an accuracy of 96% in the prediction task. [ABSTRACT FROM AUTHOR]

Aspect-based sentiment analysis (ABSA) aims to determine the sentiment polarity of aspects in a sentence. Recently, graph convolution network (GCN) model combined with attention mechanism has been used for ABSA task over graph structures, achieving promising results. However, these methods of modeling over graph structure fail to consider multiple latent information in the text, i.e., syntax, semantics, context, and so on. In addition, the attention mechanism is vulnerable to noise in sentences. To tackle these problems, in this paper, we construct an efficient text graph and propose a matrix fusion-based graph convolution network (MFLGCN) for ABSA. First, the graph structure is constructed by combining statistics, semantics, and part of speech. Then, we use the sequence model combined with the multi-head self-attention mechanism to obtain the feature representation of the context. Subsequently, the text graph structure and the feature representation of context are fed into GCN to aggregate the information around aspect nodes. The attention matrix is obtained by combining sequence model, GCN and the attention mechanism. Besides, we design a filter layer to alleviate the noise problem in the sentence introduced by the attention mechanism. Finally, in order to make the context representation more effective, attention and filtering matrices are integrated into the model. Experimental results on four public datasets show that our model is more effective than the previous models, demonstrating that using our text graph and matrix fusion can significantly empower ABSA models. [ABSTRACT FROM AUTHOR]

Accurate segmentation of skin lesions in dermoscopic images is essential for early diagnosis and prevention of skin cancer. However, it is still a challenging task due to the large number of artifacts (hairs, bubbles, ink stains, etc.) in dermoscopic images and irregular shapes and blurred borders. In this paper, we propose a novel skin lesion segmentation network-DBCGN. In the feature extraction stage, We use the Transformer branch to build global contextual dependencies across scales against semantic features using three layers of high-level semantic features from the CNN branch. To more effectively couple local–global features, we design the Dual-branch Cascade-feature Fusion Module (DCFM) to implement multi-scale feature fusion of CNN and Transformer. In addition, to outline the boundaries of the target region more accurately, we design the Reverse Graph Reasoning Module (RGRM) which can refine the boundaries of the target region using low-level features. Extensive experiments have shown that our method outperforms the currently popular methods and provides more accurate segmentation of the target region. Furthermore, our method is computationally less complex and more efficient, which is more in line with the practical requirements of actual clinical segmentation. [ABSTRACT FROM AUTHOR]

Predicting human motion based on past observed motion is one of the challenging issues in computer vision and graphics. Existing research works are dealing with this issue by using discriminative models and showing the results for cases that follow a homogeneous distribution (in distribution) and not discussing the issues of the domain shift problem, where training and testing data follow a heterogeneous (out of distribution) problem, which is the reality when such models are used in practice. However, recent research proposed addressing domain shift issues by augmenting the discriminative model with a generative model and obtained better results. In the present investigation, we propose regularizing the extended network by inserting linear layers to minimize the rank of the latent space and train the entire end-to-end network. We regularize the network to strengthen the model to deal effectively with domain shift scenarios. Both training and testing data come from different distribution sets; to deal with this, we toughen our network by adding the extra linear layers to the network encoder. We tested our model with the benchmark datasets, CMU Motion Capture and Human3.6M, and proved that our model outperforms 14 OoD actions of H3.6M and 7 OoD actions of CMU MoCap in terms of the Euclidean distance calculated between predicted and ground truth joint angle values. Our average results of 14 OoD actions for short-term (80, 160, 320, 400) are 0.34, 0.6, 0.96, 1.07, and for CMU MoCap of 7 OoD actions for short-term and long term (80, 160, 320, 400, 1000) are 0.28, 0.45, 0.77, 0.89, 1.46. All these results are much better than the other state-of-the-art results. [ABSTRACT FROM AUTHOR]

As an essential part of traffic management, traffic flow prediction attracts worldwide attention to intelligent traffic systems (ITSs). Complicated spatial dependencies due to the well-connected road networks and time-varying traffic dynamics make this problem extremely challenging. Recent works have focused on modeling this complicated spatial-temporal dependence through graph neural networks with a fixed weighted graph or an adaptive adjacency matrix. However, fixed graph methods cannot address data drift due to changes in the road network structure, and adaptive methods are time consuming and prone to be overfitting because the learning algorithm thoroughly optimizes the adaptive matrix. To address this issue, we propose a principal graph embedding convolutional recurrent network (PGECRN) for accurate traffic flow prediction. First, we propose the adjacency matrix graph embedding (AMGE) generation algorithm to solve the data drift problem. AMGE can model the distribution of spatiotemporal series after data drift by extracting the principal components of the original adjacency matrix and performing an adaptive transformation. At the same time, it has fewer parameters, alleviating overfitting. Then, except for the essential spatial correlations, traffic flow data are also temporally dynamic. We utilize temporal variation by integrating gated recurrent units (GRU) and AMGE to comprise the proposed model. Finally, PGECRN is evaluated on two real-world highway datasets, PeMSD4 and PeMSD8. Compared with the existing baselines, the better prediction accuracy of our model shows that it can accurately and efficiently model traffic flow. [ABSTRACT FROM AUTHOR]

• A few-shot model HDGN is proposed to improve the acquisition of on-graph filtering information in the spectral domain. • A mixed metric combining nonlinear and linear metrics is designed to update edge features more robustly. • HDGN enhances the acquisition of low-frequency information on the graph through the heat kernel function. • The low-frequency information learned from the graph can achieve better performance in few-shot classification tasks. A heat diffusion graph network (HDGN) is proposed in this paper, which retains more similar graph signals in the spectral domain, for few-shot learning. Convolution on the graph is essentially the filtering of the graph signal. Most existing graph-network-based few-shot learning methods process graph signals with high-pass filters to get the difference in information. However, the low-frequency similar information is usually more valuable in the few-shot tasks. A joint low-pass filter is constructed to filter low-frequency graph signals, that is, heat kernel convolution aggregates similar information from neighboring nodes. The obtained low-frequency similarity information is utilized to update the representations of nodes on the graph. In addition, a more robust mixed metric is designed to dynamically update the edge feature of the graph. Predicting Unknown Nodes on Graphs by Alternating Updates of Node Representation and Edge Matrix. The experimental results also demonstrate that HDGN achieves better performance for the few-shot classification task. [ABSTRACT FROM AUTHOR]

Automatic modulation classification (AMC) of underwater acoustic communication signals is of great significance in national defense and marine military. Accurate modulation classification methods can make great contributions to accurately grasping the parameters and characteristics of enemy communication systems. While a poor underwater acoustic channel makes it difficult to classify the modulation types correctly. Feature extraction and deep learning methods have proven to be effective methods for the modulation classification of underwater acoustic communication signals, but their performance is still limited by the complex underwater communication environment. Graph convolution networks (GCN) can learn the graph structured information of the data, making it an effective method for processing structured data. To improve the stability and robustness of AMC in underwater channels, we combined the feature extraction and deep learning methods by fusing the multi-domain features and deep features using GCN. The proposed method takes the relationships among the different multi-domain features and deep features into account. Firstly, a feature graph was built using the properties of the features. Secondly, multi-domain features were extracted from the received signals and deep features were extracted from the signals using a deep neural network. Thirdly, we constructed the input of GCN using these features and the graph. Then, the multi-domain features and deep features were fused by the GCN. Finally, we classified the modulation types using the output of GCN by way of a softmax layer. We conducted the experiments on a simulated dataset and a real-world dataset, respectively. The results show that the AMC based on GCN can achieve a significant improvement in performance compared to the current state-of-the-art methods. Our approach is robust in underwater acoustic channels. [ABSTRACT FROM AUTHOR]

In the area of recommendation systems, one of the fundamental tasks is rating prediction. Most existing neural network methods independently extract user's and item's review features utilizing a parallel convolutional neural network(CNN) and use them as the representation of users and items to predict rating scores. There are two main drawbacks of these methods: (1) They typically only leverage user or item reviews but ignore the latent information provided by user-item interactions. (2) The historical rating scores are not integrated into the representation of users and items, they are simply used as labels to train models. Thus the rating information is not adequately utilized, leading to the prediction performance of these methods is not superior. To remedy these drawbacks mentioned above, in this paper, we build a unified graph convolutional network(GCN) to capture the interaction information between user and item, also obtain additional information provided by reviews and rating scores. As both reviews and ratings carry interactive messages among users and items, they would magnify the learning performance of user-item features. Specifically, we first construct a multi-attributed bipartite graph(MA-bipartite graph) to represent users, items, and their interactions through reviews and ratings. Then we divide the MA-bipartite graph into two sub-graphs according to the attributes of the edge types which represent the user-item interaction in review domain and item domain respectively. Next, an attributed GCN model is explicitly designed to learn latent features of users and items by incorporating review embeddings and rating score weights. Finally, the attention mechanism is carried to fuse user and item features dynamically to conduct the rating prediction. We conduct our experiments on two real-world datasets. The results demonstrate that the proposed model achieved the state-of-the-art performance, which increases the prediction accuracy by more than 3%, compared with baseline methods. [ABSTRACT FROM AUTHOR]

Copyright of China Rural Water & Hydropower is the property of China Rural Water & Hydropower Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Skeleton-based action recognition (SBAR) is an important task in the field of computer vision. Learning effective action representations from skeleton sequences and improving the performance of action recognition models remain challenging problems. To capture effective features from skeleton sequences, a novel feature called a short-term path signature (STPS) is proposed in this work. Based on the STPS, a plug-and-play module is proposed to achieve improved SBAR. In this module, the STPS is applied as input, and a spatial-temporal graph convolutional network (ST-GCN) is used to learn action features. Finally, a multistream ST-GCN is built to achieve SBAR. The proposed method is verified on the NTU-RGB+D dataset. Several ablation experiments are conducted to verify the effectiveness of the proposed module. The experimental results show that the proposed STPS is beneficial for improving the accuracy of action recognition networks. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Well logs play a significant role in geological exploration and oil resource development. Due to the limitation of measurement tools and expensive economic cost, obtaining certain well logs can be challenging. The inhomogeneity of the medium and the complexity of the subsurface conditions also make it difficult for linear regression and traditional neural network methods to provide accurate predictions. Multiple existing methods have difficulty in obtaining and utilizing spatiotemporal dependencies among known well logs to predict unavailable well logs. In this letter, an adaptive spatiotemporal graph neural network (ASTGNN) is designed to capture the spatial dependencies and temporal dependencies among various available well logs. A graph convolution neural network (GCN) is introduced into the field of well logs prediction to extract non-Euclidean spatial characteristics among well logs. In addition, a temporal convolution network (TCN) is employed to capture the temporal correlations. Then, a prediction block (PB) is designed to map the hidden states mined by the GCN and TCN to the target predicted well logs. Experiments on single regression well logs prediction and multiple regression well logs prediction are conducted on the dataset from the Norwegian field and the dataset from Daqing, China, respectively. The results demonstrate the accuracy and superior performance of the designed approach. [ABSTRACT FROM AUTHOR]

The core of large-scale drug virtual screening is to select the binders accurately and efficiently with high affinity from large libraries of small molecules in which non-binders are usually dominant. The binding affinity is significantly influenced by the protein pocket, ligand spatial information, and residue types/atom types. Here, we used the pocket residues or ligand atoms as the nodes and constructed edges with the neighboring information to comprehensively represent the protein pocket or ligand information. Moreover, the model with pre-trained molecular vectors performed better than the one-hot representation. The main advantage of DeepBindGCN is that it is independent of docking conformation, and concisely keeps the spatial information and physical–chemical features. Using TIPE3 and PD-L1 dimer as proof-of-concept examples, we proposed a screening pipeline integrating DeepBindGCN and other methods to identify strong-binding-affinity compounds. It is the first time a non-complex-dependent model has achieved a root mean square error (RMSE) value of 1.4190 and Pearson r value of 0.7584 in the PDBbind v.2016 core set, respectively, thereby showing a comparable prediction power with the state-of-the-art affinity prediction models that rely upon the 3D complex. DeepBindGCN provides a powerful tool to predict the protein–ligand interaction and can be used in many important large-scale virtual screening application scenarios. [ABSTRACT FROM AUTHOR]

Object detection using convolutional neural networks addresses the recognition problem solely in terms of feature extraction and disregards knowledge and experience to explore higher-level relationships between objects. This paper proposed a knowledge graph network based on a graph convolution network to improve the accuracy of baseline detectors. This network can be integrated into any object detection framework. First, this paper created an experience memory module to store information about categories in the database. When inputting the image to the database, an experience vector for it was obtained. The experience data graph was then constructed by counting the co-occurrences of labels in the dataset. Finally, a graph convolutional neural network was used to extract the relationship between the experience vector and the data graph matrix. This relational pattern can help the baseline detector perform better. Several classical object detectors were then evaluated using the COCO, VOC, and KITTI datasets. The results indicated a significant increase for the baseline detector in mAP using the knowledge graph network. [ABSTRACT FROM AUTHOR]

Knowledge base question answering (KBQA) can be divided into two types according to the type of complexity: questions with constraints and questions with multiple hops of relationships. Previous work on knowledge base question answering have mostly focused on entities and relations. In a multihop question, it is insufficient to focus solely on topic entities and their relations since the relation between words also contains some important information. In addition, because the question contains constraints or multiple relationships, the information is difficult to capture, or the constraints are missed. In this paper, we applied a dependency structure to questions that capture relation information (e.g., constraint) between the words in question through a graph convolution network. The captured relation information is integrated into the question for re-encoding, and the information is used to generate and rank query graphs. Compared with existing sequence models and query graph generation models, our approach achieves a 0.8–3% improvement on two benchmark datasets. [ABSTRACT FROM AUTHOR]

Human Motion Prediction is a crucial task in computer vision and robotics. It has versatile application potentials, such as human-robot interactions, human action tracking for airport security systems, autonomous car navigation, and computer gaming, to name a few. However, predicting human motion based on past actions is extremely challenging due to the difficulties in correctly detecting spatial and temporal features. We propose an Inception Residual Block(IRB) to detect temporal features in human poses due to its inherent capability of processing multiple kernels to capture salient features. Here, we propose to use multiple 1-D Convolution Neural Networks (CNN) with different kernel sizes and input sequence lengths and concatenate them to get proper embedding. As kernels stride over different receptive fields, they detect smaller and bigger salient features at multiple temporal scales. Our main contribution is to propose a residual connection between input and the output of the inception block to have a continuity between the previously observed pose and the next predicted pose. With this proposed architecture, it learns prior knowledge much better about human poses, and we achieve much higher prediction accuracy as detailed in the paper. Subsequently, we further propose to feed the output of the IRB as an input to the Graph Convolution Neural Network (GCN) due to its better spatial feature learning capability. We perform a parametric analysis for a better design of our model. Subsequently, we evaluate our approach on the Human 3.6M dataset and CMU MoCap dataset and compare our short-term and long-term predictions with the state-of-the-art papers, where our model outperforms most of the pose results, the detailed reasons of which have been elaborated in the paper. [ABSTRACT FROM AUTHOR]

• This is a comprehensive review of Transformer for Skeleton-based Action Recognition. • This paper proposes a new taxonomy of transformer-style techniques. • This survey aims to help researchers systematically select promising future directions. Skeleton-based action recognition has rapidly become one of the most popular and essential research topics in computer vision. The task is to analyze the characteristics of human joints and accurately classify their behaviors through deep learning technology. Skeleton provides numerous unique advantages over other data modalities, such as robustness, compactness, noise immunity, etc. In particular, the skeleton modality is extremely lightweight, which is especially beneficial for deep learning research in low-resource environments. Due to the non-European nature of skeleton data, Graph Convolution Network (GCN) has become mainstream in the past few years, leveraging the benefits of processing topological information. However, with the explosive development of transformer methods in natural language processing and computer vision, many works have applied transformer into the field of skeleton action recognition, breaking the accuracy monopoly of GCN. Therefore, we conduct a survey using transformer method for skeleton-based action recognition, forming of a taxonomy on existing works. This paper gives a comprehensive overview of the recent transformer techniques for skeleton action recognition, proposes a taxonomy of transformer-style techniques for action recognition, conducts a detailed study on benchmark datasets, compares the algorithm accuracy of standard methods, and finally discusses the future research directions and trends. To the best of our knowledge, this study is the first to describe skeleton-based action recognition techniques in the style of transformers and to suggest novel recognition taxonomies in a review. We are confident that Transformer-based action recognition technology will become mainstream in the near future, so this survey aims to help researchers systematically learn core tasks, select appropriate datasets, understand current challenges, and select promising future directions. [ABSTRACT FROM AUTHOR]

In order to analyze the content contained in the text more accurately and provide suggestions for people' s production and life, a multi-label classification model that combines multi-granularity features and label semantics on the basis of the traditional multilabel classification method based on deep learning was proposed. Bidirectional long short-term memory network(Bi-LSTM) was used in this model to extract multi-grained text features and obtain text features at different levels. The label diagram was constructed by calculating pmi, and the graph convolution network(GCN) was used to extract hidden relationships of labels and obtain text representation with label information. Finally, multi-granularity text features were fused to carry out multi-label text classification. Experiments were conducted on AAPD and news data sets. The results show that the Micro-F1 values of the proposed model are up to 0. 704 and 0. 729, respectively, which verifies the effectiveness of the model. [ABSTRACT FROM AUTHOR]

Traffic forecasting has always been an important part of intelligent transportation systems. At present, spatiotemporal graph neural networks are widely used to capture spatiotemporal dependencies. However, most spatiotemporal graph neural networks use a single predefined matrix or a single self-generated matrix. It is difficult to obtain deeper spatial information by only relying on a single adjacency matrix. In this paper, we present a progressive multi-graph convolutional network (PMGCN), which includes spatiotemporal attention, multi-graph convolution, and multi-scale convolution modules. Specifically, we use a new spatiotemporal attention multi-graph convolution that can extract extensive and comprehensive dynamic spatial dependence between nodes, in which multiple graph convolutions adopt progressive connections and spatiotemporal attention dynamically adjusts each item of the Chebyshev polynomial in graph convolutions. In addition, multi-scale time convolution was added to obtain an extensive and comprehensive dynamic time dependence from multiple receptive field features. We used real datasets to predict traffic speed and traffic flow, and the results were compared with a variety of typical prediction models. PMGCN has the smallest Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE) results under different horizons (H = 15 min, 30 min, 60 min), which shows the superiority of the proposed model. [ABSTRACT FROM AUTHOR]

The malware detection methods based on data flow graph only focused on the data flow information caused by API calls, which ignored the sequence information among API calls. To address this issue, this paper integrated the sequence information of API calls into data flow graph, put forward the concept of dual data flow graph combined with sequence information, gave the method of constructing event logs from the API call sequence, and proposed the mining method of dual data flow graph combined with sequence information. Finally, this paper proposed an improved graph convolution network to classify dual data flow graph combined with sequential information. \The classification result could be used to detect and classify malwares. The experimental results show that the proposed method has better detection accuracy than traditional methods based on data flow graphs. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Signal Processing is the property of Journal of Signal Processing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Cybersecurity has become the most significant research area in the domain of the Internet of Things (IoT) owing to the ever-increasing number of cyberattacks. The rapid penetration of Android platforms in mobile devices has made the detection of malware attacks a challenging process. Furthermore, Android malware is increasing on a daily basis. So, precise malware detection analytical techniques need a large number of hardware resources that are signifi- cantly resource-limited for mobile devices. In this research article, an optimal Graph Convolutional Neural Network-based Malware Detection and classification (OGCNN-MDC) model is introduced for an IoT-cloud environment. The proposed OGCNN-MDC model aims to recognize and categorize malware occurrences in IoT-enabled cloud platforms. The presented OGCNN-MDC model has three stages in total, such as data pre-processing, malware detection and parameter tuning. To detect and classify the malware, the GCNN model is exploited in this work. In order to enhance the overall efficiency of the GCNN model, the Group Mean-based Optimizer (GMBO) algorithm is utilized to appropriately adjust the GCNN parameters, and this phenomenon shows the novelty of the current study. A widespread experimental analysis was conducted to establish the superiority of the proposed OGCNN-MDC model. A comprehensive comparison study was conducted, and the outcomes highlighted the supreme performance of the proposed OGCNN-MDC model over other recent approaches. [ABSTRACT FROM AUTHOR]

Social recommendation systems leverage the abundant social information of users existing in the current Internet to mitigate the problem of data sparsity, ultimately enhancing recommendation performance. However, most existing recommendation systems that introduce social information ignore the negative messages passed by high-order neighbor nodes and aggregate messages without filtering, which results in a decline in the performance of the recommendation system. Considering this problem, we propose a novel social recommendation model based on graph neural networks (GNNs) called the preference-aware light graph convolutional network (PLGCN), which contains a subgraph construction module using unsupervised learning to classify users according to their embeddings and then assign users with similar preferences to a subgraph to filter useless or even negative messages from users with different preferences to attain even better recommendation performance. We also designed a feature aggregation module to better combine user embeddings with social and interaction information. In addition, we employ a lightweight GNN framework to aggregate messages from neighbors, removing nonlinear activation and feature transformation operations to alleviate the overfitting problem. Finally, we carried out comprehensive experiments using two publicly available datasets, and the results indicate that PLGCN outperforms the current state-of-the-art (SOTA) method, especially in dealing with the problem of cold start. The proposed model has the potential for practical applications in online recommendation systems, such as e-commerce, social media, and content recommendation. [ABSTRACT FROM AUTHOR]

Graph neural network-based representation models have shown extraordinary potential in numerous recommender system applications. Previous studies mainly considered high-order connectivity information in a single view from an interaction graph but ignored the individualized information of the users or items, making them vulnerable to over-smoothing problems. In this study, we proposed a dynamic multi-view fusion-based graph convolution network model for recommendation systems. Multiple views were generated for learning on the basis of latent user interest topics from the decomposed matrix, and a continuous awareness mechanism was proposed to maintain the model's focus on the individualized features of the nodes. During the graph learning process, a dynamic aggregation mechanism was designed to adjust the fusion weight of different propagation layers. Lastly, the different features from multiple views were dynamically fused through an attention mechanism and a principal component control mechanism to predict the similarities between users and items. Experimental results of three popular datasets of recommendation systems demonstrated that our method could effectively alleviate the over-smoothing problem and achieved better performance than four state-of-the-art baselines. [ABSTRACT FROM AUTHOR]

Tuberculosis (TB) has long been recognized as a significant health concern worldwide. Recent advancements in noninvasive wearable devices and machine learning (ML) techniques have enabled rapid and cost-effective testing for the real-time detection of TB. However, small datasets are often encountered in biomedical and chemical engineering domains, which can hinder the success of ML models and result in overfitting issues. To address this challenge, we propose various data preprocessing methods and ML approaches, including long short-term memory (LSTM), convolutional neural network (CNN), Gramian angular field-CNN (GAF-CNN), and multivariate time series with MinCutPool (MT-MinCutPool), for classifying a small TB dataset consisting of multivariate time series (MTS) sensor signals. Our proposed methods are compared with state-of-the-art models commonly used in MTS classification (MTSC) tasks. We find that lightweight models are more appropriate for small-dataset problems. Our experimental results demonstrate that the average performance of our proposed models outperformed the baseline methods in all aspects. Specifically, the GAF-CNN model achieved the highest accuracy of 0.639 and the highest specificity of 0.777, indicating its superior effectiveness for MTSC tasks. Furthermore, our proposed MT-MinCutPool model surpassed the baseline MTPool model in all evaluation metrics, demonstrating its viability for MTSC tasks. [ABSTRACT FROM AUTHOR]

A recommender system is an important subject in the field of data mining, where the item rating information from users is exploited and processed to make suitable recommendations with all other users. The recommender system creates convenience for e-commerce users and stimulates the consumption of items that are suitable for users. In addition to e-commerce, a recommender system is also used to provide recommendations on books to read, movies to watch, courses to take or websites to visit. Similarity between users is an important impact for recommendation, which could be calculated from the data of past user ratings of the item by methods of collaborative filtering, matrix factorization or singular vector decomposition. In the development of graph data mining techniques, the relationships between users and items can be represented by matrices from which collaborative filtering could be done with the larger database, more accurate and faster in calculation. All these data can be represented graphically and mined by today’s highly developed graph neural network models. On the other hand, users’ social friendship data also influence consumption habits because recommendations from friends will be considered more carefully than information sources. However, combining a user’s friend influence and the similarity between users whose similar shopping habits is challenging. Because the information is noisy and it affects each particular data set in different ways. In this study, we present the input data processing method to remove outliers which are single reviews or users with little interaction with the items; the next proposed model will combine the social relationship data and the similarity in the rating history of users to improve the accuracy and recall of the recommender system. We perform a comparative assessment of the influence of each data set and calculation method on the final recommendation. We also propose and implement a model and compared it with base line models which include NGCF, LightGCN, WiGCN, SocialLGN and SEPT.

Traditionally, research on three-dimensional (3D) facial reconstruction has focused heavily on methods that use 3D Morphable Models (3DMMs) based on principal component analysis (PCA). Because such methods are linear, they are robust to external noise. The PCA method has limitations when restoring faces that deviate from the training data distribution, particularly when recovering fine details. By contrast, restoration methods utilizing Graph Convolution Networks (GCN) offer the advantages of non-linearity and direct regression of vertex coordinates and colors. However, GCN-based approaches can be prone to overfitting, making them less stable. This study presents a face restoration approach that aims to regress the vertex coordinates and colors of a 3D face model directly from a single wilds 2D facial image. This method demonstrates greater stability and higher accuracy compared to conventional techniques. In addition, Graph Attention Networks (GAT) enhance the restoration performance while separating the networks responsible for facial shape and color, reducing noise caused by interference between different data attributes. Through experiments, we demonstrate the most optimized network structures and training methods and demonstrate improved performance compared to existing approaches.

Recent advances in deep learning have led to increased adoption of convolutional neural networks (CNN) for structural magnetic resonance imaging (sMRI)-based Alzheimer’s disease (AD) detection. AD results in widespread damage to neurons in different brain regions and destroys their connections. However, current CNN-based methods struggle to relate spatially distant information effectively. To solve this problem, we propose a graph reasoning module (GRM), which can be directly incorporated into CNN-based AD detection models to simulate the underlying relationship between different brain regions and boost AD diagnosis performance. Specifically, in GRM, an adaptive graph Transformer (AGT) block is designed to adaptively construct a graph representation based on the feature map given by CNN, a graph convolutional network (GCN) block is adopted to update the graph representation, and a feature map reconstruction (FMR) block is built to convert the learned graph representation to a feature map. Experimental results demonstrate that the insertion of the GRM in the existing AD classification model can increase its balanced accuracy by more than 4.3%. The GRM-embedded model achieves state-of-the-art performance compared with current deep learning-based AD diagnosis methods, with a balanced accuracy of 86.2%.

This paper presents a novel fault detection and identification method for low-voltage direct current (DC) microgrid with meshed configuration. The proposed method is based on graph convolutional network (GCN), which utilizes the explicit spatial information and measurement data of the network topology to identify a fault. It has a more substantial feature extraction ability even in the presence of noise and bad data. The adjacency matrix for GCN is developed by considering the network topology as an inherent graph. The bus voltage and line current samples after faults are regarded as the node attributes. Moreover, the DC microgrid model is developed using PSCAD/EMTDC simulation, and fault simulation is carried out by considering different possible events that include environmental and physical conditions. The performance of the proposed method under different conditions is compared with those of different machine learning techniques such as convolutional neural network (CNN), support vector machine (SVM), and fully connected network (FCN). The results reveal that the proposed method is more effective than others at detecting and classifying faults. This method also possesses better robustness under the presence of noise and bad data.

The development of new drugs is a long and costly process, Computer-aided drug design reduces development costs while computationally shortening the new drug development cycle, in which DTA (Drug-Target binding Affinity) prediction is a key step to screen out potential drugs. With the development of deep learning, various types of deep learning models have achieved notable performance in a wide range of fields. Most current related studies focus on extracting the sequence features of molecules while ignoring the valuable structural information; they employ sequence data that represent only the elemental composition of molecules without considering the molecular structure maps that contain structural information. In this paper, we use graph neural networks to predict DTA based on corresponding graph data of drugs and proteins, and we achieve competitive performance on two benchmark datasets, Davis and KIBA. In particular, an MSE of 0.227 and CI of 0.895 were obtained on Davis, and an MSE of 0.127 and CI of 0.903 were obtained on KIBA.

Abstract Aspect-based sentiment analysis (ABSA) aims to determine the sentiment polarity of aspects in a sentence. Recently, graph convolution network (GCN) model combined with attention mechanism has been used for ABSA task over graph structures, achieving promising results. However, these methods of modeling over graph structure fail to consider multiple latent information in the text, i.e., syntax, semantics, context, and so on. In addition, the attention mechanism is vulnerable to noise in sentences. To tackle these problems, in this paper, we construct an efficient text graph and propose a matrix fusion-based graph convolution network (MFLGCN) for ABSA. First, the graph structure is constructed by combining statistics, semantics, and part of speech. Then, we use the sequence model combined with the multi-head self-attention mechanism to obtain the feature representation of the context. Subsequently, the text graph structure and the feature representation of context are fed into GCN to aggregate the information around aspect nodes. The attention matrix is obtained by combining sequence model, GCN and the attention mechanism. Besides, we design a filter layer to alleviate the noise problem in the sentence introduced by the attention mechanism. Finally, in order to make the context representation more effective, attention and filtering matrices are integrated into the model. Experimental results on four public datasets show that our model is more effective than the previous models, demonstrating that using our text graph and matrix fusion can significantly empower ABSA models.

Software bug repair is a problem that cannot be ignored in the process of software life.How to efficiently assign software bugs automatically is a very important research direction.Now,the existing research methods mainly focus on the bugreport’s text content or the low-level information of the developers’ tossing network,while ignoring the high-level topology information in the tossing network.Therefore,this paper proposes a software bug automatic assignment model MFD-GCN based on the developers’ multivariate features.Model fully considers the high-level topological features in the developers’ tossing network,and uses the powerful network feature extraction capabilities of graph convolution network to fully mine the multivariate features that represent developers’ deep cooperation relationship and fixing preferences,and train the classifier together with the bug text features.The proposed method is evaluated on two large open-source software projects,i.e.,Eclipse and Mozilla.Expe-rimental results show that compared with the mainstream bug-assignment methods proposed in recent years,the MFD-GCN mo-del has achieved state-of-art results in recommending the top K developers.The accuracy rate of top-1 recommendation on the Eclipse and Mozilla project reaches 69.8% and 59.7%,respectively.

Abstract Background Anticancer peptide (ACP) inhibits and kills tumor cells. Research on ACP is of great significance for the development of new drugs, and the prediction of ACPs and non-ACPs is the new hotspot. Results We propose a new machine learning-based method named GCNCPR-ACPs (a Graph Convolutional Neural Network Method based on collapse pooling and residual network to predict the ACPs), which automatically and accurately predicts ACPs using residual graph convolution networks, differentiable graph pooling, and features extracted using peptide sequence information extraction. The GCNCPR-ACPs method can effectively capture different levels of node attributes for amino acid node representation learning, GCNCPR-ACPs uses node2vec and one-hot embedding methods to extract initial amino acid features for ACP prediction. Conclusions Experimental results of ten-fold cross-validation and independent validation based on different metrics showed that GCNCPR-ACPs significantly outperformed state-of-the-art methods. Specifically, the evaluation indicators of Matthews Correlation Coefficient (MCC) and AUC of our predicator were 69.5% and 90%, respectively, which were 4.3% and 2% higher than those of the other predictors, respectively, in ten-fold cross-validation. And in the independent test, the scores of MCC and SP were 69.6% and 93.9%, respectively, which were 37.6% and 5.5% higher than those of the other predictors, respectively. The overall results showed that the GCNCPR-ACPs method proposed in the current paper can effectively predict ACPs.

Abstract Robust 3D mesh watermarking is a traditional research topic in computer graphics, which provides an efficient solution to the copyright protection for 3D meshes. Traditionally, researchers need manually design watermarking algorithms to achieve sufficient robustness for the actual application scenarios. In this paper, we propose the first deep learning-based 3D mesh watermarking network, which can provide a more general framework for this problem. In detail, we propose an end-to-end network, consisting of a watermark embedding sub-network, a watermark extracting sub-network and attack layers. We employ the topology-agnostic graph convolutional network (GCN) as the basic convolution operation, therefore our network is not limited by registered meshes (which share a fixed topology). For the specific application scenario, we can integrate the corresponding attack layers to guarantee adaptive robustness against possible attacks. To ensure the visual quality of watermarked 3D meshes, we design the curvature consistency loss function to constrain the local geometry smoothness of watermarked meshes. Experimental results show that the proposed method can achieve more universal robustness while guaranteeing comparable visual quality.

Human behavior recognition has always been a hot spot for research in computer vision. In this paper, we propose a novel video behavior recognition method based on Actional-Structural Graph Convolution and a Temporal Extension Module under the framework of a Spatio-Temporal Graph Convolution Neural Network, which can optimize the spatial and temporal features simultaneously. The basic network framework of our method consists of three parts: spatial graph convolution module, temporal extension module and attention mechanism module. In the spatial dimension, the action graph convolution is utilized to obtain abundant spatial features by capturing the correlations of distant joint features, and the structural graph convolution expands the existing skeleton graph to acquire the spatial features of adjacent joints. In the time dimension, the sampling range of the temporal graph is expanded for extracting the same and adjacent joints of adjacent frames. Furthermore, attention mechanisms are introduced to improve the performance of our method. In order to verify the effectiveness and accuracy of our method, a large number of experiments were carried out on two standard behavior recognition datasets: NTU-RGB+D and Kinetics. Comparative experiment results show that our proposed method can achieve better performance.

The technology of change detection from a pair of registered satellite remote sensing images is of great importance for the study of landform changes and thus has attracted much attention. Recently, convolutional neural networks have been favoured in many applications due to their excellent feature extraction performance. However, for high-resolution satellite remote sensing images, previous methods are difficult to obtain finer details of change detection. There are many complex scenery and objects in satellite remote sensing images, which leads to the decline of the performance of the network model. In order to better solve the above problems, we propose a graph convolution-based multi-scale network with a channel attention mechanism. Specifically, we first project the initial features at different levels as vertex features, and there is a mutual relationship between the vertices. After that, we learn the correlations between these vertices through a graph reasoning. We then reproject the newly obtained vertex features to the original pixel space to obtain fine feature maps of the same size and fuse fine features of different scales through the attention fusion module. Finally, we realize the change detection of the input bi-temporal image through the feature difference module of the decoder. We separately conduct extensive experiments on three different high-resolution datasets to demonstrate that our method is effective. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is a neurodegenerative disorder that causes gait abnormalities. Early and accurate recognition of PD gait is crucial for effective treatment. Recently, deep learning techniques have shown promising results in PD gait analysis. However, most existing methods focus on severity estimation and frozen gait detection, while the recognition of Parkinsonian gait and normal gait from the forward video has not been reported. In this paper, we propose a novel spatiotemporal modeling method for PD gait recognition, named WM–STGCN, which utilizes a Weighted adjacency matrix with virtual connection and Multi-scale temporal convolution in a Spatiotemporal Graph Convolution Network. The weighted matrix enables different intensities to be assigned to different spatial features, including virtual connections, while the multi-scale temporal convolution helps to effectively capture the temporal features at different scales. Moreover, we employ various approaches to augment skeleton data. Experimental results show that our proposed method achieved the best accuracy of 87.1% and an F1 score of 92.85%, outperforming Long short-term memory (LSTM), K-nearest neighbors (KNN), Decision tree, AdaBoost, and ST–GCN models. Our proposed WM–STGCN provides an effective spatiotemporal modeling method for PD gait recognition that outperforms existing methods. It has the potential for clinical application in PD diagnosis and treatment. [ABSTRACT FROM AUTHOR]