Search

Recommendation systems address content delivery system information overload. Recent recommender systems include knowledge-aware deep recommender models. These models train user and item embeddings using Graph Neural Networks and operate on collaborative knowledge graphs that combine user-item interactions with item-entity relationships to deliver more item information. Certain models employ a two-step learning procedure wherein the primary objective is to generate recommendations while simultaneously training knowledge graph embeddings. By co-learning, the models are able to comprehend the structural connections that exist between entities and relations, while also improving user embeddings through the use of previous encounters. The objective of this research endeavour is to assess the efficacy of cutting-edge models through the integration of knowledge graph embedding co-learning and the manipulation of embedding techniques. Co-learning boosts quicker learning and enhances the precision of recommendations, according to empirical findings. When a KG embedding layer was incorporated into KGCN, the recall metric outperformed the initial model by as much as 37%. Compared to translational embedding techniques, RotatE decreased training duration by 18% and demonstrated a noteworthy 15% improvement in the recall metric. The study's outcome offer valuable guidance for enhancing recommendation accuracy and reducing training time for models incorporating Knowledge Graph (KG) embeddings. Furthermore, the study introduces a new application for complex-valued KG embeddings. [ABSTRACT FROM AUTHOR]

Brain-computer interfaces (BCIs) have demonstrated immense potential in aiding stroke patients during their physical rehabilitation journey. By reshaping the neural circuits connecting the patient's brain and limbs, these interfaces contribute to the restoration of motor functions, ultimately leading to a significant improvement in the patient's overall quality of life. However, the current BCI primarily relies on Electroencephalogram (EEG) motor imagery (MI), which has relatively coarse recognition granularity and struggles to accurately recognize specific hand movements. To address this limitation, this paper proposes a hybrid BCI framework based on Electroencephalogram and Electromyography (EEG-EMG). The framework utilizes a combination of techniques: decoding EEG by using Graph Convolutional LSTM Networks (GCN-LSTM) to recognize the subject's motion intention, and decoding EMG by using a convolutional neural network (CNN) to accurately identify hand movements. In EEG decoding, the correlation between channels is calculated using Standardized Permutation Mutual Information (SPMI), and the decoding process is further explained by analyzing the correlation matrix. In EMG decoding, experiments are conducted on two task paradigms, both achieving promising results. The proposed framework is validated using the publicly available WAL-EEG-GAL (Wearable interfaces for hand function recovery Electroencephalography Grasp-And-Lift) dataset, where the average classification accuracies of EEG and EMG are 0.892 and 0.954, respectively. This research aims to establish an efficient and user-friendly EEG-EMG hybrid BCI, thereby facilitating the hand rehabilitation training of stroke patients. [ABSTRACT FROM AUTHOR]

3D reconstruction based on single view aims to reconstruct the entire 3D shape of an object from one perspective. When existing methods reconstruct the mesh surface of complex objects, the surface details are difficult to predict and the reconstruction visual effect is poor because the mesh representation is not easily integrated into the deep learning framework; the 3D topology is easily limited by predefined templates and inflexible, and unnecessary mesh self-intersections and connections will be generated when reconstructing complex topology, thus destroying the surface details; the training of the reconstruction network is limited by the large amount of information attached to the mesh vertices, and the training time of the reconstructed network is too long. In this paper, we propose a method for fast mesh reconstruction from single view based on Graph Convolutional Network (GCN) and topology modification. We use GCN to ensure the generation of high-quality mesh surfaces and use topology modification to improve the flexibility of the topology. Meanwhile, a feature fusion method is proposed to make full use of the features of each stage of the image hierarchically. We use 3D open dataset ShapeNet to train our network and add a new weight parameter to speed up the training process. Extensive experiments demonstrate that our method can not only reconstruct object meshes on complex topological surfaces, but also has better qualitative and quantitative results. [ABSTRACT FROM AUTHOR]

• The co-precipitation method and hydrothermal method were used for the synthesis of CuO and MoS 2 respectively. • The sonication method was used for the synthesis of ternary composite CuO/MoS 2 with gCN. • CuO/MoS 2 @gCN showed bandgap energy of 1.52 eV. • CuO/MoS 2 @gCN degraded 85.14 % of MO dye and 63.50 % of phenol. • Scavenging experiment was performed to investigate the active species in degradation reaction. Water from the industries contains various organic effluents which are needed to be removed. Photocatalysis is an efficient method to degrade organic effluents that contaminate water and harmfully affect the environment. This study presents the structural, optical characteristics, and photocatalytic activity of the CuO/MoS 2 and their composite with gCN. The hydrothermal approach was employed to fabricate MoS 2 nanoflowers, and the co-precipitation method to fabricate CuO. The composite with gCN was prepared successfully using an ultrasonication approach. The physio-chemical properties of fabricated photocatalysts were investigated by various characterization techniques. SEM analysis revealed the nanoflower morphology of the prepared MoS 2 sample. The photocatalytic activity was examined by the photodegradation of methyl orange dye and phenol. CuO/MoS 2 @gCN photocatalyst showed 85.14 % degradation of methyl orange and 63.50 % degradation of phenol. Also, the rate constant of CuO/MoS 2 @gCN was found to be 0.051 min−1 for MO dye and 0.02 min−1 for phenol, which is higher than other synthesized samples. The scavenging experiment was carried out to identify the most photoactive species involved in the organic pollutants photocatalytic degradation. Electrons served as the primary photoactive species in photocatalysis. Consequently, it is concluded from the results that CuO/MoS 2 @gCN could be employed as an efficient photocatalyst for wastewater treatment. [ABSTRACT FROM AUTHOR]

The Graph Convolutional Neural Networks (GCNs) have demonstrated a powerful capacity for relation processing. To improve the representation learning capability of GCNs, numerous studies have begun to excavate latent topological graphs based on node features, aiming to obtain richer representations of nodes. However, these graph mining methods based on global metrics, such as Euclidean distance, lack descriptions of uncertainty and fine granularity, leading to the loss of structural information, which affects the final representation of nodes. Therefore, this paper proposes the Three-way Graph Convolutional Neural Network (TW-GCN), which is based on multi-granularity and three-way decision. First, from the perspective of multi-granularity, each attribute is analyzed as a granule, and we establish relations of advantage, disadvantage, and uncertainty for objects under each attribute granule. Three-way relation captures the complex connections between nodes from multiple perspectives, improves the comprehensive understanding of graph data, and avoids the information loss caused by traditional two-way processing. Second, a three-way graph convolutional layer is constructed to capture uncertainty and ambiguity by performing graph convolution on different topological graphs. The final node representation is obtained through message passing on the node's neighbor graph, which considers not only the global structure, but also effectively captures local structural features. Finally, we employ the TW-GCN for multi-label classification within information systems to validate the model's validity based on multiple indicators. • We analyze features from multiple granulation spaces for multi-label systems. • We establish a three-way relation to explain uncertain relationships. • The three-way graph convolutional layer forms the foundation of TW-GCNs. [ABSTRACT FROM AUTHOR]

The direct thermal polymerization techniques were applied to prepare the graphene oxide (GO)-graphitic carbon nitride (gCN) hybrid structure. The prepared hybrid heterojunction GO-gCN nanosheets were utilized as a photocatalyst to remove model pollutants methylene blue (MB) dye. The basic physio-chemical properties of GO-gCN layered materials have been analyzed by various characterization techniques. In addition, the proposed materials have a higher photocatalytic ability toward the degradation of aqueous solution of MB dye under visible light irradiation within a short treatment time. This is because it's the synergistic effects of GO-gCN layer-by-layer structures produced by π─π stacking with charge-transfer interactions. The gCN with GO composite can able to enhance the charge transfer and light-harvesting properties. Under the influence of photocatalyst, the surface of Graphene oxide undergoes the separation and combination of carbonyl radicals, hydroxyl radicals, epoxy radicals, and electron-hole pairs. This enhances the absorption of visible light and improves the degradation of MB, when GO is incorporated into gCN. The removal efficiency of MB reached up to 82.311% within the short treatment time. [Display omitted] • GO-doped gCN layered materials was achieved using thermal polymerization method. • GO-gCN heterojunction photocatalyst was achieved. • The materials have higher visible light absorption property. • Successfully degraded the methylene blue dyes within short treatment time. [ABSTRACT FROM AUTHOR]

Predicting human mobility has always been an important task in Location-based Social Network. Previous efforts fail to capture spatial dependence effectively, mainly reflected in weakening the location topology information. In this paper, we propose a neural network-based method which can capture spatial-temporal dependence to predict the next location of a person. Specifically, we involve a graph convolutional network (GCN) based on a seq2seq framework to capture the location topology information and temporal dependence, respectively. The encoder of the seq2seq framework first generates the hidden state and cell state of the historical trajectories. The GCN is then used to generate graph embeddings of the location topology graph. Finally, we predict future trajectories by aggregated temporal dependence and graph embeddings in the decoder. For evaluation, we leverage two real-world datasets, Foursquare and Gowalla. The experimental results demonstrate that our model has a better performance than the compared models. [ABSTRACT FROM AUTHOR]

Graph Convolutional Network (GCN) has been extensively utilized to extract relations among electroencephalography (EEG) electrode channels for its strong ability to handle non-Euclidean data. However, GCN still has some issues when it comes to extracting features from EEG signals: (1) GCN with more layers may experience over-smoothing, restricting its ability to mine longer dependency relations. (2) At the moment, most GCNs used to process EEG signals construct adjacency matrices by Euclidean distance, only considering the correlations on the feature domain while ignoring changes of signals over the entire time window. To address the issues above, we introduce an Ordinary Differential Equation (ODE) based GCN, which can perfectly eliminate the over-smoothing problem of the traditional GCN. Besides, we also propose a method based on Dynamic Time Wrapping (DTW) algorithm to construct an adjacency matrix in the time domain. To handle adjacency matrices calculated by Euclidean distance and DTW distance respectively, we apply a temporal–spatial model composed of two parallel modules each containing an ODE-based GCN and Long short-term memory neural networks (LSTM) network in turn. We conducted experiments on three public datasets. The results show that our methods have achieved an improvement of 2.19%/2.77%/2.13%/2.01% on Arousal/Valence/Dominance/Liking on DEAP dataset, 1.43% on SEED dataset and 3.06%/3.27% on Arousal/Valence on DREAMER dataset compared with state-of-the-art (SOTA) baseline methods. It demonstrates that our method can effectively approve the performance to handle the relations between EEG channels. The premise of the ODE-based GCN is that signal changes of all EEG channels should be continuous rather than abrupt. We believe that it conforms to the EEG mode, as it is activated by the same emotion stimulation while being collected. • A method based on Dynamic Time Wrapping (DTW) algorithm to measure the similarities among signals of EEG channels on time domain is proposed. • A graph neural network based on Ordinary Differential Equation (ODE) is introduced to extract longer dependency relations among EEG channels. • A temporal–spatial model with two parallel modules composed of an ODE based GCN and an LSTM network is proposed. [ABSTRACT FROM AUTHOR]

• In this study, we constructed a population-based GCN framework by expressing the subject population as adjacency matrix in graph to achieve the diagnosis of AD. The nodes in graph are associated with individual features, and edges are weighted by combining the phenotypic information. • To compensate for the low specificity of ROI features from sMRI or PET image in graph analysis, we acquired individual features by constructing brain network of subject via health group. • Compared with the method of acquiring the ROI features of the brain regions as the input features of GCN, our proposed method remarkably improved the prediction accuracy by about 5 to 10 percentage based on both sMRI and PET images. • We further discussed the influence of phenotypic information on the classification performance of GCN model. Through our testing and experimental analysis on the ADNI dataset, our method achieved improved performance for AD diagnosis and mild cognitive impairment conversion prediction tasks. • Our proposed method also provides technical support for AD diagnosis of GCN method based on three-dimensional brain images. In addition, GCN can provide better flexibility for integrating multimodal features, which also informs AD diagnosis methods using multimodal data. As a deep learning method, graph convolution network (GCN) has the advantage of dealing with non-Euclidean domain problems and is constantly applied in the research of computer-aided diagnosis of Alzheimer's disease (AD). In graph-based methods for AD diagnosis, nodes can represent potential subjects with a set of vectors, and edges combine the interaction and similarity between subjects. However, for the three-dimensional neuroimage like structural Magnetic Resonance Imaging (sMRI) or Positron Emission Tomography (PET), due to the non-sequential of ROI (Region of Interest) features (compared with four-dimensional neuroimage), which makes the graph-based analysis approach more difficult. In this study, we obtained individual features by constructing brain network via health group indirectly, and then constructed a population-based GCN framework by expressing the subject population as adjacency matrix in graph to achieve the diagnosis of AD. The nodes in graph are associated with individual features, and edges are weighted by combining the phenotypic information, we further discuss the influence of phenotypic information on the classification performance of GCN. Compared with acquiring the ROI features of the brain regions as the input features of GCN, our proposed method remarkably improved the prediction accuracy based on both sMRI and PET images by about 5 to 10 percentage. Through our testing and experimental analysis on the public ADNI dataset, our method achieved improved performance for AD diagnosis and mild cognitive impairment conversion prediction tasks. Our proposed method also provides technical support for AD diagnosis using GCN method based on three-dimensional brain images. [ABSTRACT FROM AUTHOR]

Graphical abstract Highlights • Graphitic carbon nitride (GCN) was coupled with NiFe 2 O 4. • GCN/NiFe 2 O 4 was effectively used to mineralize oxytetracycline antibiotic. • Photocatalysis was substantially affected by adsorption process. • GCN/NiFe 2 O 4 was magnetically separated in 2 min from reaction solution. • GCN/NiFe 2 O 4 has significant stability for ten catalytic cycles. Abstract In this work, we report the applicability of NiFe 2 O 4 loaded graphitic carbon nitride (GCN/NiFe 2 O 4) as a magnetic photocatalyst for wastewater treatment. Herein, GCN was prepared by thermal condensation of urea and successfully loaded with NiFe 2 O 4 nanoparticles. GCN/NiFe 2 O 4 was characterized by FESEM, HRTEM, IR, XRD, VSM and UV visible spectral analysis. High surface area of GCN/NiFe 2 O 4 was suitable for adsorption coupled photocatalysis. The GCN/NiFe 2 O 4 exhibited significant photocatalytic activity for mineralization of oxytetracycline (OTC) antibiotic under solar light. The OTC was completely mineralized in 8 h during simultaneous adsorption and photocatalysis process. The rate of OTC degradation was optimal at pH 5. GCN/NiFe 2 O 4 possessed ferromagnetic behavior and could be easily separated by external magnetic field from waste water. The excellent activity of GCN/NiFe 2 O 4 was due to efficient separation of photogenertaed electron and hole pair. GCN/NiFe 2 O 4 exhibited significant recycle efficiency for ten catalytic cycles. [ABSTRACT FROM AUTHOR]

The structure of a sentence conveys rich linguistic knowledge and has proven useful for natural language understanding. In this paper, we aim to incorporate syntactical constraints and long-range word dependencies into the sentence encoding procedure using the widely applied Graph Convolutional Network (GCN) and word dependency trees. Existing syntax-aware GCN methods construct the adjacency matrix by referring to whether two words are connected in the dependency tree. But they fail to model the word dependency type, which reflects how the words are linked in dependency trees. They cannot distinguish the different contributions of different word dependency paths. To avoid introducing redundant word dependencies that harm language understanding, we propose a GCN version that is extended by a novel Word Dependency Gate mechanism. Word Dependency Gate can adaptively maintain the balance between the inclusion and exclusion of specific word dependency paths based on the word dependency type and its word context. Experiments show that our approach can effectively incorporate the relevant syntactical dependency in BERT and achieve a state-of-the-art performance in the End-to-End Aspect-Based Sentiment Analysis and Relation Triple Extraction tasks. • We propose exploiting syntactical structures through a GCN. • Design Word Dependency Gate that encode the edges and labels in the dependency trees. • Our model can effectively incorporate the syntactical dependency in BERT. • Achieve a state-of-the-art performance in End-to-End Aspect-Based Sentiment Analysis. [ABSTRACT FROM AUTHOR]

Traffic prediction on the road, as a vital part of the Intelligent Transportation System (ITS) has attracted much attention recently. It is always one of the hot topics about how to implement an efficient, robust, and accurate vehicular traffic prediction system. With the help of Machine Learning-based (ML) methods, especially Deep Learning-based (DL) methods, the accuracy of the prediction model is increased. However, we also noticed that there are still many open challenges under ML-based vehicular traffic prediction model real-world implementation. Firstly, the time consumption for training DL model is relatively large when compared to parametric models, such as ARIMA, SARIMA. Second, it is still a hot topic for road traffic prediction that how to capture the spacial relationship between road detectors, which is affected by the geographic correlation, as well as the time change. The last but not the least, it is important for us to implement the prediction system into the real world; meanwhile, we should find a way to make use of the advanced technology applied in ITS to improve the prediction system itself. In this paper, we focus on improving the features of the prediction model, which can be helpful for implementing the model in the real world. We present a new hybrid deep learning model by using Graph Convolutional Network (GCN) and the deep aggregation structure (i.e., the sequence to sequence structure) of Gated Recurrent Unit (GRU). Meanwhile, in order to solve the real-world prediction problem, i.e., the online prediction task, we present a new online prediction strategy by using refinement learning. In order to further improve the model's accuracy and efficiency when applied to ITS, we make use of an efficient parallel training strategy while taking advantage of the vehicular cloud structure. [ABSTRACT FROM AUTHOR]