Computational drug repositioning is a promising strategy in discovering new indicators for approved or experimental drugs. However, most of computational-based methods fall short of taking into account the non-Euclide...
详细信息
ISBN:
(数字)9789819947492
ISBN:
(纸本)9789819947485;9789819947492
Computational drug repositioning is a promising strategy in discovering new indicators for approved or experimental drugs. However, most of computational-based methods fall short of taking into account the non-Euclidean nature of biomedical network data. To address this challenge, we propose a graph representation learning model, called DDAGTP, for drug repositioning using graph transition probability matrix in heterogenous information networks (HINs), In particular, DDAGTP first integrates three different types of drug-disease, drug-protein and protein-disease association networks and their biological knowledge to construct a heterogeneous information network (HIN). Then, a graph convolution autoencoder model is adopted by combining graph transfer probabilities to learn the feature representation of drugs and diseases. Finally, DDAGTP incorporates a CatBoost classifier to complete the task of drug-disease association prediction. Experimental results demonstrate that DDAGTP achieves the excellent performance on all benchmark datasets when compared with state-of-the-art prediction models in terms of several evaluation metrics.
Recent reports by the Mine Safety and Health Administration suggest that several injuries and fatalities could be attributed to the inability to accurately locate miners in case of disasters. Since underground mines h...
详细信息
ISBN:
(纸本)9781450395298
Recent reports by the Mine Safety and Health Administration suggest that several injuries and fatalities could be attributed to the inability to accurately locate miners in case of disasters. Since underground mines have a complicated geometrical landscape and technological constraints such as no GPS information available, it is difficult to predict the location of a miner and hence may cause delays and inefficiencies in rescue operations during a disaster. A significant amount of research has been done to capture complex spatio-temporal relationships of movement of the nodes/people/things with time, spatial and temporal features to separately extract these relationships for location prediction. Although Markov Chains (MC) and Recurrent Neural Network (RNN) based methods have been used to predict locations, not all of them specifically mention the spatial locations, their connections and the aggregation techniques which would allow for the actual representations of the trajectory of miners. Addressing these concerns, we develop a first-of-its-kind end-to-end system entitled MinerFinder to predict the future location of the miners by incorporating Long Short Term Memory (LSTM) for trajectory information with graph autoencoder (GAE) for spatial environmental information representing the node connectivity. In addition, our approach will combine the miners' previous trajectories and daily repetitive patterns enhancing the prediction robustness. We evaluated MinerFinder over synthetic dataset to analyze the structure and location topology of an underground mine compared with foreground locations. Our model outperforms state of the art models and achieves an AP score ranging from (0.62 - 0.68) and Receiver Operating Characteristics (ROC) ranging from (0.63-0.68) with increasing percentage of prominent locations (most visited) to 50%.
How can we exploit Label Propagation (LP) to improve the performance of GNN models on heterophilic graphs? graph Neural Network (GNN) models have received a lot of attention as a powerful deep learning technology that...
详细信息
ISBN:
(纸本)9798350370027;9798350370034
How can we exploit Label Propagation (LP) to improve the performance of GNN models on heterophilic graphs? graph Neural Network (GNN) models have received a lot of attention as a powerful deep learning technology that uses graph structure and features, and has achieved an archived state-of-the-art performance for graph-related tasks. LP has been applied in various studies to improve performance of GNN models. However, LP does not perform well on heterophilic graphs, where nodes of different types are linked with each other, since LP assumes that the graphs inherently exhibits homophily, where similar nodes tend to be linked. Such heterophilic graphs are increasingly common nowadays. In this paper, we propose LPkG (Label Propagation on kNearest Neighbor graphs of graph autoencoder), a simple but effective method to engage LP to improve the performance of GNN models even on heterophilic graphs. LPkG constructs a supplementary homophilic graph, peforms LP on this graph, and uses the results together with the results of GNN models. The supplementary graph is a k-Nearest Neighbor (k-NN) graph genereated from a latent space computed by graph autoencoder (GAE). Experimental results demonstrate that LPkG consistently achieves performance improvement on various heterophilic graph datasets: 2.75% on the Wisconsin dataset, 2.23% on the Texas dataset, and 2.55% on the Cornell dataset.
graph embedding has shown its effectiveness to represent graph information and capture deep relationships in graph data. Most recent graph embedding methods focus on attributed graphs, since they preserve both structu...
详细信息
ISBN:
(纸本)9781728183169
graph embedding has shown its effectiveness to represent graph information and capture deep relationships in graph data. Most recent graph embedding methods focus on attributed graphs, since they preserve both structure and content information in the network. However, corruption can exist in the graph structure as well as the node content of the graph, and both can lead to inferior embedding results. Unfortunately, few existing graph embedding algorithms have considered the corruption problem, and to the best of our knowledge, none has studied structural corruption in attributed graphs, including missing and redundant edges. This field is difficult for previous methods, mainly due to two challenges: (1) the existence of various corruption causes has made it difficult to recognize corruptions in graphs, and (2) the complexity of graph-structured data has increased the difficulty of handling corruption therein for graph embedding methods. These facts lead us here to propose a novel autoencoder-based graph embedding approach, which is robust against structural corruption. Our idea comes from the recent discovery of memorization effects in deep learning. Namely, deep neural networks prefer to fit clean data first, before they over-fit corrupted data. Specifically, we train two autoencoders simultaneously and let them learn the reliability of the edges in the graph from each other. The two autoencoders would evaluate the edges according to their reconstructed structure and manipulate this by devaluing those distrusted edges to update the structure information. The updated structure would be used further in the next iteration as the ground-truth of its peer-network. Experiments on different versions of real-world graphs show state-of-the-art results and demonstrate the robustness of our model against structural corruption.
In this paper, we propose a novel approach to capture inter-company relationships from banking transaction data using graph neural networks with a special attention mechanism and textual industry or sector information...
详细信息
ISBN:
(纸本)9781450391306
In this paper, we propose a novel approach to capture inter-company relationships from banking transaction data using graph neural networks with a special attention mechanism and textual industry or sector information. Transaction data owned by financial institutions can be an alternative source of information to comprehend real-time corporate activities. Such transaction data can be applied to predict stock price and miscellaneous macroeconomic indicators as well as to sophisticate credit and customer relationship management. Although the inter-company relationship is important, traditional methods for extracting information have not captured that enough. With the recent advances in deep learning on graphs, we can expect better extraction of inter-company information from banking transaction data. Especially, we analyze common issues that arise when we represent banking transactions as a network and propose an efficient solution to such problems by introducing a novel edge weight-enhanced attention mechanism, using textual information, and designing an efficient combination of existing graph neural networks.
With the development of graph convolutional network (GCN), which is powerful in graph embedding learning meanwhile can capture node feature information, deep multi-view graph clustering methods based on graph autoenco...
详细信息
With the development of graph convolutional network (GCN), which is powerful in graph embedding learning meanwhile can capture node feature information, deep multi-view graph clustering methods based on graph autoencoder have emerged as a new stream. Although they achieve satisfactory performance, they still have the following weaknesses: (1) some of them model the weights for different views in the encoder part by using attention in the multi-view embedding fusion layer, but fail to consider the weighting in the decoder part to measure the contributions of different views to the reconstruction loss. (2) Most of them directly conduct clustering on the multi-view common embedding layer, but fail to guarantee the alignment of clustering results of different views. To this end, we propose a novel GCN-based deep multi-view graph clustering network with weighting mechanism and collaborative training (DMVGC). The model is composed of multiple view-specific graph encoders and a unified graph decoder. Besides the specially-designed attention module in the encoder part, we construct a reconstruction loss with adaptive weighting mechanism in the decoder part. Additionally, a collaborative self-training clustering objective is jointly conducted on each view-specific embedding layer and the common embedding layer, to make the embedding of each view clustering-friendly toward a common partition. Experiments on several datasets demonstrate the effectiveness of our model.
The accessibility and popularity of online learning have aided the spread of modern learning systems, which provide numerous opportunities for studying the behavior of learners and improving their learning quality. In...
详细信息
The accessibility and popularity of online learning have aided the spread of modern learning systems, which provide numerous opportunities for studying the behavior of learners and improving their learning quality. In online platforms, different learners have different learning styles based on their learning behavior. As a result, analyzing learners' behavior and detecting their learning styles is important in order to provide appropriate resource recommendations, thereby improving their learning outcomes. While several approaches for detecting learning styles have been suggested, behavioral data is collected on a large scale, making it difficult to capture learners' behavior efficiently. Furthermore, current approaches neglect relationships between learner-resource interaction and use techniques based on individual learners' features. However, considering these relations can provide additional benefits. Motivated by these limitations, we propose GRL-LS, a learning style detection approach based on graph representation learning techniques. We first construct a bipartite graph representing the interaction between the set of learners and the set of learning resources. We then introduce a graph embedding technique that learns the latent representations of learners and resources. The learned representation is mapped with the Felder-Silverman learning style model (FSLSM) to identify and group learners using K-means algorithms. The GRL-LS technique can be used in a wide range of educational settings and can be customized to fit a variety of learning style models. An experiment on real-world KDDCup datasets was conducted to evaluate the effectiveness of the proposed GRL-LS approach. The results show that the proposed GRL-LS is efficient at detecting learning style, performs best for all FSLSM dimensions with average accuracy and precision of 88.25% and 78.50% respectively, and outperforms existing approaches with an average improvement of 8.55% accuracy and 2.33% precision
Early DNN-based collaborative filtering (CF) approaches have demonstrated their superior performance than traditional CF such as Matrix Factorization. However, such approaches treat each user-item interaction as separ...
详细信息
Early DNN-based collaborative filtering (CF) approaches have demonstrated their superior performance than traditional CF such as Matrix Factorization. However, such approaches treat each user-item interaction as separate data and thus overlook the intrinsic relationships among data instances. Inspired by the discovery that the autoencoder architecture can force the hidden representation to capture information about the structure of the graph data, in this work, we propose a novel framework called High-order autoencoder based Collaborative Filtering (HACF) that enhances the classic NeuMF framework with autoencoders for capturing latent high-order connectivity signals in the user-item interaction graph. Specifically, each user-item pair is augmented with higher-order neighbours and input to two sets of autoencoders, one set for the users and the other for the items. All the autoencoders in one set share parameters so increasing the number of autoencoders does not increase the model *** have conducted extensive experiments on four popular public benchmark datasets with different sparsity. The overall comparison results demonstrate the advantages of autoencoder-based methods and show that our framework outperforms some state-of-the-art DNN-based collaborative filtering approaches.(c) 2022 Elsevier B.V. All rights reserved.
Anomaly detection is distinguishing unusual objects from normal patterns. It is a complex task due to unpredictable nature of anomalies, which can appear in many forms or they can be hidden by mimicking normal behavio...
详细信息
Anomaly detection is distinguishing unusual objects from normal patterns. It is a complex task due to unpredictable nature of anomalies, which can appear in many forms or they can be hidden by mimicking normal behaviors in a graph structure. Such diversity makes this Deep learning approaches can solve these problems by extracting complex patterns from networks. However, addressing different forms of anomalous instances is essential for successfully implementing these approaches, as different anomaly types require further analysis. Additionally, it is challenging to interpret anomalies beforehand without focusing on every aspect of anomalies. Our objective is to propose an architecture capable of handling all types of anomalous entities by tackling challenges across various domains. In this paper, we introduce ARNAD, a novel framework that integrates three deep models to identify anomalies in graphs: graph neural network, autoencoder, and adversarial autoencoder. ARNAD approaches graph anomaly detection by utilizing the features of the deep parts, and four key elements stand out: (1) the autoencoder learns the overall graph structure and identifies highly deviated ones, (2) the graph neural network exploits graph structure to detect anomalies among the communities, (3) a fixed -size randomized neighborhood that prevents overfitting while reducing complexity (4) the adversarial autoencoder improves the robustness of the framework and discriminates anomalies. To detect anomalies, four receptive components assign risk scores to objects in the attributed network. We evaluated the framework with three synthetic datasets that simulate different behaviors of anomalies and six widely used real attributed networks. Our experimental results show that ARNAD performs competitively with other state-of-the-art models in detecting anomalous entities while minimizing false positives, demonstrating ARNAD's effectiveness in detecting graph anomalies.
Hyperspectral image (HSI) clustering is a fundamental yet challenging task that groups image pixels with similar features into distinct clusters. Among various approaches, contrastive learning methods, which employ th...
详细信息
Hyperspectral image (HSI) clustering is a fundamental yet challenging task that groups image pixels with similar features into distinct clusters. Among various approaches, contrastive learning methods, which employ the concept of encouraging semantically similar samples to move closer together while pushing semantically inconsistent samples apart, have garnered significant attention due to their promising performance. However, the most prevalent approaches face two major limitations: 1) treating all samples indiscriminately during optimization, where the abundance of well-categorized samples overwhelms the feature learning process and 2) tending to introduce noise when constructing positive sample pairs through view augmentation or searching the nearest neighbors, which would cause semantic drift of sample features. To solve these issues, we propose a graph autoencoder-based deep clustering framework named spatial-spectral graph contrastive clustering with hard sample mining (SSGCC) that constructs spatial-spectral dual views without data augmentation and focuses more on hard samples rather than treating all samples equally with the aid of spatial-spectral features. Concretely, we extract the spectral features and the neighborhood spatial features of the samples as dual branches to avoid the noise caused by data augmentation and develop the cluster-oriented consistency learning to facilitate the exchange of knowledge between the two spectral-spatial perspectives. In addition, we propose a hard sample mining-based contrastive learning scheme with the aid of spatial-spectral features. To better measure the importance of the samples, we combine spatial features and spectral features to calculate the similarity between sample pairs. The weights of hard sample pairs are dynamically up-weight while the easy ones are down-weighting to improve the discriminative capability. Extensive experiments on four benchmark HSI datasets demonstrate the effectiveness and superiority of
暂无评论