Search

The instability is shown in the existing methods of representation learning based on Euclidean distance under a broad set of conditions. Furthermore, the scarcity and high cost of labels prompt us to explore more expressive representation learning methods which depends on the labels as few as possible. To address these issues, the small-perturbation ideology is firstly introduced on the representation learning model based on the representation probability distribution. The positive small-perturbation information (SPI) which only depend on two labels of each cluster is used to stimulate the representation probability distribution and then two variant models are proposed to fine-tune the expected representation distribution of RBM, namely, Micro-supervised Disturbance GRBM (Micro-DGRBM) and Micro-supervised Disturbance RBM (Micro-DRBM) models. The Kullback-Leibler (KL) divergence of SPI is minimized in the same cluster to promote the representation probability distributions to become more similar in Contrastive Divergence(CD) learning. In contrast, the KL divergence of SPI is maximized in the different clusters to enforce the representation probability distributions to become more dissimilar in CD learning. To explore the representation learning capability under the continuous stimulation of the SPI, we present a deep Micro-supervised Disturbance Learning (Micro-DL) framework based on the Micro-DGRBM and Micro-DRBM models and compare it with a similar deep structure which has not any external stimulation. Experimental results demonstrate that the proposed deep Micro-DL architecture shows better performance in comparison to the baseline method, the most related shallow models and deep frameworks for clustering., Comment: 14 pages

Network embedding (NE) aims to encode the relations of vertices into a low-dimensional space. After NE, we can obtain the learned vectors of vertices that preserve the proximity of network structures for subsequent applications, e.g., vertex classification and link prediction. In existing NE models, they usually exploit the skip-gram with a negative sampling method to optimize their objective functions. Generally, this method learns the vertex representation only from the local connectivity of vertices (i.e., neighbors). However, there is a larger scope of vertex connectivity in real-world scenarios: a vertex may have multifaceted aspects and should belong to overlapping communities. Taking a social network as the overlapping example, a user may subscribe to the channels of politics, economy, and sports simultaneously, but the politics share more common attributes with the economy and less with the sports. In this article, we propose an adversarial learning approach (ACNE) for modeling overlapping communities of vertices. Specifically, we map the association between communities and vertices into an embedding space. Moreover, we take further research on enhancing our ACNE with the following two operations. First, in the initialization stage, we adopt a walking strategy with perception to obtain paths containing more possible boundary vertices to improve overlapping community detection. Then, after representation learning with ACNE, we use soft community assignments from a simple classifier as supervision to update the weights of ACNE. This self-training mechanism referred to as ACNE-ST can help ACNE to achieve better performance. Experimental results demonstrate that the proposed methods, including ACNE and ACNE-ST, can outperform the state-of-the-art models on the subsequent tasks of vertex classification and overlapping community detection.

Citation recommendation is important in the environment of scholarly big data, where finding relevant papers has become more difficult because of information overload. Applying traditional collaborative filtering (CF) to citation recommendation is rather challenging due to the cold start problem and the lack of paper ratings. To address these two challenges, in this paper, we propose a Collaborative filtering with Network representation learning framework for Citation Recommendation dubbed as CNCRec, which is a hybrid user-based CF considering both paper content and network topology. It aims at recommending citations in heterogeneous academic information networks. CNCRec creates the paper rating matrix based on attributed citation network representation learning, where the attributes are topics extracted from the paper text information. Meanwhile, the learned representations of attributed collaboration network is utilized to improve the selection of nearest neighbors. By harnessing the power of network representation learning, CNCRec is able to make full use of the whole citation network topology compared with previous context-aware network-based models. Extensive experiments on both DBLP and APS datasets show that the proposed method outperforms state-of-the-art methods in terms of precision, recall, and Mean Reciprocal Rank. Moreover, CNCRec can better solve the data sparsity problem compared with other CF-based baselines.

In this study, we investigated the roles of bacterial outer membrane component BLP in regulating inflammatory responses and lethality in mice that were induced by a ubiquitous and serious pathogen, Escherichia coli . BLP could alleviate the mortality of mice and organ damage, as well as decrease proinflammatory cytokines and chemokine production and enhance anti-inflammatory cytokine production in mouse serum and organs.

Acute lung injury (ALI) is an inflammatory lung disease that is caused by bacterial infection.

The host TLR2 and TLR4 play critical roles in defense against Escherichia coli (E. coli) infection. The NLRP3 inflammasome is the most thoroughly characterized and responds to numerous physically and chemically diverse stimuli. However, the underlying molecular mechanism of TLR2, TLR4 and NLRP3 in the host inflammatory response to E. coli infection remains unclear. This study aimed to explore the roles of TLR2, TLR4 and NLRP3 in regulating the inflammatory response in E. coli infection-induced mice. Our result indicated that macrophages from NLRP3-deficient mice showed significantly reduced secretion of TNF-α and IL-1β and cyclooxygenase-2 (COX-2) expression in response to stimulation with lipopolysaccharide (LPS), braun lipoprotein (BLP), or WT E. coli compared with macrophages from wild-type mice. In addition, TNF-α and IL-1β production in mouse serum after stimulation agreed with the macrophage data. Liver damage in TLR2-deficient, TLR4-deficient, and NLRP3-deficient mice significantly decreased compared to wild-type mice after stimulation with LPS, BLP, or WT E. coli. These results indicate that besides TLR2 and TLR4, NLRP3 is also plays a critical role in host inflammatory responses to defense against E. coli infection, and might provide a therapeutic target in combating disease with bacterium infection.

Density-based clustering is a traditional research topic with the capability of determining clusters of arbitrary shapes. Besides, through the Density Estimator (DE), density-based methods such as MeanShift, and QuickShift can find the local density maximums as Modes that are excellent representatives of the clusters. However, concentrating on the modes only may suffer from the over-segmentation problem. On the other hand, most density-based methods cannot satisfy the scenario requiring partitioning the data samples into exactly K clusters. To overcome these issues, QuickDSC: a novel and efficient clustering algorithm that groups the samples through the Quick Density Subgraph Estimation, is proposed in this work. It firstly identifies the high-density-connected samples as the Density Subgraphs (DSs). And then, the importance of DSs is estimated from two aspects: density and geometric weight. The top- K important DSs are selected as the cluster centers and based on which the cluster memberships of remaining samples are determined. QuickDSC incorporates three crucial clustering attributes: (1) the cluster centroids are modes (as in density-based methods); (2) able to efficiently return results by utilizing the underlying density structure (as in hierarchical clustering methods); and (3) it explicitly returns K clusters (e.g., K-Means, K -Modes). In addition, QuickDSC is theoretically and empirically efficient. It is only slightly slower than classical clustering methods such as K -Means and DBSCAN. Experiments on artificial and real-world datasets demonstrate the advantages of the proposed method, and the clustering quality outperforms the state-of-the-art approaches.

With the advancement of Cyber Physic Systems and Social Internet of Things, the tourism industry is facing challenges and opportunities. We can now able to collect, store, and analyze large amounts of travel data. With the help of data science and artificial intelligence, smart tourism enables tourists with great autonomy and convenience for an intelligent trip. It is of great significance to make full use of these massive data to provide better services for smart tourism. However, due to the skewed and imbalanced visiting for point of interest located at different places, it is of great significance to predict the tourist flow of each place, which can help the service providers for designing a better schedule visiting strategy in advance. Against this background, this article proposes a multi-graph convolutional network framework, named AMOUNT, for tourist flow prediction. To capture the diverse relationships among POIs, AMOUNT first constructs three subgraphs, including the geographical graph, interaction graph, and the co-relation graph. Then, a multi-graph convolution network is utilized to predict the future tourist flow. Experimental results on two real-world datasets indicate that the proposed AMOUNT model outperforms all other baseline tourist flow prediction approaches.

With the advancement of communication technology and location acquisition technology in the context of modern smart cities, the sharing bike systems offer users the great autonomy and convenience for the last/first-kilometer trip. Meanwhile, we can now able to collect, store, and analyze a large amount of sharing bike data. How to effectively use these massive data to provide better services is an emerging task. However, due to the skewed and imbalanced bike usages for stations located at different places, it is of great significance yet very challenging to predict the potential destinations of each individual trip beforehand so that the service providers can better schedule manual bike re-dispatch in advance. To address this issue, this paper proposes an attention-based deep learning framework for trip destination prediction (AFTER). AFTER first learns the low-dimension representations of users and sharing bike stations via negative sampling strategies. Then, a convolution neural network with an attention mechanism is utilized to predict the future trip destination. Experimental results on a real-world dataset indicate that the proposed framework outperforms several state-of-the-art approaches in terms of precision, recall, and F1.

Recently, dynamic social network research has attracted a great amount of attention, especially in the area of anomaly analysis that analyzes the anomalous change in the evolution of dynamic social networks. However, most of the current research focused on anomaly analysis of the macro representation of dynamic social networks and failed to analyze the nodes that have anomalous structural changes at a micro level. To identify and evaluate anomalous structural change-based nodes in generalized dynamic social networks that only have limited structural information, this research considers undirected and unweighted graphs and develops a multiple-neighbor superposition similarity method ( ), which mainly consists of a multiple-neighbor range algorithm ( ) and a superposition similarity fluctuation algorithm ( ). introduces observation nodes, characterizes the structural similarities of nodes within multiple-neighbor ranges, and proposes a new multiple-neighbor similarity index on the basis of extensional similarity indices. Subsequently, maximally reflects the structural change of each node, using a new superposition similarity fluctuation index from the perspective of diverse multiple-neighbor similarities. As a result, based on and , not only identifies anomalous structural change-based nodes by detecting the anomalous structural changes of nodes but also evaluates their anomalous degrees by quantifying these changes. Results obtained by comparing with state-of-the-art methods via extensive experiments show that can accurately identify anomalous structural change-based nodes and evaluate their anomalous degrees well.

With the widely used Internet of Things, 5G, and smart city technologies, we are able to acquire a variety of vehicle trajectory data. These trajectory data are of great significance which can be used to extract relevant information in order to, for instance, calculate the optimal path from one position to another, detect abnormal behavior, monitor the traffic flow in a city, and predict the next position of an object. One of the key technology is to cluster vehicle trajectory. However, existing methods mainly rely on manually designed metrics which may lead to biased results. Meanwhile, the large scale of vehicle trajectory data has become a challenge because calculating these manually designed metrics will cost more time and space. To address these challenges, we propose to employ network representation learning to achieve accurate vehicle trajectory clustering. Specifically, we first construct the k-nearest neighbor-based internet of vehicles in a dynamic manner. Then we learn the low-dimensional representations of vehicles by performing dynamic network representation learning on the constructed network. Finally, using the learned vehicle vectors, vehicle trajectories are clustered with machine learning methods. Experimental results on the real-word dataset show that our method achieves the best performance compared against baseline methods.

Mining topics on social media (e.g., Twitter and Facebook) is an important task for various applications, such as hot topic discovery, advertising, and promotion activities. Topic modeling techniques are helpful to find out topics that people are talking about. However, current full-analysis models cannot perform well on a focused analysis task-find out all topics related to one particular area in short documents. One reason is that the targeted topic is usually sparse in the corpus of short texts. Another one is, during clustering, even minor errors may compound and render the model useless. This article studies these problems and proposes a targeted analysis model (TAM) with reinforcement learning (RL) to extract any specific topic in a given corpus and perform fine-grained topic generation. In this work, we design a reward function of RL to prevent the false propagation problem induced by Gibbs sampling during the clustering. We amend the targeted topic modeling techniques to the case of RL and use policy search combined with the Gibbs EM algorithm for parameter estimation. Metrics of F1 score and the proposed normalized mutual information-F1 are exploited for the evaluation of clustering and topic generation, respectively. Our experiments have demonstrated that TAM can outperform state-of-the-art models-specifically achieving 25.7% improvement on the F1 score for binary clustering on average.

Evaluating similarities between node pairs in a graph is an important task for data analytics and mining. Among various similarity measures proposed in recent years, SimRank is regarded as one of the most influential measures. However, the computation of SimRank is very expensive especially for large graphs. Although pruning technique and random walk based methods were proposed to accelerate the computation, the accuracy of SimRank score is still very low. In this paper, we propose a novel matrix random sampling approach to accelerate computation speed and reduce memory cost. The matrix random sampling technique not only guarantees the sparsity of the involved matrices, but also enhances the precision of estimated SimRank scores. Moreover, we design a fast sparse matrix–matrix multiplication technique which makes the time complexity of single-source query free of the graph size. We further exploit the Steepest Decent technique to accelerate the speed of convergence. The experimental results show our proposed algorithms outperform the state-of-the-art SimRank algorithms.

While scientific collaboration is critical for a scholar, some collaborators can be more significant than others, e.g., lifetime collaborators. It has been shown that lifetime collaborators are more influential on a scholar’s academic performance. However, little research has been done on investigating predicting such special relationships in academic networks. To this end, we propose Scholar2vec, a novel neural network embedding for representing scholar profiles. First, our approach creates scholars’ research interest vector from textual information, such as demographics, research, and influence. After bridging research interests with a collaboration network, vector representations of scholars can be gained with graph learning. Meanwhile, since scholars are occupied with various attributes, we propose to incorporate four types of scholar attributes for learning scholar vectors. Finally, the early-stage similarity sequence based on Scholar2vec is used to predict lifetime collaborators with machine learning methods. Extensive experiments on two real-world datasets show that Scholar2vec outperforms state-of-the-art methods in lifetime collaborator prediction. Our work presents a new way to measure the similarity between two scholars by vector representation, which tackles the knowledge between network embedding and academic relationship mining.

Network representation learning (NRL) aims at modeling network graph by encoding vertices and edges into a low-dimensional space. These learned representations can be used for subsequent applications, such as vertex classification and link prediction. Negative Sampling (NS) is the most widely used method for boosting the performance of NRL. However, most of the existing work only randomly draws negative samples based on vertex frequencies, i.e., the vertices with higher frequency are more likely to be drawn, which ignores the situation that the sampled one may not be a true negative sample, thus, lead to undesirable embeddings. In this paper, we propose a new negative sampling method, called Hierarchical Negative Sampling (HNS), which is able to model the latent structures of vertices and learn the relations among them. During sampling, HNS can draw more appropriate negative samples and thereby obtain better performance on network embeddings. Firstly, we theoretically demonstrate the superiority of HNS over NS. And then we use experimental results to show that our proposed method outperforms the state-of-the-art models on vertex classification tasks at different training scales in real-world networks.

Hypergraph is popularly used for describing multi-relationships among objects in a unified manner, and spectral clustering is regarded as one of the most effective algorithms for partitioning those objects (vertices) into different communities. However, the traditional spectral clustering for hypergraph (HC) incurs expensive costs in terms of both time and space. In this paper, we propose a framework called GraphLSHC to tackle the scalability problem faced by the large scale hypergraph spectral clustering. In our solution, the hypergraph used in GraphLSHC is expanded into a general format to capture complicated higher-order relationships. Moreover, GraphLSHC is capable to simultaneously partition both vertices and hyperedges according to the “eigen-trick”, which provides an approach for reducing the computational complexity of the clustering. To improve the performance further, several hyperedge-based sampling techniques are proposed, which can supplement the sampled matrix with the whole graph information. We also give a theoretical guarantee for the error boundary of the supplement. Several experiments show the superiority of the proposed framework over the state-of-the-art algorithms.

Natural language processing technologies, such as topic models, have been proven to be effective for scholarly recommendation tasks with the ability to deal with content information. Recently, venue recommendation is becoming an increasingly important research task due to the unprecedented number of publication venues. However, traditional methods focus on either the author’s local network or author-venue similarity, where the multiple relationships between scholars and venues are overlooked, especially the venue–venue interaction. To solve this problem, we propose an author topic model–enhanced joint graph modeling approach that consists of venue topic modeling, venue-specific topic influence modeling, and scholar preference modeling. We first model the venue topic with Latent Dirichlet Allocation. Then, we model the venue-specific topic influence in an asymmetric and low-dimensional way by considering the topic similarity between venues, the top-influence of venues, and the top-susceptibility of venues. The top-influence characterizes venues’ capacity of exerting topic influence on other venues. The top-susceptibility captures venues’ propensity of being topically influenced by other venues. Extensive experiments on two real-world datasets show that our proposed joint graph modeling approach outperforms the state-of-the-art methods.

Prostaglandin E

Finding both efficient and effective quantitative representations for scholars in scientific digital libraries has been a focal point of research. The unprecedented amounts of scholarly datasets, combined with contemporary machine learning and big data techniques, have enabled intelligent and automatic profiling of scholars from this vast and ever-increasing pool of scholarly data. Meanwhile, recent advance in network embedding techniques enables us to mitigate the challenges of large scale and sparsity of academic collaboration networks. In real-world academic social networks, scholars are accompanied with various attributes or features, such as co-authorship and publication records, which result in attributed collaboration networks. It has been observed that both network topology and scholar attributes are important in academic relationship mining. However, previous studies mainly focus on network topology, whereas scholar attributes are overlooked. Moreover, the influence of different scholar attributes are unclear. To bridge this gap, in this work, we present a novel framework of Attributed Collaboration Network Embedding (ACNE) for academic relationship mining. ACNE extracts four types of scholar attributes based on the proposed scholar profiling model, including demographics, research, influence, and sociability. ACNE can learn a low-dimensional representation of scholars considering both scholar attributes and network topology simultaneously. We demonstrate the effectiveness and potentials of ACNE in academic relationship mining by performing collaborator recommendation on two real-world datasets and the contribution and importance of each scholar attribute on scientific collaborator recommendation is investigated. Our work may shed light on academic relationship mining by taking advantage of attributed collaboration network embedding.

With the development of communication and information technologies, smart tourism is gradually changing the tourism industry. internet of Things (ioT) plays an important role in smart tourism. However, it is a challenge to apply ioT for smart tourism because of the need for dealing with a vast amount of data and low-latency communication. To this end, in this article, we outline 5G and Ai-empowered ioT systems for smart tourism. Efficient data transmission based on 5G technology and smart data processing based on Ai technology are significant to unlocking ioT based smart tourism applications. To demonstrate the superior performance of our proposed method, we perform a case study on POi recommendation. The experiment results demonstrate the efficiency and effectiveness of our proposed method.

Predicting the user's trajectory behavior sequence based on point of interests (POIs) recommendation is of great significance in the realization of the smart city with the emerging of social Internet of Things technology. One of the widely adopted frameworks is the user-based collaborative filtering, where the explicit POI rating is calculated based on similar users’ preference. However, the trust between users is seldom considered. We believe that if two users show similar preferences or personality traits, the trust level between them should be high. To this end, we propose to calculate the trust-enhanced user similarity in user-based collaborative filtering based on network representation learning. Meanwhile, due to the significance of geographic influence and temporal influence, we integrate these two factors into POI recommendation by a fusion model. Therefore, our proposed POI recommendation system is unified collaborative recommendation framework, which fuses trust-enhanced users’ preferences to potential POIs with geographic influences and temporal influence for POI recommendation. Finally, we conduct extensive experiments on two real-world datasets by comparing with several state-of-the-art methods in terms of precision@k and recall@k. Experimental results indicate that our proposed trust-enhanced collaborative filtering method outperforms other recommendation approaches.

SimRank is popularly used for evaluating similarities between nodes in a graph. Though some recent work has transformed the SimRank into a linear system which can be solved by using conjugate gradient (CG) algorithm. However, the diagonal preconditioner used in the algorithm still left some room for improvement. There is no work showing what should be an optimal preconditioner for the linear system. For such a sake, in this paper, we theoretically study the preconditioner problem and propose a polynomial preconditioner which can improve the computation speed significantly. In our investigation, we also modify the original linear system to adapt to all situations of the graph. We further prove that the condition number of a polynomial preconditioned matrix is bounded in a narrow interval. It implies that the new preconditioner can effectively accelerate the convergence rate of the CG algorithm. The experimental results show the proposed algorithm outperforms the state-of-the-art algorithms for the all-pairs SimRank computation.

In this article, we present a novel solution to automating the procurement of food materials by using electronic scales, which can automatically identify the food materials along weighing them. Although the CNN model is regarded as one of the most effective solutions to image recognition, the traditional techniques cannot handle the mismatch problem between the lab training data and the real world data. To solve the problem, we propose to embed a partial-and-imbalanced domain adaptation technique (tree adaptation network) in the deep learning model, which can borrow knowledge from sibling classes, to overcome the imbalance problem, and transfer knowledge from the source domain to the target domain, to fight the mismatch problem between the lab training data and the real world data. Experiments show that the proposed approach outperforms state-of-the-art algorithms. Furthermore, the proposed techniques have already been used in practice.

Short text stream clustering has become an important problem for mining textual data in diverse social media platforms (e.g., Twitter). However, most of the existing clustering methods (e.g., LDA and PLSA) are developed based on the assumption of a static corpus of long texts, while little attention has been given to short text streams. Different from the long texts, the clustering of short texts is more challenging since their word co-occurrence pattern easily suffers from a sparsity problem. In this paper, we propose a Dirichlet process biterm-based mixture model (DP-BMM), which can deal with the topic drift problem and the sparsity problem in short text stream clustering. The major advantages of DP-BMM include (1) DP-BMM explicitly exploits the word-pairs constructed from each document to enhance the word co-occurrence pattern in short texts; (2) DP-BMM can deal with the topic drift problem of short text streams naturally. Moreover, we further propose an improved algorithm of DP-BMM with forgetting property called DP-BMM-FP, which can efficiently delete biterms of outdated documents by deleting clusters of outdated batches. To perform inference, we adopt an online Gibbs sampling method for parameter estimation. Our extensive experimental results on real-world datasets show that DP-BMM and DP-BMM-FP can achieve a better performance than the state-of-the-art methods in terms of NMI metrics.

Phthalate esters (PAEs) are ubiquitous pollutants in the environment with toxicological and epidemiological effects for humans. As one of the daily necessities, edible plant oil is an important exposure source of PAEs, due to the inevitable contact with PAE-containing materials and the intrinsic lipid solubility of PAEs. However, limited information is currently available on the exposure risk of PAEs in commercial plant oil. This study was aimed at investigating the occurrence and risk assessment of PAEs in plant oils with a high-frequency import rate in west China. The analysis method was referenced to the Chinese national standard for the determination of PAEs in food. Results indicated that PAEs (mainly including DBP and DEHP) were ubiquitous contaminants in imported plant oils with the detectable rate being up to 56.83% in 366 samples. The detected concentrations were in the range of 0.10-3.20 mg kg

Staphylococcus aureus (S. aureus) is a gram-positive pathogen that can cause infectious diseases in mammals. S. aureus-induced host innate immune responses have a relationship with Toll-like receptor 2 (TLR2), TLR4, and Nod-like receptor pyrin domain-containing protein 3 (NLRP3). However, the detailed roles of TLR2, TLR4, and NLRP3 in regulating the host inflammatory response to S. aureus infection remain unclear. Our data indicated that the S. aureus-induced mortality was aggravated by deficiency of TLR2, TLR4, and NLRP3 in mice. In the subsequent experiment, we found that during S. aureus infection, the roles of TLR2, TLR4, and NLRP3 seemed to be different at multiple timepoints. The deficiency of TLR2, TLR4, or NLRP3 attenuated the expression of High-mobility group box protein 1 (HMGB1) and Hyaluronic acid-binding protein 2 (HABP2), which is accompanied by decreased proinflammatory cytokine (TNF-α), chemokine (RANTES), and anti-inflammatory cytokine (IL-10) production in lungs and serum at 3 h and 6 h post-infection. However, with S. aureus infection prolonged (24 h post-infection), the trend was diametrically opposite. The results showed that deficiency of TLR2, TLR4, or NLRP3 aggravated HABP2 and HMGB1 expression, which is accompanied by enhanced proinflammatory cytokine (TNF-α), chemokine (RANTES), and anti-inflammatory cytokine (IL-10) production in lungs and serum. These results were consistent with the data observed in S. aureus-infected bone marrow-derived macrophages (BMDMs). All these results suggested that during S. aureus infection, TLR2, TLR4, and NLRP3 has time-dependent effect in regulating the balance between immune-driven resistance and tolerance.

With the explosive growth of short documents generated from streaming textual sources (e.g., Twitter), latent topic discovery has become a critical task for short text stream clustering. However, most online clustering models determine the probability of producing a new topic by manually setting some hyper-parameter/threshold, which becomes barrier to achieve better topic discovery results. Moreover, topics generated by using existing models often involve a wide coverage of the vocabulary which is not suitable for online social media analysis. Therefore, we propose a nonparametric model (NPMM) which exploits auxiliary word embeddings to infer the topic number and employs a “spike and slab” function to alleviate the sparsity problem of topic-word distributions in online short text analyses. NPMM can automatically decide whether a given document belongs to existing topics, measured by the squared Mahalanobis distance. Hence, the proposed model is free from tuning the hyper-parameter to obtain the probability of generating new topics. Additionally, we propose a nonparametric sampling strategy to discover representative terms for each topic. To perform inference, we introduce a one-pass Gibbs sampling algorithm based on Cholesky decomposition of covariance matrices, which can further be sped up using a Metropolis-Hastings step. Our experiments demonstrate that NPMM significantly outperforms the state-of-the-art algorithms.

Node classification is a central task in graph data analysis. Scarce or even no labeled data of emerging classes is a big challenge for existing methods. A natural question arises: can we classify the nodes from those classes that have never been seen? In this paper, we study this zero-shot node classification (ZNC) problem which has a two-stage nature: (1) acquiring high-quality class semantic descriptions (CSDs) for knowledge transfer, and (2) designing a well generalized graph-based learning model. For the first stage, we give a novel quantitative CSDs evaluation strategy based on estimating the real class relationships, so as to get the "best" CSDs in a completely automatic way. For the second stage, we propose a novel Decomposed Graph Prototype Network (DGPN) method, following the principles of locality and compositionality for zero-shot model generalization. Finally, we conduct extensive experiments to demonstrate the effectiveness of our solutions., Comment: Accepted by KDD 2021