Search

The identification of compound-protein interactions is crucial for drug discovery, target identification, network pharmacology, and elucidation of protein function. In this paper, we develop a representation learning based graph neural network model for predicting compound-protein interactions. Firstly, Word2vec representation learning method is used to extract features of compounds and proteins automatically. Then the features are input to construct a graph neural network prediction model. Compared with traditional machine learning methods and previous advanced methods, this model shows better results in AUC, accuracy and other model evaluation indicators. Predict the probability of all unknown compound-protein interactions in the Binding-DB database, with four of the top five compound-protein interactions with the highest prediction score confirmed by external evidence. The robustness and effectiveness of the model are further proved. This model can fully utilize aggregated neighbor information, node features, and adaptively capture the topological structure of the compound protein space, thereby achieving high model accuracy. The results of this study provide a new idea and method for the study of compound-protein interaction identification. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer-Aided Design & Computer Graphics / Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao is the property of Gai Kan Bian Wei Hui 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.)

Copyright of Engineering Mechanics / Gongcheng Lixue is the property of Engineering Mechanics Editorial Department 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.)

Copyright of Journal of Frontiers of Computer Science & Technology 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.)

Person re-identification (re-ID) involves the cross-camera retrieval and matching of target pedestrian images, facilitating pedestrian association in scenarios where biometric features such as faces and fingerprints may prove ineffective. It has become a pivotal technology in intelligent video surveillance systems, playing a crucial role in domains like smart security and smart cities. Traditional re-ID algorithms typically employ either representation learning or metric learning methods. A novel approach was proposed which combined representation learning and metric learning methods based on the multi-task learning machine learning paradigm. By capitalizing on the advantages of both feature representation and distance metric, and concurrently training the model using classification loss and triplet loss, comprehensive training for both feature extraction and similarity measurement was ensured. Experimental results validate the effectiveness of the proposed approach, demonstrating superior performance in re-ID tasks and underscoring the robustness and superior generalization capability.

In medical image segmentation tasks, the domain shift problem affects the performance of trained segmentation models in the unseen domain. Therefore, improving model generalization is crucial for the practical application of intelligent models for medical images. Representation learning is currently one of the dominant methods for solving domain generalization problems, mostly using image-level loss and consistency loss to supervise image generation. However, it is not sensitive enough to the deviation of small morphological features of medical images, which can lead to unclear edges of the generated images and affect the subsequent learning of the model. In order to improve the generalization of the model, this paper proposed a semi-supervised feature-level loss and learnable noise domain generalization(FLLN-DG) method for medical image segmentation. Firstly, the introduction of feature level loss improved the problem of unclear boundaries of the generated images. Secondly, the introduction of the learnable noise components further increased the data diversity and improved the model generalization. Compared with the baseline model, FLLN-DG improves the performance in the unseen domain by 2% to 4%, which demonstrates the effectiveness of to feature-level loss and to learnable noise components. FLLN-DG also has the best generalization and segmentation results compared to typical generalization models such as nnUNet, SDNet+AUG, LDDG, SAML and Meta. [ABSTRACT FROM AUTHOR]

Copyright of Telecommunications Science is the property of Beijing Xintong Media 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.)

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.)

Most few-shot knowledge graph completion models have some problems, such as low ability to learn relation representation and rarely attaching importance to the relative location and interaction between query entity pair when the relation between entities is complex or triples’ neighborhood is sparse. A selective attention mechanism and interaction awareness (SAIA) based few-shot knowledge graph completion algorithm is proposed. Firstly, by introducing selective attention mechanism in the process of aggregating neighbor information, the neighbor encoder pays more attention to important neighbors to reduce adverse effects of noise neighbors. Secondly, SAIA utilizes the information related to task relation in the background knowledge graph to learn more accurate relation embedding in the process of relationship representation learning. Finally, in order to mine the interaction information and location information between entities in knowledge graph, a common interaction rate index (CIR) of entity pair is designed to measure the degree of association between entities in 3-hop path. Then, SAIA combines entity pair semantic information to predict new fact. Experimental results show that SAIA outperforms the state-of-the-art few-shot knowledge graph completion methods. Compared with the optimal results of baseline models, the proposed method achieves 5-shot link prediction performance improvement of 0.038, 0.011, 0.028 and 0.052 on NELL-one dataset and 0.034，0.037，0.029 and 0.027 on Wiki-one dataset by the metric MRR, Hits@10, Hits@5 as well as Hits@1, which verifies the effectiveness and feasibility of SAIA.

Copyright of Journal of Frontiers of Computer Science & Technology 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.)

Copyright of Journal of Computer-Aided Design & Computer Graphics / Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao is the property of Gai Kan Bian Wei Hui 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.)

[Purpose/Significance] In the era of big data, the volume of data is growing at an exponential rate. One of the most prominent areas affected by this growth is the field of agriculture. The use of agricultural knowledge graphs, which serve as key infrastructures for managing agricultural knowledge, has expanded significantly. However, as the number of nodes and relationships within these graphs increase, so too does their complexity. This complexity gives rise to new challenges in training and representing such large-scale knowledge graphs. It is therefore of great significance to investigate methods for speeding up the embedding process of agricultural knowledge graphs, while preserving their structural integrity and minimizing resource consumption. This research embarks on a novel exploration to address this issue. It stands out from previous studies by concentrating on a hierarchical representation model for agricultural knowledge graphs. The potential impacts of this research on propelling the advancement of the field and on addressing significant real-world problems are substantial. [Method/Process] To confront this challenge, we propose a hierarchical representation model for agricultural knowledge graphs rooted in the HARP framework. Our model leverages the inherent hierarchical features of the agricultural knowledge graph. It incorporates an improved random walk strategy based on relational paths to semantically model relationship objects within the agricultural knowledge graph. This innovative approach effectively retains the hierarchy and asymmetrical relationship structure of the nodes in the graph, setting our work apart from previous research. The validity of our proposed model is fortified by a strong foundation of theoretical and empirical evidence. [Results/Conclusions] Our experimental results reveal several key findings. First, the hierarchical random walk with path (HRWP) model using the LEIDEN algorithm can preserve the spatial structure more effectively and converge more quickly to the maximum modularity, in comparison to the HARP framework. Second, the fusion model employing HRWP takes less training time than the total training time of both models combined, without significantly affecting the time complexity of the original algorithm. Third, we observed that when traditional algorithms are integrated with HRWP, there is an average improvement of 2% across various indicators, with a substantial enhancement in non-neural network models. Therefore, our proposed model not only accurately represents the agricultural knowledge graph but also effectively reduces the training time. Despite the promising outcomes of our study, there remain areas of potential improvement. One such area is the need for a more detailed discussion on the hierarchical nature of relationship objects in future research. This provides potential avenues for future exploration in this field. The findings of this research carry profound implications for the development of agricultural knowledge management systems, offering an effective approach to handle the burgeoning complexity of knowledge graphs.

As graph neural networks continue to develop, knowledge graph embedding methods based on graph neural networks are receiving increasing attention from researchers. Compared with traditional methods, they can better handle the diversity and complexity of entities, and capture the multiple features and complex relationships of entities, thereby improving the representation ability and application value of knowledge graphs. This paper firstly outlines the development history of knowledge graphs and the basic concepts of knowledge graphs and graph neural networks. Secondly, it focuses on discussing the design ideas and algorithm frameworks of knowledge graph embedding based on graph convolution, graph neural networks, graph attention, and graph autoencoders. Then, it describes the performance of graph neural network knowledge graph embedding in tasks such as link prediction, entity alignment, knowledge graph reasoning, and knowledge graph completion, while supplementing some research on commonsense knowledge graphs with graph neural networks. Finally, this paper makes a comprehensive summary, and future research directions are outlined with respect to some challenges and issues in knowledge graph embedding.

The existing knowledge graph representation learning studies generally suffer from the problems of neglecting the semantic space of specific relations, or difficulty in modeling non-injective complex relations, or difficulty in modeling multiple relation patterns, especially poor performance on two relation patterns of non-commutative combinations as well as subrelations. To address this problem, based on the adaptive projection of entities, this paper proposed a new model with strong representation ability, called ATR3DKRL. By extending the rotation operation from 2D to 3D using the Rodrigues' rotation formula with translation optimization, it could be demonstrated through theoretical derivation that the model could model non-injective complex relationships and multiple relation patterns. The experimental results on several generic datasets show that the model can effectively improve link prediction accuracy, leading existing baseline models in four metrics in dataset DB100K and FB15K-237. Comparing to the baseline model RotatE on the evaluation indicators MRR and H@1 in DB100K, it can significantly increase by 3.3% and 6.5%. [ABSTRACT FROM AUTHOR]

With the combination and development of “Internet plus Education”, online education has become an important teaching mode at present. Research shows that the interaction in online education provides effective help for learners. And the evaluation of interaction is the key to achieving high-quality online learning. The interaction between learners and learning resources in online education builds a bipartite interactive network, and network representation learning technology is a powerful tool for network modeling and prediction. Based on the above analysis, an evaluation algorithm based on bipartite interactive network representation learning (EABINRL) is proposed. This algorithm combines the topological structure information of the bipartite interactive network and the interactive information between nodes, and the aim of this algorithm is to learn the low-dimensional vector representations of two types of nodes by modeling the explicit interaction behavior and the implicit interaction behavior. For different interaction types, different weights are used. Then the model is further optimized and the interactive evaluation results are obtained through Frobenius norm calculation. The results of the learner state prediction experiments conducted on the real public datasets prove the effectiveness of this algorithm.

Entity alignment is an important step in knowledge fusion, which aims to identify equivalent entities in different knowledge graphs. In order to accurately determine the equivalent entities, the existing methods first perform representation learning to map the entities into a low-dimensional vector space, and then infer the equivalence of the entities by the similarity between the vectors. Recent works on entity alignment focus on the improvement of representation learning methods. In order to better understand the mechanism of these models, mine valuable design directions, and provide reference for subsequent optimization and improvement work, this paper reviews the research on representation learning methods for entity alignment. Firstly, based on the existing methods, a general framework for representation learning is proposed, and several representative works are summarized and analyzed. Then, these works are compared and analyzed through experiments, and the common methods of each module in the framework are compared. Through the results, the advantages and disadvantages of various methods are summarized, and the use suggestions are put forward. Finally, the feasibility of the alignment and fusion of large language models and knowledge graphs is preliminarily discussed, and the existing problems and challenges are analyzed. [ABSTRACT FROM AUTHOR]

[Purpose/Significance] In the era of big data, the volume of data is growing at an exponential rate. One of the most prominent areas affected by this growth is the field of agriculture. The use of agricultural knowledge graphs, which serve as key infrastructures for managing agricultural knowledge, has expanded significantly. However, as the number of nodes and relationships within these graphs increase, so too does their complexity. This complexity gives rise to new challenges in training and representing such large-scale knowledge graphs. It is therefore of great significance to investigate methods for speeding up the embedding process of agricultural knowledge graphs, while preserving their structural integrity and minimizing resource consumption. This research embarks on a novel exploration to address this issue. It stands out from previous studies by concentrating on a hierarchical representation model for agricultural knowledge graphs. The potential impacts of this research on propelling the advancement of the field and on addressing significant real-world problems are substantial. [Method/Process] To confront this challenge, we propose a hierarchical representation model for agricultural knowledge graphs rooted in the HARP framework. Our model leverages the inherent hierarchical features of the agricultural knowledge graph. It incorporates an improved random walk strategy based on relational paths to semantically model relationship objects within the agricultural knowledge graph. This innovative approach effectively retains the hierarchy and asymmetrical relationship structure of the nodes in the graph, setting our work apart from previous research. The validity of our proposed model is fortified by a strong foundation of theoretical and empirical evidence. [Results/Conclusions] Our experimental results reveal several key findings. First, the hierarchical random walk with path (HRWP) model using the LEIDEN algorithm can preserve the spatial structure more effectively and converge more quickly to the maximum modularity, in comparison to the HARP framework. Second, the fusion model employing HRWP takes less training time than the total training time of both models combined, without significantly affecting the time complexity of the original algorithm. Third, we observed that when traditional algorithms are integrated with HRWP, there is an average improvement of 2% across various indicators, with a substantial enhancement in non-neural network models. Therefore, our proposed model not only accurately represents the agricultural knowledge graph but also effectively reduces the training time. Despite the promising outcomes of our study, there remain areas of potential improvement. One such area is the need for a more detailed discussion on the hierarchical nature of relationship objects in future research. This provides potential avenues for future exploration in this field. The findings of this research carry profound implications for the development of agricultural knowledge management systems, offering an effective approach to handle the burgeoning complexity of knowledge graphs. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Frontiers of Computer Science & Technology 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.)

Knowledge graphs(KGs)has gradually become valuable asset in the field of AI.However,a major problem is that there are many missing edges in the existing KGs.KGs representation learning can effectively solve this problem.The quality of representation learning depends on how well the geometry of the embedding space matches the structure of the data.Euclidean space has been the main force for embeddings;hyperbolic andspherical spaces gaining popularity due to their ability to better embed new types of structured data.However,most data are highly heterogeneous,the single-space modeling leads to large information distortion.To solve this problem,inspired by MuRP model,mixed-curve space model is proposed to provide representations suitable for heterogeneous structural data.Firstly,the Descartes product of Euclidean hyperbolic and spherical spaces is used to construct mixed space.Then,a graph attention mechanism is designed to obtain the importance of relationship.Experimental results on three KGs benchmark datasets show that the proposed model can effectively alleviate the problems caused by heterostructural embedding in low-dimensional spaces with constant curvature.The proposed method is applied to the cold start problem of recommender system,and the corresponding indicators have been improved to a certain extent.

Knowledge graphs describe objective knowledge in the real world in a structured form,and are confronted with issues of completeness and newly-added knowledge.As an important means of complementing and updating knowledge graphs,know-ledge graph reasoning aims to infer new knowledge based on existing knowledge.In recent years,the research on knowledge graph reasoning based on repre-sentation learning has received extensive attention.The main idea is to convert the traditional reasoning process into semantic vector calculation based on the distributed representation of entities and relations.It has the advantages of fast calculation efficiency and high reasoning performance.In this paper,we review the knowledge graph reasoning based on repre sentation learning.Firstly,this paper summarizes the symbolic representation,data set,evaluation metric,training method,and evaluation task of knowledge graph reasoning.Secondly,it introduces the typical methods of knowledge graph reasoning,including translational distance and semantic matching methods.Thirdly,multi-source information fusion-based knowledge graph reasoning methods are classified.Then,neural network-based reasoning methods are introduced including convolutional neural network,graph neural network,recurrent neural network,and capsule network.Finally,this paper summarizes and forecasts the future research direction of knowledge graph reasoning.

Patent knowledge graph plays a important role in patent accurate retrieval,patent in-depth analysis and patent know-ledge training.This paper proposes a practical patent knowledge graph construction method based on seed knowledge graph,text mining and relationship completion.In this method,to ensure the quality,a seed patent knowledge graph is first established ma-nually,then the concept and relation extraction method of patent text pattern is used to expand the seed patent knowledge graph,and finally the extended patent knowledge graph is quantitatively evaluated.In this paper,artificial extraction of seed knowledge and manual summarization of lexical and syntactic patterns are carried out for patents in the field of traditional Chinese medicine.After obtaining new lexical and syntactic patterns by machine learning,the knowledge graph of seed patent is expanded and completed.Experimental results show that the number of nodes and relationships in the knowledge graph of traditional Chinese medicine are 19 453 and 194 775 respectively.After expansion,they reach 558 461 and 7 275 958 respectively,representing an increase of 27.7 and 36.3 folds respectively.

Rule mining for knowledge graph is to extract the implied rule knowledge from the knowledge graph, the rules which have the advantages of high accuracy and strong interpretability can be used to knowledge graph completion, denoising, data interpretion and other issues. The latest research progress of rule mining methods in the field of knowledge graph are reviewed, the rule mining methods based on path traversal, frequent itemset and representation learning are introduced in detail. The characteristics, performance and existing problems of various methods are analyzed. The quality evaluation function of rules are discussed. Finally, the future research direction and prospect of this field are discussed and prospected. [ABSTRACT FROM AUTHOR]

Model generalization is an important indicator of safety in autonomous driving and medicine. A domain generalization algorithm selection method can guide users to select the most suitable algorithm quickly and accurately. To address the lack of effective algorithm selection methods, a feature contrast (FeCo) selection method based on a domain generalization algorithm based on comparative learning was proposed. Features were selected based on positive and negative example selection strategy. Feature similarity was calculated using the dot product. Finally, the score was calculated by info noise contrastive estimation (InfoNCE). The score was used to evaluate the degree of aggregation of the same class of features and the degree of separation of different class of features, and was verified on a total of 200 domain generalization models in 3 datasets. The experimental results show that FeCo is the only method with stable results among all the methods. The correlation between the results of FeCo and the real generalization error of the model is up to 0. 89, and the running time is shortened by more than 60 times. [ABSTRACT FROM AUTHOR]

Research on the related analysis tasks of graph representation methods was carried out, starting from the formal definition, divide graph representation learning methods into five categories with different core technologies as the classification criteria, including dimensionality reduction analysis, matrix decomposition, random walk, neural network and other representation learning methods. Secondly, through induction and comparative analysis, the paper combed the development context of various technologies, and then showed the advantages and disadvantages of various graphic representation methods in a deep level. Then combined with the inductive analysis of common data sets, evaluation methods and application fields of graph representation learning, it carried out the four-dimensional analysis of dynamics, scalability, interpretability and analyticity. Finally, it summarized and looked forward to the future research trends and development directions of graph representation learning [ABSTRACT FROM AUTHOR]

Copyright of Journal of Frontiers of Computer Science & Technology 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.)

As a structured form of human knowledge,knowledge graphs have played a great supportive role in supporting the semantic intercommunication of massive,multi-source,heterogeneous data,and effectively support tasks such as data analysis,attracting the attention of academia and industry.At present,most knowledge graphs are constructed based on non-real-time static data,without considering the temporal characteristics of entities and relationships.However,data in application scenarios such as social network communication,financial trade,and epidemic spreading network are highly dynamic and exhibit complex temporal properties.How to use time series data to build knowledge graphs and effectively model them is a challenging problem.Recently,numerous studies use temporal information in time series data to enrich the characteristics of knowledge graphs,endowing know-ledge graphs with dynamic features,expanding fact triples into quadruple representation(head entity,relationship,tail entity,time).The knowledge graph which utilizes time-related quadruples to represent knowledge are collectively referred to as temporal knowledge graph.This paper summarizes the research work of temporal knowledge graph representation learning by sorting out and analyzing the corresponding literature.Specifically,it first briefly introduce the background and definition of temporal know-ledge graph.Next,it summarizes the advantages of the temporal knowledge graph representation learning method compared with the traditional knowledge graph representation learning method.Then it elaborates on the recent method of temporal knowledge graph representation learning from the perspective of the method modeling facts,introduces the dataset used by the above method and summarizes the main challenges of this technology.Finally,the future research direction is prospected.

The analysis and inference of program play an important role in software development, maintenance and migration. How to efficiently obtain high quality information from program code has become a hot research topic. In recent years, a large number of researchers have introduced the deep learning-based representation technology into the code analysis tasks. The deep learning model can automatically extract the implicit and useful features implicit in the source code, which can alleviate the dependence on the manual construct feature. This paper first introduces the background and basic concepts of code representation, and summarizes the recent research work on deep learning-based code representation learning from the perspective of code static information analysis. Furthermore, this paper introduces the application of code representation on three tasks, code clone detection, code search and code completion. Finally, it discusses the challenges of deep learning-based code representation and the possible research directions in this field.

Although the sparseness and irregularity of point cloud data have been successfully solved by deep neural networks.However,how to learn the local features of point clouds is still a challenging problem.Existing networks for point cloud representation learning have the problem of extracting features independently between points and points.To this end,a new spatial graph convolution is proposed.Firstly,an adaptive hole K-nearest neighbor algorithm is proposed when constructing the graph structure to maximize local topo-logical structure information.Secondly,the angle feature between each edge of the convolution kernel and the receptive field map is added to the convolution,which ensures more discriminative feature extraction.Finally,in order to make full use of local features,a novel graph pyramid pooling is proposed.This algorithm is tested on the standard public data sets ModelNet40 and ShapeNet,and the accuracy is 93.2% and 86.5% respectively.Experimental results show that the proposed algorithm is at a leading level in point cloud representation learning.

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.)

Copyright of Journal of Remote Sensing is the property of Editorial Office of Journal of Remote Sensing & Science Publishing Co. 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.)

Modeling the probability distribution of rows in structured electronic health records and generating realistic synthetic data is a non-trivial task.Tabular data usually contains discrete columns, and traditional encoding approaches may suffer from the curse of feature dimensionality.Poincaré Ball model was utilized to model the hierarchical structure of nominal variables and Gaussian copula-based generative adversarial network was employed to provide synthetic structured electronic health records.The generated training data are experimentally tested to achieve only 2% difference in utility from the original data yet ensure privacy.

Copyright of Journal of Beijing University of Technology is the property of Journal of Beijing University of Technology, Editorial Department 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.)

Copyright of Journal of Frontiers of Computer Science & Technology 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.)

To improve the accuracy of malicious domain name detection based on the associated information, a detection method combining resolution information and query time was proposed.Firstly, the resolution information was mapped to nodes and edges in a heterogeneous information network, which improved the utilization rate.Secondly, considering the problem of high computational complexity in extracting associated information with matrix multiplication, an efficiency breadth-first network traversal algorithm based on meta-path was proposed.Then, the query time was used to detect the domain names lacking meta-path information, which improved the coverage rate.Finally, domain names were vectorized by representation learning with adaptive weight.The Euclidean distance between domain name feature vectors was used to quantify the correlation between domain names.Based on the vectors learned above, a supervised classifier was constructed to detect malicious domain names.Theoretical analysis and experimental results show that the proposed method preforms well in extraction domain name associated information.The coverage rate and F1 score are 97.7% and 0.951 respectively.

With the development of mobile networks and APPs,user generated contents (UGC) containing multi-source heterogeneous data such as evaluations,markings,scoring,images and videos are greatly valuable information for improving the quality of personalized services.The representation learning of fusion and vectorization on the multi-source heterogeneous UGC is the most critical issue for the successful application.Motivated by this,we propose a representation learning method for effectively fusing and vectorizing the comments and image data.We utilize the Doc2vec and LDA models to sufficiently extract the features of the multi-source comments.The images correlated with the comments are represented with deep convolutional network.A hybrid vectorized representation learning for fusing comments and a convolution strategy for integrating images and comments are presented.The feasibility and effectiveness of the proposed method is demonstrated by applying it to typical Amazon public data sets with heterogeneous UGC,in which the vectorized multi-source heterogeneous UGC is taken as the representation of each product and the classification accuracy of the products are compared.

Alzheimer's disease (AD) is a complex neurodegenerative disease involving a variety of pathogenic factors.So far,the cause of Alzheimer's disease is not clear,the course of the disease is irreversible,and there is no cure.Its early diagnosis and treatment have always been the focus of attention.The neuroimaging data of subjects has an important auxiliary role in the diagnosis of this disease,and the combination of multimodal data can further improve the diagnostic effect.At present,the multimodal data representation learning of the disease has gradually become an emerging research field,which has attracted wide attention from researchers.An autoencoder based multimodal representation learning method for Alzheimer's disease diagnosis is proposed.Firstly,the multimodal data are initially fused to obtain the primary common representation.Then,it is input into the autoencoder network to learn the final common representation in latent space.Finally,the common representation in latent space is classified to obtain the disease result.The proposed method,which achieves the best diagnostic results compared with comparison algorithms,has an accuracy of 88.9% in the classification of AD and healthy subjects in the ADNI dataset.Extensive experimental results verify its effectiveness.

In location-based social networks (LBSN),users share their location and content related to location information.Point-of-interest (POI) recommendation is an important application in LBSN which recommends locations that might be of interest to users.However,compared with other recommendation problems (such as product and movie recommendation),the users' prefe-rence for POI is particularly determined by the time feature.In this paper,the influence of time feature on POI recommendation task is explored,and a time-aware POI recommendation method is proposed,called TAPR (Time Aware POI Recommendation).Our method constructs different relation matrices based on different time scales,and uses tensor decomposition to decompose the constructed multiple relation matrices to obtain the representation of the user and the POI.Finally,our method uses cosine similarity to calculate similarity scores between users and non-visited POIs,and combines the algorithm of user preference modeling to obtain the final recommendation score.Experimental results on two public datasets show that the proposed TAPR performs better than other POI recommendation methods.

Recommendation system,as a very effective technique to solve the information overload,has received a great deal of attention from researchers.However,the real application of recommending systems can be modeled as heterogeneous networks with multi-typed nodes and relations.Thus,heterogeneous network embedding based recommendation becomes a very hot research topic in recent years.However,most of the existing studies do not fully explore the auxiliary information and complex relations which are valuable for enhancing recommending performance.To address the above problems,a recommendation algorithm based on heterogeneous information network embedding and attention neural network is proposed.First,this paper proposes a heterogeneous information network embedding method that maintains semantic relationship and topological structure simultaneously.Then,it designs a meta-path based random walk strategy to extract node sequences from heterogeneous information networks.All the sequences are filtered and then employed to learn the embeddings for each user and item in different meta-paths.At last,this paper presents a recommendation algorithm based on attention neural network with the above embeddings as input.The attention network composed of attention layers and hidden layers is able to explore the complex relationships and hence enhance the performance of recommendation.To verify the effectiveness of the proposed method,this paper conducts experiments on two kinds of real-world datasets and makes a comparison with three competitive algorithms.The results show that the proposed algorithm improves the recommending performance in terms of MAE and RMSE,with a maximum increase of 8.9%.

To solve the problem that the nonlinear relationship between speaker representations cannot be accurately captured in speaker verification, an objective function based on depth information divergence maximization was proposed.It could implicitly represent the nonlinear relationship between speaker representations by calculating the similarity between their distributions.Under the supervision of the optimization goal of maximizing the statistical correlation, the deep neural network was optimized towards the direction that the within-class data was more compact and the between-class data were far away from each other, and finally the discrimination of deep speaker representation space could be effectively improved.Experimental results show that compared with other deep learning methods, the relative EER of the proposed method is reduced by 15.80% at most, which significantly improves the system performance.

Citation recommendation (CR) is able to intelligently generate a paper list related to a query paper, which is of great value to researches. CR problem is related to papers?? semantic and structural information. Recently, network representation learning (NRL) based CR problem has gained extensive attention. However, existing studies all use single granularity networks to model CR, which have the disadvantages of high computational complexity and large memory consumption. For overcoming this challenge, this paper proposes a citation recommendation algorithm based on hierarchical attributed network representation learning (CR-HANRSL), which can greatly improve the efficiency of NRL while taking into account semantic and structural features of papers. First, the network is coarsened into a series of smaller networks based on papers?? attributes and author relationship repeatedly, and the super-nodes are made to contain child-nodes?? attributes after coarsening for further constructing semantic links. Second, this paper uses the single granularity NRL to learn the roughened network and graph convolutional network to refine the learned network representations. Finally, a multi-modal feature representation similarity between papers is generated to get a paper recommendation list. A mass of experimental results on AAN and DBLP two datasets show that the proposed method can improve NRL??s efficiency while learning high-quality network feature representations.

Copyright of Journal of Zhejiang University (Science Edition) is the property of Journal of Zhejiang University (Science Edition) 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.)

Copyright of Journal of Guangdong University of Technology is the property of Journal of Guangdong University of Technology 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.)

Most of the real-life networks are heterogeneous networks that contain multiple types of nodes and edges, and heterogeneous networks integrate more information and contain richer semantic information than homogeneous networks. Heterogeneous network representation learning to have powerful modeling capabilities, enables to solve the heterogeneity of heterogeneous networks effectively, and to embed the rich structure information and semantic information of heterogeneous networks into low-dimensional node representations to facilitate downstream task applications. Through sorting out and classifying the current heterogeneous network representation learning methods at home and abroad, reviewed the current research status of heterogeneous network representation learning methods, compared the characteristics of each category model , introduced the related applications of heterogeneous network representation learning, and summarized and prospected the development trend of heterogeneous network representation learning methods. It is proposed that in-depth discussion can be carried out in the following aspects in future: First, avoid predefined meta-paths and fully release the automatic learning capabilities of the model; Second, design heterogeneous network representation learning method suitable for dynamic and large-scale networks.

In attribute network, the attribute information associated with nodes is essential to improve the performance of network embedding tasks. Nevertheless, network is a graph structure, in which nodes not only contain attribute information but also embrace the rich structural information. In order to make full use of the structural information, firstly, this paper defined influential node characteristics, spatial relationships, and constructed similarity matrix based on the definition of link prediction. Then it mapped correlation similarity vector associated with nodes in the binary group to the relationship space of the adjacency matrix, so as to maintain the node vector matrix structure information feature. Based on the definition of normalized graph Laplacian, it fused the attribute information and label feature and integrated the above three kinds of information into an optimization framework. Finally, it inferenced the projection matrix by calculating the local maximum value through a second order derivative. Experimental results indicate that the proposed algorithm can effectively utilize information of the adjacency structure with the binary group of nodes, and compared with other benchmark network embedding algorithms, it also can achieve better results on the node classification task. [ABSTRACT FROM AUTHOR]

Copyright of Journal on Communication / Tongxin Xuebao is the property of Journal on Communications 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.)

When recommending items to groups on social networks, most of the existing studies use the complete preferences of group members and some aggregation strategies to generate group recommendation. However, in practice, a user may join a group just because of a part of his/her complete preferences related to the group. In this way, using the complete preferences of group members for recommendation will bring a lot of noise, and damage the effect of recommendation. This paper proposed the HINGR_GT method to solve the above problem. It constructed a group oriented heterogeneous social network sub graph for each group, and excluded members' group independent preferences in advance. On this basis, it used the heterogeneous network representation learning method to generate item recommendation for groups. In order to verify the effectiveness of the proposed method, this paper carried out experiments on Douban Movie dataset, and the results show that the proposed method achieves better results in all evaluation metrics. [ABSTRACT FROM AUTHOR]

Factorization machine (FM) has been widely applied in recommender systems since its publication, and to capture feature interactions, FM represents its 2nd-order coefficient of any two features as inner product of the corresponding embedding vectors in Euclidean space. Considering that objects in recommender systems such as items, users, properties and contexts can be described as a heterogeneous network exhibiting hierarchical structures, whereas flat Euclidean space is not able to capture this kind of structure, restricting the feature representation ability of FM, this paper proposes hyperbolic FM (HFM). It represents each feature as a vector in hyperbolic space rather than in Euclidean space, and evaluates the 2nd-order feature interaction strength with hyperbolic distance measure. The reason for adopting hyperbolic geometry is that it has been shown to be the underlying embedding space of hierarchical structures, like trees, graphs and vocabulary. This paper designs two HFMs based on Poincaré ball model and hyperboloid model, respectively, and derives the corresponding Riemannian gradient descent algorithm for optimization. Experiments conducted on various datasets indicate that HFM achieves better performance than original FM with identical number of trainable parameters, and reveals the hierarchical structure of features which is missing in FM, offering explanability to some extent.

The network representation learning algorithm represents the information network as a low-dimensional dense real vector carrying the characteristic information of network nodes, and is applied to the input of downstream machine learning tasks. With the development of machine learning and deep learning, network representation learning has been widely used due to its powerful modeling capabilities and extensive applications. The network representation learning methods and their application were summarized. Firstly, the current network representation learning methods at home and abroad were categorized into different groups, including traditional methods, network structure-based embedding, embedding with attribute information, and spectral-based Graph Convolutional Networks, spatial-based Graph Convolutional Networks and Graph Attention Networks. Then various models were expounded, and the applicability and characteristics of the models were compared. Secondly, the related applications of network representation learning, including recommender system and biomedical field, were introduced. Commonly used data sets are also given, and open source implementations of representation learning models and powerful graph deep learning libraries for reference were organized. Finally, the developing trend of network representation learning was summarized and forecasted. Deep graph neural network, dynamic and heterogeneous network representation, and generalization ability of network model will need to be further studied.

Objective: To evaluate the performance of magnetic resonance imaging (MRI) multi-sequence feature imputation and fusion mutual model based on sequence deletion in differentiating high-grade glioma (HGG) from low-grade glioma (LGG)., Methods: We retrospectively collected multi-sequence MR images from 305 glioma patients, including 189 HGG patients and 116 LGG patients. The region of interest (ROI) of T1-weighted images (T1WI), T2-weighted images (T2WI), T2 fluid attenuated inversion recovery (T2_FLAIR) and post-contrast enhancement T1WI (CE_T1WI) were delineated to extract the radiomics features. A mutual-aid model of MRI multi-sequence feature imputation and fusion based on sequence deletion was used for imputation and fusion of the feature matrix with missing data. The discriminative ability of the model was evaluated using 5-fold cross-validation method and by assessing the accuracy, balanced accuracy, area under the ROC curve (AUC), specificity, and sensitivity. The proposed model was quantitatively compared with other non-holonomic multimodal classification models for discriminating HGG and LGG. Class separability experiments were performed on the latent features learned by the proposed feature imputation and fusion methods to observe the classification effect of the samples in twodimensional plane. Convergence experiments were used to verify the feasibility of the model., Results: For differentiation of HGG from LGG with a missing rate of 10%, the proposed model achieved accuracy, balanced accuracy, AUC, specificity, and sensitivity of 0.777, 0.768, 0.826, 0.754 and 0.780, respectively. The fused latent features showed excellent performance in the class separability experiment, and the algorithm could be iterated to convergence with superior classification performance over other methods at the missing rates of 30% and 50%., Conclusion: The proposed model has excellent performance in classification task of HGG and LGG and outperforms other non-holonomic multimodal classification models, demonstrating its potential for efficient processing of non-holonomic multimodal data.

Copyright of Journal of South China University of Technology (Natural Science Edition) is the property of South China University of Technology 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.)