首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献，搜索用时 31 毫秒
1.
Multi-modal hashing can encode the large-scale social geo-media multimedia data from multiple sources into a common discrete hash space, in which the heterogeneous correlations from multiple modalities could be well explored and preserved into the objective semantic-consistent hash codes. The current researches on multi-modal hashing mainly focus on performing common data reconstruction, but they fail to effectively distill the intrinsic and consensus structures of multi-modal data and fully exploit the inherent semantic knowledge to capture semantic-consistent information across multiple modalities, leading to unsatisfactory retrieval performance. To facilitate this problem and develop an efficient multi-modal geographical retrieval method, in this article, we propose a discriminative multi-modal hashing framework named Cognitive Multi-modal Consistent Hashing (CMCH), which can progressively pursue the structure consensus over heterogeneous multi-modal data and simultaneously explore the informative transformed semantics. Specifically, we construct a parameter-free collaborative multi-modal fusion module to incorporate and excavate the underlying common components from multi-source data. Particularly, our formulation seeks for a joint multi-modal compatibility among multiple modalities under a self-adaptive weighting manner, which can take full advantages of their complementary properties. Moreover, a cognitive self-paced learning policy is further leveraged to conduct progressive feature aggregation, which can coalesce multi-modal data onto the established common latent space in a curriculum learning mode. Furthermore, deep semantic transform learning is elaborated to generate flexible semantics for interactively guiding collaborative hash codes learning. An efficient discrete learning algorithm is devised to address the resulting optimization problem, which obtains stable solutions when dealing with large-scale multi-modal retrieval tasks. Sufficient experiments performed on four large-scale multi-modal datasets demonstrate the encouraging performance of the proposed CMCH method in comparison with the state-of-the-arts over multi-modal information retrieval and computational efficiency. The source codes of this work could be available at https://github.com/JunfengAn1998a/CMCH .  相似文献   

2.
Deep hashing has been an important research topic for using deep learning to boost performance of hash learning. Most existing deep supervised hashing methods mainly focus on how to effectively preserve the similarity in hash coding solely depending on pairwise supervision. However, such pairwise similarity-preserving strategy cannot fully explore the semantic information in most cases, which results in information loss. To address this problem, this paper proposes a discriminative dual-stream deep hashing (DDDH) method, which integrates the pairwise similarity loss and the classification loss into a unified framework to take full advantage of label information. Specifically, the pairwise similarity loss aims to preserve the similarity and structural information of high-dimensional original data. Meanwhile, the designed classification loss can enlarge the margin between different classes which improves the discrimination of learned binary codes. Moreover, an effective optimization algorithm is employed to train the hash code learning framework in an end-to-end manner. The results of extensive experiments on three image datasets demonstrate that our method is superior to several state-of-the-art deep and non-deep hashing methods. Ablation studies and analysis further show the effectiveness of introducing the classification loss in the overall hash learning framework.  相似文献   

3.
As a prevalent problem in online advertising, CTR prediction has attracted plentiful attention from both academia and industry. Recent studies have been reported to establish CTR prediction models in the graph neural networks (GNNs) framework. However, most of GNNs-based models handle feature interactions in a complete graph, while ignoring causal relationships among features, which results in a huge drop in the performance on out-of-distribution data. This paper is dedicated to developing a causality-based CTR prediction model in the GNNs framework (Causal-GNN) integrating representations of feature graph, user graph and ad graph in the context of online advertising. In our model, a structured representation learning method (GraphFwFM) is designed to capture high-order representations on feature graph based on causal discovery among field features in gated graph neural networks (GGNNs), and GraphSAGE is employed to obtain graph representations of users and ads. Experiments conducted on three public datasets demonstrate the superiority of Causal-GNN in AUC and Logloss and the effectiveness of GraphFwFM in capturing high-order representations on causal feature graph.  相似文献   

4.
Graph neural networks have been frequently applied in recommender systems due to their powerful representation abilities for irregular data. However, these methods still suffer from the difficulties such as the inflexible graph structure, sparse and highly imbalanced data, and relatively shallow networks, limiting rate prediction ability for recommendations. This paper presents a novel deep dynamic graph attention framework based on influence and preference relationship reconstruction (DGA-IPR) for recommender systems to learn optimal latent representations of users and items. The entire framework involves a user branch and an item branch. An influence-based dynamic graph attention (IDGA) module, a preference-based dynamic graph attention (PDGA) module, and an adaptive fine feature extraction (AFFE) module are respectively constructed for each branch. Concretely, the first two attention modules concentrate on reconstructing influence and preference relationship graphs, breaking imbalanced and fixed constraints of graph structures. Then a deep feature aggregation block and an adaptive feature fusion operation are built, improving the network depth and capturing potential high-order information expressions. Besides, AFFE is designed to acquire finer latent features for users and items. The DGA-IPR architecture is formed by integrating IDGA, PDGA, and AFFE for users and items, respectively. Experiments reveal the superiority of DGA-IPR over existing recommendation models.  相似文献   

5.
  相似文献   

6.
Recently, graph neural network (GNN) has been widely used in sequential recommendation because of its powerful ability to capture high-order collaborative relations, greatly promoting recommendation performance. However, some existing GNN-based methods fail to make full use of multiple relevant features of nodes and ignore the impact of semantic association between nodes on extracting user preferences. To this end, we propose a multi-feature fused collaborative attention network MASR, which sufficiently learns the temporal and positional features of nodes, and innovatively measures the importance of these two features for analyzing the nodes’ dynamic patterns. In addition, we incorporate semantic-enriched contrastive learning into collaborative filtering to enhance the semantic association between nodes and reduce the noise from the structural neighborhood, which has a positive effect on the sequential recommendation. Compared with the baseline models, the performance of MASR on MovieLens, CDs and Beauty datasets is improved by 2.0%, 2.1% and 1.7% respectively, proving its effectiveness in the sequential recommendation.  相似文献   

7.
Clustering is a basic technique in information processing. Traditional clustering methods, however, are not suitable for high dimensional data. Thus, learning a subspace for clustering has emerged as an important research direction. Nevertheless, the meaningful data are often lying on a low dimensional manifold while existing subspace learning approaches cannot fully capture the nonlinear structures of hidden manifold. In this paper, we propose a novel subspace learning method that not only characterizes the linear and nonlinear structures of data, but also reflects the requirements of following clustering. Compared with other related approaches, the proposed method can derive a subspace that is more suitable for high dimensional data clustering. Promising experimental results on different kinds of data sets demonstrate the effectiveness of the proposed approach.  相似文献   

8.
[目的/意义]实体语义关系分类是信息抽取重要任务之一，将非结构化文本转化成结构化知识，是构建领域本体、知识图谱、开发问答系统、信息检索系统的基础工作。[方法/过程]本文详细梳理了实体语义关系分类的发展历程，从技术方法、应用领域两方面回顾和总结了近5年国内外的最新研究成果，并指出了研究的不足及未来的研究方向。[结果/结论]热门的深度学习方法抛弃了传统浅层机器学习方法繁琐的特征工程，自动学习文本特征，实验发现，在神经网络模型中融入词法、句法特征、引入注意力机制能有效提升关系分类性能。  相似文献   

9.
This paper proposes a method to improve retrieval performance of the vector space model (VSM) in part by utilizing user-supplied information of those documents that are relevant to the query in question. In addition to the user's relevance feedback information, information such as original document similarities is incorporated into the retrieval model, which is built by using a sequence of linear transformations. High-dimensional and sparse vectors are then reduced by singular value decomposition (SVD) and transformed into a low-dimensional vector space, namely the space representing the latent semantic meanings of words. The method has been tested with two test collections, the Medline collection and the Cranfield collection. In order to train the model, multiple partitions are created for each collection. Improvement of average precision of the averages over all partitions, compared with the latent semantic indexing (LSI) model, are 20.57% (Medline) and 22.23% (Cranfield) for the two training data sets, and 0.47% (Medline) and 4.78% (Cranfield) for the test data, respectively. The proposed method provides an approach that makes it possible to preserve user-supplied relevance information for the long term in the system in order to use it later.  相似文献   

10.
In synthetic aperture radar (SAR) image change detection, the deep learning has attracted increasingly more attention because the difference images (DIs) of traditional unsupervised technology are vulnerable to speckle noise. However, most of the existing deep networks do not constrain the distributional characteristics of the hidden space, which may affect the feature representation performance. This paper proposes a variational autoencoder (VAE) network with the siamese structure to detect changes in SAR images. The VAE encodes the input as a probability distribution in the hidden space to obtain regular hidden layer features with a good representation ability. Furthermore, subnetworks with the same parameters and structure can extract the spatial consistency features of the original image, which is conducive to the subsequent classification. The proposed method includes three main steps. First, the training samples are selected based on the false labels generated by a clustering algorithm. Then, we train the proposed model with the semisupervised learning strategy, including unsupervised feature learning and supervised network fine-tuning. Finally, input the original data instead of the DIs in the trained network to obtain the change detection results. The experimental results on four real SAR datasets show the effectiveness and robustness of the proposed method.  相似文献   

11.
Graph neural networks (GNN) have emerged as a new state-of-the-art for learning knowledge graph representations. Although they have shown impressive performance in recent studies, how to efficiently and effectively aggregate neighboring features is not well designed. To tackle this challenge, we propose the simplifying heterogeneous graph neural network (SHGNet), a generic framework that discards the two standard operations in GNN, including the transformation matrix and nonlinear activation. SHGNet, in particular, adopts only the essential component of neighborhood aggregation in GNN and incorporates relation features into feature propagation. Furthermore, to capture complex structures, SHGNet utilizes a hierarchical aggregation architecture, including node aggregation and relation weighting. Thus, the proposed model can treat each relation differently and selectively aggregate informative features. SHGNet has been evaluated for link prediction tasks on three real-world benchmark datasets. The experimental results show that SHGNet significantly promotes efficiency while maintaining superior performance, outperforming all the existing models in 3 out of 4 metrics on NELL-995 and in 4 out of 4 metrics on FB15k-237 dataset.  相似文献   

12.
To improve the effect of multimodal negative sentiment recognition of online public opinion on public health emergencies, we constructed a novel multimodal fine-grained negative sentiment recognition model based on graph convolutional networks (GCN) and ensemble learning. This model comprises BERT and ViT-based multimodal feature representation, GCN-based feature fusion, multiple classifiers, and ensemble learning-based decision fusion. Firstly, the image-text data about COVID-19 is collected from Sina Weibo, and the text and image features are extracted through BERT and ViT, respectively. Secondly, the image-text fused features are generated through GCN in the constructed microblog graph. Finally, AdaBoost is trained to decide the final sentiments recognized by the best classifiers in image, text, and image-text fused features. The results show that the F1-score of this model is 84.13% in sentiment polarity recognition and 82.06% in fine-grained negative sentiment recognition, improved by 4.13% and 7.55% compared to the optimal recognition effect of image-text feature fusion, respectively.  相似文献   

13.
We present an image retrieval framework based on automatic query expansion in a concept feature space by generalizing the vector space model of information retrieval. In this framework, images are represented by vectors of weighted concepts similar to the keyword-based representation used in text retrieval. To generate the concept vocabularies, a statistical model is built by utilizing Support Vector Machine (SVM)-based classification techniques. The images are represented as "bag of concepts" that comprise perceptually and/or semantically distinguishable color and texture patches from local image regions in a multi-dimensional feature space. To explore the correlation between the concepts and overcome the assumption of feature independence in this model, we propose query expansion techniques in the image domain from a new perspective based on both local and global analysis. For the local analysis, the correlations between the concepts based on the co-occurrence pattern, and the metrical constraints based on the neighborhood proximity between the concepts in encoded images, are analyzed by considering local feedback information. We also analyze the concept similarities in the collection as a whole in the form of a similarity thesaurus and propose an efficient query expansion based on the global analysis. The experimental results on a photographic collection of natural scenes and a biomedical database of different imaging modalities demonstrate the effectiveness of the proposed framework in terms of precision and recall.  相似文献   

14.
Hashing has been an emerging topic and has recently attracted widespread attention in multi-modal similarity search applications. However, most existing approaches rely on relaxation schemes to generate binary codes, leading to large quantization errors. In addition, amounts of existing approaches embed labels into the pairwise similarity matrix, leading to expensive time and space costs and losing category information. To address these issues, we propose an Efficient Discrete Matrix factorization Hashing (EDMH). Specifically, EDMH first learns the latent subspaces for individual modality through matrix factorization strategy, which preserves the semantic structure representation information of each modality. In particular, we develop a semantic label offset embedding learning strategy, improving the stability of label embedding regression. Furthermore, we design an efficient discrete optimization scheme to generate compact binary codes discretely. Eventually, we present two efficient learning strategies EDMH-L and EDMH-S to pursue high-quality hash functions. Extensive experiments on various widely-used databases verify that the proposed algorithms produce significant performance and outperform some state-of-the-art approaches, with an average improvement of 2.50% (for Wiki), 2.66% (for MIRFlickr) and 2.25% (for NUS-WIDE) over the best available results, respectively.  相似文献   

15.
With the information explosion of news articles, personalized news recommendation has become important for users to quickly find news that they are interested in. Existing methods on news recommendation mainly include collaborative filtering methods which rely on direct user-item interactions and content based methods which characterize the content of user reading history. Although these methods have achieved good performances, they still suffer from data sparse problem, since most of them fail to extensively exploit high-order structure information (similar users tend to read similar news articles) in news recommendation systems. In this paper, we propose to build a heterogeneous graph to explicitly model the interactions among users, news and latent topics. The incorporated topic information would help indicate a user’s interest and alleviate the sparsity of user-item interactions. Then we take advantage of graph neural networks to learn user and news representations that encode high-order structure information by propagating embeddings over the graph. The learned user embeddings with complete historic user clicks capture the users’ long-term interests. We also consider a user’s short-term interest using the recent reading history with an attention based LSTM model. Experimental results on real-world datasets show that our proposed model significantly outperforms state-of-the-art methods on news recommendation.  相似文献   

16.
Deep learning methods have been widely applied for disease diagnosis on resting-state fMRI (rs-fMRI) data, but they are incapable of investigating global relationships between different brain regions as well as ignoring the interpretability. To address these issues, this paper presents a new graph neural network framework for brain disease diagnosis via jointly learning global relationships and selecting the most discriminative brain regions. Specifically, we first design a self-attention structure learning to capture the global interactions between brain regions for achieving diagnosis effectiveness, and theoretically integrate a feature selection method to reduce the noise influence as well as achieve interpretability. Experiment results on three neurological diseases datasets show the effectiveness of our method, compared to the comparison methods, in terms of diagnostic performance and interpretability.  相似文献   

17.
Both node classification and link prediction are popular topics of supervised learning on the graph data, but previous works seldom integrate them together to capture their complementary information. In this paper, we propose a Multi-Task and Multi-Graph Convolutional Network (MTGCN) to jointly conduct node classification and link prediction in a unified framework. Specifically, MTGCN consists of multiple multi-task learning so that each multi-task learning learns the complementary information between node classification and link prediction. In particular, each multi-task learning uses different inputs to output representations of the graph data. Moreover, the parameters of one multi-task learning initialize the parameters of the other multi-task learning, so that the useful information in the former multi-task learning can be propagated to the other multi-task learning. As a result, the information is augmented to guarantee the quality of representations by exploring the complex constructure inherent in the graph data. Experimental results on six datasets show that our MTGCN outperforms the comparison methods in terms of both node classification and link prediction.  相似文献   

18.
Image–text matching is a crucial branch in multimedia retrieval which relies on learning inter-modal correspondences. Most existing methods focus on global or local correspondence and fail to explore fine-grained global–local alignment. Moreover, the issue of how to infer more accurate similarity scores remains unresolved. In this study, we propose a novel unifying knowledge iterative dissemination and relational reconstruction (KIDRR) network for image–text matching. Particularly, the knowledge graph iterative dissemination module is designed to iteratively broadcast global semantic knowledge, enabling relevant nodes to be associated, resulting in fine-grained intra-modal correlations and features. Hence, vector-based similarity representations are learned from multiple perspectives to model multi-level alignments comprehensively. The relation graph reconstruction module is further developed to enhance cross-modal correspondences by constructing similarity relation graphs and adaptively reconstructing them. We conducted experiments on the datasets Flickr30K and MSCOCO, which have 31,783 and 123,287 images, respectively. Experiments show that KIDRR achieves improvements of nearly 2.2% and 1.6% relative to Recall@1 on Flicr30K and MSCOCO, respectively, compared to the current state-of-the-art baselines.  相似文献   

19.
Precise prediction of Multivariate Time Series (MTS) has been playing a pivotal role in numerous kinds of applications. Existing works have made significant efforts to capture temporal tendency and periodical patterns, but they always ignore abrupt variations and heterogeneous/spatial associations of sensory data. In this paper, we develop a dual normalization (dual-norm) based dynamic graph diffusion network (DNGDN) to capture hidden intricate correlations of MTS data for temporal prediction. Specifically, we design time series decomposition and dual-norm mechanism to learn the latent dependencies and alleviate the adverse effect of abnormal MTS data. Furthermore, a dynamic graph diffusion network is adopted for adaptively exploring the spatial correlations among variables. Extensive experiments are performed on 3 real world experimental datasets with 8 representative baselines for temporal prediction. The performances of DNGDN outperforms all baselines with at least 4% lower MAPE over all datasets.  相似文献   

20.
With the development of 3D technology and the increase in 3D models, 2D image-based 3D model retrieval tasks have drawn increased attention from scholars. Previous works align cross-domain features via adversarial domain alignment and semantic alignment. However, the extracted features of previous methods are disturbed by the residual domain-specific features, and the lack of labels for 3D models makes the semantic alignment challenging. Therefore, we propose disentangled feature learning associated with enhanced semantic alignment to address these problems. On one hand, the disentangled feature learning enables decoupling the twisted raw features into the isolated domain-invariant and domain-specific features, and the domain-specific features will be dropped while performing adversarial domain alignment and semantic alignment to acquire domain-invariant features. On the other hand, we mine the semantic consistency by compacting each 3D model sample and its nearest neighbors to further enhance semantic alignment for unlabeled 3D model domain. We give comprehensive experiments on two public datasets, and the results demonstrate the superiority of the proposed method. Especially on MI3DOR-2 dataset, our method outperforms the current state-of-the-art methods with gains of 2.88% for the strictest retrieval metric NN.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号