Related references
Note: Only part of the references are listed.
Article
Computer Science, Artificial Intelligence
Shuang Liang et al.
Summary: Recently, many knowledge graph embedding models have been proposed, but they only focus on semantic information and neglect graph structure information. To address this issue, a novel model called deep relational graph infomax (DRGI) is proposed, which combines complete structure information and semantic information. DRGI consists of two encoders, ARGATs, for capturing semantic and structure information respectively. The method achieves state-of-the-art performance and shows advantages in convergence speed and predictive performance for entities with small indegree.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Jinhao Zhang et al.
Summary: Responding to complex questions is a challenging and valuable goal in Knowledge Base Question Answering (KBQA). Current approaches to extract the in-depth semantics of questions mainly follow two methods. Information retrieval-based methods encode questions directly, while semantic parsing-based methods rely on expensive query graph labels and suffer from sparse reward. This paper proposes a novel semantic parsing model named Structural Information Restraint (SIR) for KBQA. SIR utilizes the structural information of questions for reinforcement-based path reasoning, and introduces a composite structural attention mechanism to improve the efficiency of path reasoning. The model also incorporates a type-assisted reward based on answer concepts, which effectively alleviates the sparse reward problem. Experimental results demonstrate the state-of-the-art performance of our model on CWQ, CQ, and WQSP datasets.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Yifan Qian et al.
Summary: The classification performance of graph convolutional networks (GCNs) is related to the alignment between features, graph, and ground truth, quantified using a subspace alignment measure (SAM). The study showcases the relationship between SAM and classification performance through various experiments, revealing the relative importance of graph and features for classification purposes.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Weimin Li et al.
Summary: This paper proposes a new heterogeneous graph neural network model called CoNR for link prediction task. The model learns node representations and relation representations collaboratively to improve the performance of downstream tasks.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Information Systems
Hui Yin et al.
Summary: The proposed deep learning model DFMF effectively predicts the retweet time by integrating multimodal features of tweets, social relationships, and posting time. Compared to Logistic Regression (LR) and Support Vector Machine (SVM), DFMF achieves up to 11.25% performance improvement in predicting the retweet time accurately.
WORLD WIDE WEB-INTERNET AND WEB INFORMATION SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Xuqian Huang et al.
Summary: Knowledge graph embedding models aim to refine knowledge graphs by learning distributed representations for entities and relations, and predicting unknown triplets by scoring candidate triplets. The proposed MRotatE model exploits triplet features from the perspective of relational and entity rotations, enabling it to model various relation patterns and handle multiple relations simultaneously, achieving state-of-the-art performance.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Lisa Zhang et al.
Summary: This paper proposes a novel Knowledge-aware representation Graph Convolutional Network for Recommendation (KCRec) framework, which effectively addresses the data sparsity and cold-start problems in collaborative filtering in real recommendation scenarios. By utilizing knowledge graph and user-item relationships, the performance of the recommendation system is improved. Experimental results demonstrate that the proposed method outperforms several state-of-the-art baselines.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Wenfeng Zheng et al.
Summary: This paper constructs a knowledge base graph embedding module to extend the versatility of knowledge-based VQA models. By extracting core entities from images and text and performing sub-graph embedding, the accuracy of knowledge-based VQA models is improved.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Artificial Intelligence
Song Bai et al.
Summary: Graph neural networks have recently gained significant attention and performance, with the introduction of hypergraph convolution and hypergraph attention operators for efficient learning on high-order graph-structured data. These operators enhance representation learning capacity and enable the extension of graph neural networks to more flexible models for diverse applications with non-pairwise relationships.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Artificial Intelligence
Jiao Li et al.
Summary: This paper proposes a collective entity linking algorithm CFEL for heterogeneous information networks, which includes coarse-grained and fine-grained models. The coarse-grained model adopts a pruning strategy to reduce the number of candidate entity mentions, while the fine-grained model combines semantic information in text with structural information in HINs using a probabilistic method.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Jie Jiao et al.
Summary: This paper proposes a novel semantic matching-based approach to address the ineffectiveness of queries in knowledge base question answering. By extracting subgraphs of the knowledge base to model the query intention of input questions and conducting semantic matching in the knowledge base, the inefficiency of queries is effectively reduced.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Geochemistry & Geophysics
Yule Duan et al.
Summary: The LC-SMHL algorithm is a novel dimensionality reduction method that can simultaneously discover the manifold-based sparse structure and multivariate discriminant sparse relationship of HSI. By constructing sparse hypergraphs and learning optimal projections, the LC-SMHL method demonstrates better performance in HSI classification.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Xiaohui Wei et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2020)
Article
Computer Science, Artificial Intelligence
Takuma Ebisu et al.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2020)
Article
Geochemistry & Geophysics
Fulin Luo et al.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2020)
Article
Geochemistry & Geophysics
Xueqi Ma et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2019)
Article
Computer Science, Artificial Intelligence
Jens Lehmann et al.
