Representation Learning of Knowledge Graphs with Entity Attributes and Multimedia Descriptions

Zuo, Yukun; Fang, Quan; Qian, Shengsheng; Zhang, Xiaorui; Xu, Changsheng

doi:10.1109/bigmm.2018.8499179

Cited by 18 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The generation of invisible data was proposed to generate node embedding by utilizing the feature information of nodes. Zuo et al [22] proposed a new model for learning entity attributes and multimedia description of knowledge representation, and verified its effectiveness in knowledge map indexing and multi-model embedding. Hamilton et al [23] introduced the method of embedding a single node and the whole subgraph in detail, considering the recommendation in the field of medicine as an example.…”

Section: B: Methods Of Representing R As a Matrixmentioning

confidence: 99%

Review and Trend Analysis of Knowledge Graphs for Crop Pest and Diseases

et al. 2019

View full text Add to dashboard Cite

Current techniques of knowledge management have some common defects in efficiency, scalability, and applicability. Knowledge graph provides a new way for knowledge management and is a more flexible knowledge management method. Considering the specific features of crop diseases and pest data, this paper analyzed and classified the key techniques and methods of knowledge graph technology in the field of crop diseases and pest in recent years. We introduced the definition and connotation of the crop diseases and pest knowledge, and the current construction method of it was analyzed in detail from four aspects: knowledge representation, extraction, fusion, and reasoning. Furthermore, the application of crop diseases and pest knowledge graph was introduced in detail in the expert system, search engine, and knowledge question-answering system. At last, this paper summarized the challenges and important problems of diseases and pest knowledge graph and forecasted the prospect of knowledge graph according to the key points and difficulties of current knowledge graph research.INDEX TERMS Knowledge graph, crop disease and pest, knowledge extraction, knowledge fusion, knowledge reasoning.

show abstract

Section: B: Methods Of Representing R As a Matrixmentioning

confidence: 99%

Review and Trend Analysis of Knowledge Graphs for Crop Pest and Diseases

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Therefore, it is natural to take into account multi‐source information available to capture richer features for entities, and relations (Wang et al, 2017), for example, triple, graph structure and textual information. There exists some works utilizing textual information in the entity representation and entity alignment (Chen, Tian, et al, 2018; Socher et al, 2013; Zuo et al, 2018). For example, Socher et al (2013) considers textual information to initialize entity representation.…”

Section: Related Workmentioning

confidence: 99%

“…A variety of models have been proposed for link prediction (Bordes et al, 2013; Glorot et al, 2013; Lin et al, 2015; Nathani et al, 2019; Schlichtkrull et al, 2018; Zuo et al, 2018). Recent representation learning (RL) models have gained much attention The vast majority of link prediction models nowadays use original KG elements to learn low‐dimensional representations dubbed RL (Wang et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

“…However, we observe that previous methods of GRL suffer from two shortcomings: These models only preserve the information from triples and graph structures, whereas textual descriptions were mostly ignored. Recent studies (Socher et al, 2013; Zuo et al, 2018; Yao, Mao & Luo, 2019) have shown the importance of interaction between textual description and triples in entity representation, hence incorporating textual information seems to be beneficial. Previous models encounter the ‘neighbour explosion’ problem when scaling GRL to large KG. Various sampling techniques have been proposed to deal with this problem (Chen, Ma, & Xiao, 2018; Chiang et al, 2019; Hamilton et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

“…These models only preserve the information from triples and graph structures, whereas textual descriptions were mostly ignored. Recent studies (Socher et al, 2013;Zuo et al, 2018;Yao, Mao & Luo, 2019) have shown the importance of interaction between textual description and triples in entity representation, hence incorporating textual information seems to be beneficial.…”

mentioning

confidence: 99%

See 2 more Smart Citations

MIFAS: Multi‐source heterogeneous information fusion with adaptive importance sampling for link prediction

et al. 2021

View full text Add to dashboard Cite

Link prediction plays an important role in constructing knowledge graph. Recently, graph representation learning models yield state‐of‐the‐art results. However, existing models concentrate merely on triples or graph structures and mostly ignore textual descriptions, resulting in incomplete or partial information. In this paper, we propose a novel graph representation learning model to address this challenge, namely multi‐source heterogeneous information fusion with adaptive importance sampling. Our model leverages multiple sources, such triple, graph structure and textual description, and generate rich‐attribute embeddings for entities, encapsulating relations simultaneously. We also propose an adaptive importance sampling algorithm to boost aggregation of useful features from local neighbours. Additionally, we also boost node aggregation of useful features from local neighbours by adaptive importance sampling algorithm in our model. Experimental results on two benchmark datasets show that our proposed model significantly outperforms state‐of‐the‐art methods.

show abstract

Text-Enhanced and Relational Context Based Hyperbolic Knowledge Graph Embedding

Ying

et al. 2022

Knowledge Science, Engineering and Management

View full text Add to dashboard Cite

Representation Learning of Knowledge Graphs with Entity Attributes and Multimedia Descriptions

Cited by 18 publications

References 7 publications

Review and Trend Analysis of Knowledge Graphs for Crop Pest and Diseases

Review and Trend Analysis of Knowledge Graphs for Crop Pest and Diseases

MIFAS: Multi‐source heterogeneous information fusion with adaptive importance sampling for link prediction

Text-Enhanced and Relational Context Based Hyperbolic Knowledge Graph Embedding

Contact Info

Product

Resources

About