2019/06/04 by Deepak Nathani, Nathani, Deepak, Jatin Chauhan +5 · 1 citation
Computer Science · Decision Sciences · #Advanced Graph Neural Networks #Computation and Language (cs.CL) #Data Quality and Management #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Topic Modeling
paper · pdf · doi:10.48550/arxiv.1906.01195
openalex publication_date 2019/06/04 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
The recent proliferation of knowledge graphs (KGs) coupled with incomplete or\npartial information, in the form of missing relations (links) between entities,\nhas fueled a lot of research on knowledge base completion (also known as\nrelation prediction). Several recent works suggest that convolutional neural\nnetwork (CNN) based models generate richer and more expressive feature\nembeddings and hence also perform well on relation prediction. However, we\nobserve that these KG embeddings treat triples independently and thus fail to\ncover the complex and hidden information that is inherently implicit in the\nlocal neighborhood surrounding a triple. To this effect, our paper proposes a\nnovel attention based feature embedding that captures both entity and relation\nfeatures in any given entity's neighborhood. Additionally, we also encapsulate\nrelation clusters and multihop relations in our model. Our empirical study\noffers insights into the efficacy of our attention based model and we show\nmarked performance gains in comparison to state of the art methods on all\ndatasets.\n