مقاله انگلیسی رایگان در مورد تعبیه نمودار دانش با راهنمایی تعاملی – IEEE 2019

 

مشخصات مقاله
ترجمه عنوان مقاله تعبیه نمودار دانش با راهنمایی تعاملی از توضیحات نهادی
عنوان انگلیسی مقاله Knowledge Graph Embedding With Interactive Guidance From Entity Descriptions
انتشار مقاله سال 2019
تعداد صفحات مقاله انگلیسی 8 صفحه
هزینه دانلود مقاله انگلیسی رایگان میباشد.
پایگاه داده نشریه IEEE
نوع نگارش مقاله
مقاله پژوهشی (Research Article)
مقاله بیس این مقاله بیس نمیباشد
نمایه (index) Scopus – Master Journals List – JCR
نوع مقاله ISI
فرمت مقاله انگلیسی  PDF
ایمپکت فاکتور(IF)
4.641 در سال 2018
شاخص H_index 56 در سال 2019
شاخص SJR 0.609 در سال 2018
شناسه ISSN 2169-3536
شاخص Quartile (چارک) Q2 در سال 2018
مدل مفهومی ندارد
پرسشنامه ندارد
متغیر ندارد
رفرنس دارد
رشته های مرتبط مهندسی کامپیوتر
نوع ارائه مقاله
ژورنال
مجله / کنفرانس دسترسی – IEEE Access
دانشگاه  School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
کلمات کلیدی تعبیه نمودار دانش، توضیحات نهادی، راهنمایی تعاملی
کلمات کلیدی انگلیسی  Knowledge graph embedding, entity descriptions, interactive guidance
شناسه دیجیتال – doi
https://doi.org/10.1109/ACCESS.2019.2950015
کد محصول  E13933
وضعیت ترجمه مقاله  ترجمه آماده این مقاله موجود نمیباشد. میتوانید از طریق دکمه پایین سفارش دهید.
دانلود رایگان مقاله دانلود رایگان مقاله انگلیسی
سفارش ترجمه این مقاله سفارش ترجمه این مقاله

 

فهرست مطالب مقاله:
Abstract
I. Introduction
II. Related Work
III. Methods
IV. Experiments and Analsis
V. Conclusions
Authors
Figures
References

 

بخشی از متن مقاله:
Abstract

Knowledge Graph (KG) embedding aims to represent both entities and relations into a continuous low-dimensional vector space. Most previous attempts perform the embedding task using only knowledge triples to indicate relations between entities. Entity descriptions, although containing rich background information, have not been well utilized in these methods. In this paper, we propose Entity Descriptions-Guided Embedding (EDGE), a novel method for learning the knowledge graph representations with semantic guidance from entity descriptions. EDGE enables an embedding model to learn simultaneously from 1) knowledge triples that have been directly observed in a given KG, and 2) entity descriptions which have rich semantic information about these entities. In the learning process, EDGE encodes the semantics of entity descriptions to enhance the learning of knowledge graph embedding, and integrates such learned KG embedding to constraint their corresponding word embeddings in entity descriptions. Through this interactive procedure, semantics of entity descriptions may be better transferred into the learned KG embedding. We evaluate EDGE in link prediction and entity classification on Freebase and WordNet. Experimental results show that: 1) with entity descriptions injected, EDGE achieves significant and consistent improvements over state-of-the-art baselines; and 2) compared to those one-time injection schemes studied before, the interactive guidance strategy maximizes the utility of entity descriptions for KG embedding, and indeed achieves substantially better performance.

Introduction

Knowledge graphs (KGs) such as Freebase [1], DBpedia [2] and YAGO [3] provide a structured representation of world knowledge and are extremely useful and crucial resources for several artificial intelligent related applications including question answering [4]–[7] and recommendation systems [8]–[11]. A typical KG is represented as a multirelational graph with entities as nodes and relations as different types of edges, and expresses knowledge as triple facts in the form of (head entity, relation, tail entity) or (h, r, t), indicates the specific relation between two entities. The symbolic representation of KGs with triples is effective in representing structured data, however, with the increased size of KGs, computation inefficiency and data sparsity become serious in various applications related with KGs that people designed in a graph-based method. Recently, a new approach named knowledge graph embedding has been proposed to embed knowledge triples which include entities and relations into a continuous low-dimensional vector space. The embedding from such representation methods contain rich semantic information and can significantly promote a broad range of downstream tasks such as knowledge acquisition and inference [12]–[14]. Most previous representation methods solely learn from fact triples observed in a KG [15]–[24].

دیدگاهتان را بنویسید

نشانی ایمیل شما منتشر نخواهد شد. بخش‌های موردنیاز علامت‌گذاری شده‌اند *

دکمه بازگشت به بالا