2019년 컴퓨터종합학술대회
Current Result Document :
| 한글제목(Korean Title) |
Bias Walk Based RDF Entity Embeddings |
| 영문제목(English Title) |
Bias Walk Based RDF Entity Embeddings |
| 저자(Author) |
Van T.T. Duong
Kisung Park
Young-Koo Lee
|
| 원문수록처(Citation) |
VOL 46 NO. 01 PP. 0178 ~ 0180 (2019. 06) |
한글내용
(Korean Abstract) |
|
영문내용
(English Abstract) |
Nowadays, extracting feature vector representation from graph structure is attracting attention of many researchers since the vectors can be applied for any machine learning task. In this paper, we focus on the specific task that is RDF entity embedding that aims to represent entities in the RDF graph as low dimensional vectors while maintaining the most significant information from the original graph. In particular, we propose a new method to map each entity in RDF to vector using word2vec as a language modeling to learn embeddings. In order to use word2vec, we produce a bias random walk, to generate sequences as node context. This bias relies on common connections of similar entities so that we make sure that similar entities will generate similar sequences. Experimental results and the case study on real graphs demonstrate that our method achieves better efficiency. |
| 키워드(Keyword) |
|
| 파일첨부 |
PDF 다운로드
|
