Logo image
Back
Non-translational Alignment for Multi-relational Networks
Conference paper   Peer reviewed

Non-translational Alignment for Multi-relational Networks

Shengnan Li, Xin Li, Rui Ye, Mingzhong Wang, Haiping Su and Yingzi Ou
Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18), pp.4180-4186
International Joint Conference on Artificial Intelligence (IJCAI), 27th (Stockholm, Sweden, 13-Jul-2018–19-Jul-2018)
International Joint Conference on Artificial Intelligence
2018
DOI: https://doi.org/10.24963/ijcai.2018/581
url
https://www.ijcai.org/proceedings/2018/0581.pdfView
Webpage

Abstract

Business and Management
Most existing solutions for the alignment of multirelational networks, such as multi-lingual knowledge bases, are "translation"-based which facilitate the network embedding via the trans-family, such as TransE. However, they cannot address triangular or other structural properties effectively. Thus, we propose a non-translational approach, which aims to utilize a probabilistic model to offer more robust solutions to the alignment task, by exploring the structural properties as well as leveraging on anchors to project each network onto the same vector space during the process of learning the representation of individual networks. The extensive experiments on four multi-lingual knowledge graphs demonstrate the effectiveness and robustness of the proposed method over a set of stateof-the-art alignment methods.

Details

Metrics

1 File views/ downloads
349 Record Views
Logo image