Nested Subspace Arrangement for Representation of Relational Data

Nozomi Hata, Shizuo Kaji, Akihiro Yoshida, Katsuki Fujisawa

Research output: Contribution to journalArticlepeer-review

Abstract

Studies on acquiring appropriate continuous representations of discrete objects, such as graphs and knowledge base data, have been conducted by many researchers in the field of machine learning. In this study, we introduce Nested SubSpace (NSS) arrangement, a comprehensive framework for representation learning. We show that existing embedding techniques can be regarded as special cases of the NSS arrangement. Based on the concept of the NSS arrangement, we implement a Disk-ANChor ARrangement (DANCAR), a representation learning method specialized to reproducing general graphs. Numerical experiments have shown that DANCAR has successfully embedded WordNet in R20 with an F1 score of 0.993 in the reconstruction task. DANCAR is also suitable for visualization in understanding the characteristics of graphs.

MSC Codes 68T30.

Original languageEnglish
JournalUnknown Journal
Publication statusPublished - Jul 4 2020

All Science Journal Classification (ASJC) codes

  • General

Fingerprint Dive into the research topics of 'Nested Subspace Arrangement for Representation of Relational Data'. Together they form a unique fingerprint.

Cite this