Knowledge Graph Completion to Solve University Campus Issues

A number of urban challenges are encountered by modern societies. Governments, businesses and public bodies need to make statistical data widely available in order to tackle these challenges. Nonetheless, current literature and data are problematic; they have inaccuracies which lead to less effectiv...

Full description

Saved in:
Bibliographic Details
Published inJournal of data intelligence Vol. 1; no. 3; pp. 333 - 350
Main Authors Tsukagoshi, Yuto, Kawamura, Takahiro, Sei, Yuichi, Tahara, Yasuyuki, Ohsuga, Akihiko
Format Journal Article
LanguageEnglish
Japanese
Published 01.09.2020
Online AccessGet full text
ISSN2577-610X
2577-610X
DOI10.26421/JDI1.3-3

Cover

More Information
Summary:A number of urban challenges are encountered by modern societies. Governments, businesses and public bodies need to make statistical data widely available in order to tackle these challenges. Nonetheless, current literature and data are problematic; they have inaccuracies which lead to less effective methods of resolving these issues. This research aims to solve this challenge by thinking of a university campus as a microcosm of society, implementing a data integration schema, and combining data into a knowledge graph. Existing completion methods will then be applied and updated. Especially in regards to bicycle environment, our knowledge graph was tailored and evaluated in line with conventional methods, and secondly with our proposed derivative methods. Roughly 650 pieces of parking data, with various dates and times, was contrasted with each time's mean absolute error. Our approach accurately projected 54.5 more bicycles than the conventional method.
ISSN:2577-610X
2577-610X
DOI:10.26421/JDI1.3-3